Ke Xu

AAAI Conference 2026 Conference Paper

DeFT-LoRA: Decoupled and Fused Tuning with LoRA Experts for Universal Cross-Domain Retrieval

• Ke Xu
• Xiaozheng Shen
• Shanshan Wang
• Mengzhu Wang
• Xun Yang

Universal Cross-Domain Retrieval (UCDR) aims to retrieve images across unseen domains and categories, a critical capability for real-world applications. While large-scale Vision-Language Models (VLMs) like CLIP offer strong zero-shot category generalization, they struggle with domain shifts. Existing methods often improve domain robustness at the cost of high computational overhead or by compromising the VLM's inherent knowledge. To address this, we propose Decoupled and Fused Tuning with LoRA (DeFT-LoRA), a novel and parameter-efficient framework that integrates Low-Rank Adaptation (LoRA) with a Mixture-of-Experts (MoE) mechanism. This approach resolves the intrinsic conflict between domain-invariant and domain-specific knowledge in a single adapter, enabling our model to construct a domain adapters for each input image. We propose a three-stage training strategy, which first learns a shared Base LoRA for domain-invariant features, then derives Domain-Specific Experts to capture specific styles, and finally fuses them dynamically with a lightweight gating network. Extensive experiments on three UCDR benchmarks demonstrate that DeFT-LoRA achieves comparable or superior performance to state-of-the-art methods while requiring only 1.46 percent of CLIP's image-encoder parameters and reducing computational overhead, thereby establishing an exceptional balance between accuracy and efficiency.

AAAI Conference 2026 Conference Paper

Graph Out-of-Distribution Detection via Test-Time Calibration with Dual Dynamic Dictionaries

• Yue Hou
• Ruomei Liu
• Yingke Su
• Junran Wu
• Ke Xu

A key challenge in graph out-of-distribution (OOD) detection lies in the absence of ground-truth OOD samples during training. Existing methods are typically optimized to capture features within the in-distribution (ID) data and calculate OOD scores, which often limits pre-trained models from representing distributional boundaries, leading to unreliable OOD detection. Moreover, the latent structure of graph data is often governed by multiple underlying factors, which remains less explored. To address these challenges, we propose a novel test-time graph OOD detection method, termed BaCa, that calibrates OOD scores using dual dynamically updated dictionaries without requiring fine-tuning the pre-trained model. Specifically, BaCa estimates graphons and applies a mix-up strategy solely with test samples to generate diverse boundary-aware discriminative topologies, eliminating the need for exposing auxiliary datasets as outliers. We construct dual dynamic dictionaries via priority queues and attention mechanisms to adaptively capture latent ID and OOD representations, which are then utilized for boundary-aware OOD score calibration. To the best of our knowledge, extensive experiments on real-world datasets show that BaCa significantly outperforms existing state-of-the-art methods in OOD detection.

AAAI Conference 2026 Conference Paper

QuEPT: Quantized Elastic Precision Transformers with One-Shot Calibration for Multi-Bit Switching

• Ke Xu
• Yixin Wang
• Zhongcheng Li
• Hao Cui
• Jinshui Hu
• Xingyi Zhang

Elastic precision quantization enables multi-bit deployment via a single optimization pass, fitting diverse quantization scenarios. Yet, the high storage and optimization costs associated with the Transformer architecture, research on elastic quantization remains limited, particularly for large language models. This paper proposes QuEPT, an efficient post-training scheme that reconstructs block-wise multi-bit errors with one-shot calibration on a small data slice. It can dynamically adapt to various predefined bit-widths by cascading different low-rank adapters, and supports real-time switching between uniform quantization and mixed precision quantization without repeated optimization. To enhance accuracy and robustness, we introduce Multi-Bit Token Merging (MB-ToMe) to dynamically fuse token features across different bit-widths, improving robustness during bit-width switching. Additionally, we propose Multi-Bit Cascaded Low-Rank adapters (MB-CLoRA) to strengthen correlations between bit-width groups, further improve the overall performance of QuEPT. Extensive experiments demonstrate that QuEPT achieves comparable or better performance to existing state-of-the-art post-training quantization methods.

AAAI Conference 2026 Conference Paper

Surrogate as Teacher: Distillation-Guided Graph Poisoning Attack

• Xingyu Peng
• Ke Xu

While leveraging pseudo-labels has become a common paradigm in untargeted gray-box graph poisoning attacks, it suffers from two critical limitations: the use of brittle hard pseudo-labels that overlook uncertainty and can amplify surrogate model errors, and static guidance that progressively becomes stale as the graph is perturbed. To resolve these issues, we propose MetaDist, a novel framework that reframes the attack as an adversarial self-knowledge distillation process. Here, a "teacher" model provides continuously refined soft pseudo-labels to a "student" model, with the attack objective being to maximize the divergence between them. MetaDist makes two synergistic innovations. It employs the Reverse KL (RKL) divergence as a more strategic attack loss that efficiently converts uncertain nodes into robust, high-confidence errors. Concurrently, it introduces the Online Adaptive Teacher (OAT) mechanism, which adapts the teacher via student feedback to ensure the guidance signal remains relevant. Extensive experiments demonstrate that MetaDist consistently and significantly outperforms strong baselines across multiple datasets, proving its effectiveness and transferability even against advanced graph defenses.

AAAI Conference 2026 Conference Paper

Video Mirror Detection with the Motion-in-Depth Cue

• Alex Warren
• Ke Xu
• Xin Tian
• Gary K. L. Tam
• Benjamin W. Wah
• Rynson W. H. Lau

Detecting mirror regions in RGB videos is essential for scene understanding in applications such as scene reconstruction and robotic navigation. Existing video mirror detectors typically rely on cues like inside-outside mirror correspondences and 2D motion inconsistencies. However, these methods often yield noisy or incomplete predictions when confronted with complex real-world video scenes, especially in areas with occlusion or limited visual features and motions. We observe that human perceive and navigate 3D occluded environments with remarkable ease, owing to Motion-in-Depth (MiD) perception. MiD integrates information from visual appearance (image colors and textures), the way objects move around us in 3D space (3D motions), and their relative distance from us (depth) to determine if something is approaching or receding and to support navigation. Motivated by this neuroscience mechanism, we introduce MiD-VMD, the first approach to explicitly model MiD for video mirror detection. MiD-VMD jointly utilizes contrastive 3D motion, depth, and image features through two novel modules based on a combinational QKV transformer architecture. The Motion-in-Depth Attention Learning (MiD-AL) module captures complementary relationships across these modalities with combinatorial attention and enforces a compact encoding to represent global 3D transformations, resulting in more accurate mirror detection and reduced motion artifacts. The Motion-in-Depth Boundary Detection (MiD-BD) module further sharpens mirror boundaries by leveraging cross-modal attention on 3D motion and depth features. Extensive experiments show that MiD-VMD outperforms current SOTAs.

AAAI Conference 2026 Conference Paper

Vision-Language Reasoning for Geolocalization: A Reinforcement Learning Approach

• Biao Wu
• Meng Fang
• Ling Chen
• Ke Xu
• Tao Cheng
• Jun Wang

Recent advances in vision-language models have opened up new possibilities for reasoning-driven image geolocalization. However, existing approaches often rely on synthetic reasoning annotations or external image retrieval, which can limit interpretability and generalizability. In this paper, we present Geo-R, a retrieval-free framework that uncovers structured reasoning paths from existing ground-truth coordinates and optimizes geolocation accuracy via reinforcement learning. We propose the Chain of Region, a rule-based hierarchical reasoning paradigm that generates precise, interpretable supervision by mapping GPS coordinates to geographic entities (e.g., country, province, city) without relying on model-generated or synthetic labels. Building on this, we introduce a lightweight reinforcement learning strategy with coordinate-aligned rewards based on Haversine distance, enabling the model to refine predictions through spatially meaningful feedback. Our approach bridges structured geographic reasoning with direct spatial supervision, yielding improved localization accuracy, stronger generalization, and more transparent inference. Experimental results across multiple benchmarks confirm the effectiveness of Geo-R, establishing a new retrieval-free paradigm for scalable and interpretable image geolocalization. To facilitate further research and ensure reproducibility, both the model and code will be made publicly available.

NeurIPS Conference 2025 Conference Paper

ALMGuard: Safety Shortcuts and Where to Find Them as Guardrails for Audio–Language Models

• Weifei Jin
• Yuxin Cao
• Junjie Su
• Minhui Xue
• Jie Hao
• Ke Xu
• Jin Song Dong
• Derui Wang

Recent advances in Audio-Language Models (ALMs) have significantly improved multimodal understanding capabilities. However, the introduction of the audio modality also brings new and unique vulnerability vectors. Previous studies have proposed jailbreak attacks that specifically target ALMs, revealing that defenses directly transferred from traditional audio adversarial attacks or text-based Large Language Model (LLM) jailbreaks are largely ineffective against these ALM-specific threats. To address this issue, we propose ALMGuard, the first defense framework tailored to ALMs. Based on the assumption that safety-aligned shortcuts naturally exist in ALMs, we design a method to identify universal Shortcut Activation Perturbations (SAPs) that serve as triggers that activate the safety shortcuts to safeguard ALMs at inference time. To better sift out effective triggers while preserving the model’s utility on benign tasks, we further propose Mel-Gradient Sparse Mask (M-GSM), which restricts perturbations to Mel-frequency bins that are sensitive to jailbreaks but insensitive to speech understanding. Both theoretical analyses and empirical results demonstrate the robustness of our method against both seen and unseen attacks. Overall, \MethodName reduces the average success rate of advanced ALM-specific jailbreak attacks to 4. 6\% across four models, while maintaining comparable utility on benign benchmarks, establishing it as the new state of the art. Our code and data are available at https: //github. com/WeifeiJin/ALMGuard.

NeurIPS Conference 2025 Conference Paper

Exploring Semantic-constrained Adversarial Example with Instruction Uncertainty Reduction

• Jin Hu
• Jiakai Wang
• linna Jing
• Haolin Li
• Liu haodong
• Haotong Qin
• Aishan Liu
• Ke Xu

Recently, semantically constrained adversarial examples (SemanticAE), which are directly generated from natural language instructions, have become a promising avenue for future research due to their flexible attacking forms, but have not been thoroughly explored yet. To generate SemanticAEs, current methods fall short of satisfactory attacking ability as the key underlying factors of semantic uncertainty in human instructions, such as $\textit{referring diversity}$, $\textit{descriptive incompleteness}$, and $\textit{boundary ambiguity}$, have not been fully investigated. To tackle the issues, this paper develops a multi-dimensional $\textbf{ins}$truction $\textbf{u}$ncertainty $\textbf{r}$eduction ($\textbf{InSUR}$) framework to generate more satisfactory SemanticAE, $\textit{i. e. }$, transferable, adaptive, and effective. Specifically, in the dimension of the sampling method, we propose the residual-driven attacking direction stabilization to alleviate the unstable adversarial optimization caused by the diversity of language references. By coarsely predicting the language-guided sampling process, the optimization process will be stabilized by the designed ResAdv-DDIM sampler, therefore releasing the transferable and robust adversarial capability of multi-step diffusion models. In task modeling, we propose the context-encoded attacking scenario constraint to supplement the missing knowledge from incomplete human instructions. Guidance masking and renderer integration are proposed to regulate the constraints of 2D/3D SemanticAE, activating stronger scenario-adapted attacks. Moreover, in the dimension of generator evaluation, we propose the semantic-abstracted attacking evaluation enhancement by clarifying the evaluation boundary based on the label taxonomy, facilitating the development of more effective SemanticAE generators. Extensive experiments demonstrate the superiority of the transfer attack performance of InSUR. Besides, it is worth highlighting that we realize the reference-free generation of semantically constrained 3D adversarial examples by utilizing language-guided 3D generation models for the first time.

NeurIPS Conference 2025 Conference Paper

FlowRefiner: A Robust Traffic Classification Framework against Label Noise

• Mingwei Zhan
• Ruijie Zhao
• Xianwen Deng
• Zhi Xue
• Qi Li
• Zhuotao Liu
• Guang Cheng
• Ke Xu

Network traffic classification is essential for network management and security. In recent years, deep learning (DL) algorithms have emerged as essential tools for classifying complex traffic. However, they rely heavily on high-quality labeled training data. In practice, traffic data is often noisy due to human error or inaccurate automated labeling, which could render classification unreliable and lead to severe consequences. Although some studies have alleviated the label noise issue in specific scenarios, they are difficult to generalize to general traffic classification tasks due to the inherent semantic complexity of traffic data. In this paper, we propose FlowRefiner, a robust and general traffic classification framework against label noise. FlowRefiner consists of three core components: a traffic semantics-driven noise detector, a confidence-guided label correction mechanism, and a cross-granularity robust classifier. First, the noise detector utilizes traffic semantics extracted from a pre-trained encoder to identify mislabeled flows. Next, the confidence-guided label correction module fine-tunes a label predictor to correct noisy labels and construct refined flows. Finally, the cross-granularity robust classifier learns generalized patterns of both flow-level and packet-level, improving classification robustness against noisy labels. We evaluate our method on four traffic datasets with various classification scenarios across varying noise ratios. Experimental results demonstrate that FlowRefiner mitigates the impact of label noise and consistently outperforms state-of-the-art baselines by a large margin. The code is available at https: //github. com/NSSL-SJTU/FlowRefiner.

NeurIPS Conference 2025 Conference Paper

MRSAudio: A Large-Scale Multimodal Recorded Spatial Audio Dataset with Refined Annotations

• Wenxiang Guo
• Changhao Pan
• Zhiyuan Zhu
• Xintong Hu
• Yu Zhang
• Li Tang
• Rui Yang
• Han Wang

Humans rely on multisensory integration to perceive spatial environments, where auditory cues enable sound source localization in three-dimensional space. Despite the critical role of spatial audio in immersive technologies such as VR/AR, most existing multimodal datasets provide only monaural audio, which limits the development of spatial audio generation and understanding. To address these challenges, we introduce MRSAudio, a large-scale multimodal spatial audio dataset designed to advance research in spatial audio understanding and generation. MRSAudio spans four distinct components: MRSLife, MRSSpeech, MRSMusic, and MRSSing, covering diverse real-world scenarios. The dataset includes synchronized binaural and ambisonic audio, exocentric and egocentric video, motion trajectories, and fine-grained annotations such as transcripts, phoneme boundaries, lyrics, scores, and prompts. To demonstrate the utility and versatility of MRSAudio, we establish five foundational tasks: audio spatialization, and spatial text to speech, spatial singing voice synthesis, spatial music generation and sound event localization and detection. Results show that MRSAudio enables high-quality spatial modeling and supports a broad range of spatial audio research. Demos and dataset access are available at https: //mrsaudio. github. io.

ICLR Conference 2025 Conference Paper

RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation

• Songming Liu
• Lingxuan Wu
• Bangguo Li
• Hengkai Tan
• Huayu Chen
• Zhengyi Wang
• Ke Xu
• Hang Su 0006

Bimanual manipulation is essential in robotics, yet developing foundation models is extremely challenging due to the inherent complexity of coordinating two robot arms (leading to multi-modal action distributions) and the scarcity of training data. In this paper, we present the Robotics Diffusion Transformer (RDT), a pioneering diffusion foundation model for bimanual manipulation. RDT builds on diffusion models to effectively represent multi-modality, with innovative designs of a scalable Transformer to deal with the heterogeneity of multi-modal inputs and to capture the nonlinearity and high frequency of robotic data. To address data scarcity, we further introduce a Physically Interpretable Unified Action Space, which can unify the action representations of various robots while preserving the physical meanings of original actions, facilitating learning transferrable physical knowledge. With these designs, we managed to pre-train RDT on the largest collection of multi-robot datasets to date and scaled it up to $1.2$B parameters, which is the largest diffusion-based foundation model for robotic manipulation. We finally fine-tuned RDT on a self-created multi-task bimanual dataset with over $6$K+ episodes to refine its manipulation capabilities. Experiments on real robots demonstrate that RDT significantly outperforms existing methods. It exhibits zero-shot generalization to unseen objects and scenes, understands and follows language instructions, learns new skills with just 1$\sim$5 demonstrations, and effectively handles complex, dexterous tasks. We refer to https://rdt-robotics.github.io/rdt-robotics/ for the code and videos.

NeurIPS Conference 2025 Conference Paper

Redundancy-Aware Test-Time Graph Out-of-Distribution Detection

• Yue Hou
• He Zhu
• Ruomei Liu
• Yingke Su
• Junran Wu
• Ke Xu

Distributional discrepancy between training and test data can lead models to make inaccurate predictions when encountering out-of-distribution (OOD) samples in real-world applications. Although existing graph OOD detection methods leverage data-centric techniques to extract effective representations, their performance remains compromised by structural redundancy that induces semantic shifts. To address this dilemma, we propose RedOUT, an unsupervised framework that integrates structural entropy into test-time OOD detection for graph classification. Concretely, we introduce the Redundancy-aware Graph Information Bottleneck (ReGIB) and decompose the objective into essential information and irrelevant redundancy. By minimizing structural entropy, the decoupled redundancy is reduced, and theoretically grounded upper and lower bounds are proposed for optimization. Extensive experiments on real-world datasets demonstrate the superior performance of RedOUT on OOD detection. Specifically, our method achieves an average improvement of 6. 7\%, significantly surpassing the best competitor by 17. 3\% on the ClinTox/LIPO dataset pair.

IJCAI Conference 2025 Conference Paper

Soft Reasoning Paths for Knowledge Graph Completion

• Yanning Hou
• Sihang Zhou
• Ke Liang
• Lingyuan Meng
• Xiaoshu Chen
• Ke Xu
• Siwei Wang
• Xinwang Liu

Reasoning paths are reliable information in knowledge graph completion (KGC) in which algorithms can find strong clues of the actual relation between entities. However, in real-world applications, it is difficult to guarantee that computationally affordable paths exist toward all candidate entities. According to our observation, the prediction accuracy drops significantly when paths are absent. To make the proposed algorithm more stable against the missing path circumstances, we introduce soft reasoning paths. Concretely, a specific learnable latent path embedding is concatenated to each relation to help better model the characteristics of the corresponding paths. The combination of the relation and the corresponding learnable embedding is termed a soft path in our paper. By aligning the soft paths with the reasoning paths, a learnable embedding is guided to learn a generalized path representation of the corresponding relation. In addition, we introduce a hierarchical ranking strategy to make full use of information about the entity, relation, path, and soft path to help improve both the efficiency and accuracy of the model. Extensive experimental results illustrate that our algorithm outperforms the compared state-of-the-art algorithms by a notable margin. Our code will be released at https: //github. com/7HHHHH/SRP-KGC.

AAAI Conference 2025 Conference Paper

Structural Entropy Guided Unsupervised Graph Out-Of-Distribution Detection

• Yue Hou
• He Zhu
• Ruomei Liu
• Yingke Su
• Jinxiang Xia
• Junran Wu
• Ke Xu

With the emerging of huge amount of unlabeled data, unsupervised out-of-distribution (OOD) detection is vital for ensuring the reliability of graph neural networks (GNNs) by identifying OOD samples from in-distribution (ID) ones during testing, where encountering novel or unknown data is inevitable. Existing methods often suffer from compromised performance due to redundant information in graph structures, which impairs their ability to effectively differentiate between ID and OOD data. To address this challenge, we propose SEGO, an unsupervised framework that integrates structural entropy into OOD detection regarding graph classification. Specifically, within the architecture of contrastive learning, SEGO introduces an anchor view in the form of coding tree by minimizing structural entropy. The obtained coding tree effectively removes redundant information from graphs while preserving essential structural information, enabling the capture of distinct graph patterns between ID and OOD samples. Furthermore, we present a multi-grained contrastive learning scheme at local, global, and tree levels using triplet views, where coding trees with essential information serve as the anchor view. Extensive experiments on real-world datasets validate the effectiveness of SEGO, demonstrating superior performance over state-of-the-art baselines in OOD detection. Specifically, our method achieves the best performance on 9 out of 10 dataset pairs, with an average improvement of 3.7% on OOD detection datasets, significantly surpassing the best competitor by 10.8% on the FreeSolv/ToxCast dataset pair.

AAAI Conference 2025 Conference Paper

Thinking in Granularity: Dynamic Quantization for Image Super-Resolution by Intriguing Multi-Granularity Clues

• Mingshen Wang
• Zhao Zhang
• Feng Li
• Ke Xu
• Kang Miao
• Meng Wang

Dynamic quantization has attracted rising attention in image super-resolution (SR) as it expands the potential of heavy SR models onto mobile devices while preserving competitive performance. Most current methods explore layer-to-bit configuration upon varying local regions, adaptively allocating the bit to each layer and patch. Despite the benefits, they still fall short in the tradeoff of SR accuracy and quantization efficiency. Apart from this, adapting the quantization level for each layer individually can disturb the original inter-layer relationships, thus diminishing the representation capability of quantized models. In this work, we propose Granular-DQ, which takes advantage of multi-granularity clues and local patch statistics, achieving a distinctive patch-wise and layer-invariant dynamic quantization paradigm. Specifically, Granular-DQ initiates by developing a granularity-bit controller to apprehend the coarse-to-fine granular representations of local patches, matching their proportional contribution to the entire image to determine the proper bit-width allocation. On this premise, we investigate the interrelationships between bit-width and information density within high-bit patches, establishing a soft gate that enables further fine-grained dynamic bit adaption. Extensive experiments validate the superiority of Granular-DQ in the trade-off between efficiency and accuracy over recent state-of-the-art methods on various SR models.

NeurIPS Conference 2025 Conference Paper

Under the Shadow: Exploiting Opacity Variation for Fine-grained Shadow Detection

• Xiaotian Qiao
• Ke Xu
• Xianglong Yang
• Ruijie Dong
• Xiaofang Xia
• Jiangtao Cui

Shadow characteristics are of great importance for scene understanding. Existing works mainly consider shadow regions as binary masks, often leading to imprecise detection results and suboptimal performance for scene understanding. We demonstrate that such an assumption oversimplifies light-object interactions in the scene, as the scene details under either hard or soft shadows remain visible to a certain degree. Based on this insight, we aim to reformulate the shadow detection paradigm from the opacity perspective, and introduce a new fine-grained shadow detection method. In particular, given an input image, we first propose a shadow opacity augmentation module to generate realistic images with varied shadow opacities. We then introduce a shadow feature separation module to learn the shadow position and opacity representations separately, followed by an opacity mask prediction module that fuses these representations and predicts fine-grained shadow detection results. In addition, we construct a new dataset with opacity-annotated shadow masks across varied scenarios. Extensive experiments demonstrate that our method outperforms the baselines qualitatively and quantitatively, enhancing a wide range of applications, including shadow removal, shadow editing, and 3D reconstruction.

AAAI Conference 2024 Conference Paper

Adversarial Robust Safeguard for Evading Deep Facial Manipulation

• Jiazhi Guan
• Yi Zhao
• Zhuoer Xu
• Changhua Meng
• Ke Xu
• Youjian Zhao

The non-consensual exploitation of facial manipulation has emerged as a pressing societal concern. In tandem with the identification of such fake content, recent research endeavors have advocated countering manipulation techniques through proactive interventions, specifically the incorporation of adversarial noise to impede the manipulation in advance. Nevertheless, with insufficient consideration of robustness, we show that current methods falter in providing protection after simple perturbations, e.g., blur. In addition, traditional optimization-based methods face limitations in scalability as they struggle to accommodate the substantial expansion of data volume, a consequence of the time-intensive iterative pipeline. To solve these challenges, we propose a learning-based model, Adversarial Robust Safeguard (ARS), to generate desirable protection noise in a single forward process, concurrently exhibiting a heightened resistance against prevalent perturbations. Specifically, our method involves a two-way protection design, characterized by a basic protection component responsible for generating efficacious noise features, coupled with robust protection for further enhancement. In robust protection, we first fuse image features with spatially duplicated noise embedding, thereby accounting for inherent information redundancy. Subsequently, a combination comprising a differentiable perturbation module and an adversarial network is devised to simulate potential information degradation during the training process. To evaluate it, we conduct experiments on four manipulation methods and compare recent works comprehensively. The results of our method exhibit good visual effects with pronounced robustness against varied perturbations at different levels.

AAAI Conference 2024 Conference Paper

Boosting Neural Cognitive Diagnosis with Student’s Affective State Modeling

• Shanshan Wang
• Zhen Zeng
• Xun Yang
• Ke Xu
• Xingyi Zhang

Cognitive Diagnosis Modeling aims to infer students' proficiency level on knowledge concepts from their response logs. Existing methods typically model students’ response processes as the interaction between students and exercises or concepts based on hand-crafted or deeply-learned interaction functions. Despite their promising achievements, they fail to consider the relationship between students' cognitive states and affective states in learning, e.g., the feelings of frustration, boredom, or confusion with the learning content, which is insufficient for comprehensive cognitive diagnosis in intelligent education. To fill the research gap, we propose a novel Affect-aware Cognitive Diagnosis (ACD) model which can effectively diagnose the knowledge proficiency levels of students by taking into consideration the affective factors. Specifically, we first design a student affect perception module under the assumption that the affective state is jointly influenced by the student's affect trait and the difficulty of the exercise. Then, our inferred affective distribution is further used to estimate the student's subjective factors, i.e., guessing and slipping, respectively. Finally, we integrate the estimated guessing and slipping parameters with the basic neural cognitive diagnosis framework based on the DINA model, which facilitates the modeling of complex exercising interactions in a more accurate and interpretable fashion. Besides, we also extend our affect perception module in an unsupervised learning setting based on contrastive learning, thus significantly improving the compatibility of our ACD. To the best of our knowledge, we are the first to unify the cognition modeling and affect modeling into the same framework for student cognitive diagnosis. Extensive experiments on real-world datasets clearly demonstrate the effectiveness of our ACD. Our code is available at https://github.com/zeng-zhen/ACD.

IJCAI Conference 2024 Conference Paper

Enhancing Boundary Segmentation for Topological Accuracy with Skeleton-based Methods

• Chuni Liu
• Boyuan Ma
• Xiaojuan Ban
• Yujie Xie
• Hao Wang
• Weihua Xue
• Jingchao Ma
• Ke Xu

Topological consistency plays a crucial role in the task of boundary segmentation for reticular images, such as cell membrane segmentation in neuron electron microscopic images, grain boundary segmentation in material microscopic images and road segmentation in aerial images. In these fields, topological changes in segmentation results have a serious impact on the downstream tasks, which can even exceed the misalignment of the boundary itself. To enhance the topology accuracy in segmentation results, we propose the Skea-Topo Aware loss, which is a novel loss function that takes into account the shape of each object and topological significance of the pixels. It consists of two components. First, a skeleton-aware weighted loss improves the segmentation accuracy by better modeling the object geometry with skeletons. Second, a boundary rectified term effectively identifies and emphasizes topological critical pixels in the prediction errors using both foreground and background skeletons in the ground truth and predictions. Experiments prove that our method improves topological consistency by up to 7 points in VI compared to 13 state-of-art methods, based on objective and subjective assessments across three different boundary segmentation datasets. The code is available at https: //github. com/clovermini/Skea_topo.

NeurIPS Conference 2024 Conference Paper

LuSh-NeRF: Lighting up and Sharpening NeRFs for Low-light Scenes

• Zefan Qu
• Ke Xu
• Gerhard P. Hancke
• Rynson W. Lau

Neural Radiance Fields (NeRFs) have shown remarkable performances in producing novel-view images from high-quality scene images. However, hand-held low-light photography challenges NeRFs as the captured images may simultaneously suffer from low visibility, noise, and camera shakes. While existing NeRF methods may handle either low light or motion, directly combining them or incorporating additional image-based enhancement methods does not work as these degradation factors are highly coupled. We observe that noise in low-light images is always sharp regardless of camera shakes, which implies an implicit order of these degradation factors within the image formation process. This inspires us to explore such an order to decouple and remove these degradation factors while training the NeRF. To this end, we propose in this paper a novel model, named LuSh-NeRF, which can reconstruct a clean and sharp NeRF from a group of hand-held low-light images. The key idea of LuSh-NeRF is to sequentially model noise and blur in the images via multi-view feature consistency and frequency information of NeRF, respectively. Specifically, LuSh-NeRF includes a novel Scene-Noise Decomposition (SND) module for decoupling the noise from the scene representation and a novel Camera Trajectory Prediction (CTP) module for the estimation of camera motions based on low-frequency scene information. To facilitate training and evaluations, we construct a new dataset containing both synthetic and real images. Experiments show that LuSh-NeRF outperforms existing approaches. Our code and dataset can be found here: https: //github. com/quzefan/LuSh-NeRF.

AAAI Conference 2024 Conference Paper

Multi-View Dynamic Reflection Prior for Video Glass Surface Detection

• Fang Liu
• Yuhao Liu
• Jiaying Lin
• Ke Xu
• Rynson W.H. Lau

Recent research has shown significant interest in image-based glass surface detection (GSD). However, detecting glass surfaces in dynamic scenes remains largely unexplored due to the lack of a high-quality dataset and an effective video glass surface detection (VGSD) method. In this paper, we propose the first VGSD approach. Our key observation is that reflections frequently appear on glass surfaces, but they change dynamically as the camera moves. Based on this observation, we propose to offset the excessive dependence on a single uncertainty reflection via joint modeling of temporal and spatial reflection cues. To this end, we propose the VGSD-Net with two novel modules: a Location-aware Reflection Extraction (LRE) module and a Context-enhanced Reflection Integration (CRI) module, for the position-aware reflection feature extraction and the spatial-temporal reflection cues integration, respectively. We have also created the first large-scale video glass surface dataset (VGSD-D), consisting of 19,166 image frames with accurately-annotated glass masks extracted from 297 videos. Extensive experiments demonstrate that VGSD-Net outperforms state-of-the-art approaches adapted from related fields. Code and dataset will be available at https://github.com/fawnliu/VGSD.

AAAI Conference 2024 Conference Paper

PTMQ: Post-training Multi-Bit Quantization of Neural Networks

• Ke Xu
• Zhongcheng Li
• Shanshan Wang
• Xingyi Zhang

The ability of model quantization with arbitrary bit-width to dynamically meet diverse bit-width requirements during runtime has attracted significant attention. Recent research has focused on optimizing large-scale training methods to achieve robust bit-width adaptation, which is a time-consuming process requiring hundreds of GPU hours. Furthermore, converting bit-widths requires recalculating statistical parameters of the norm layers, thereby impeding real-time switching of the bit-width. To overcome these challenges, we propose an efficient Post-Training Multi-bit Quantization (PTMQ) scheme that requires only a small amount of calibration data to perform block-wise reconstruction of multi-bit quantization errors. It eliminates the influence of statistical parameters by fusing norm layers, and supports real-time switching bit-widths in uniform quantization and mixed-precision quantization. To improve quantization accuracy and robustness, we propose a Multi-bit Feature Mixer technique (MFM) for fusing features of different bit-widths to enhance robustness across varying bit-widths. Moreover, we introduced the Group-wise Distillation Loss (GD-Loss) to enhance the correlation between different bit-width groups and further improve the overall performance of PTMQ. Extensive experiments demonstrate that PTMQ achieves comparable performance to existing state-of-the-art post-training quantization methods, while optimizing it speeds up by 100$\times$ compared to recent multi-bit quantization works. Code can be available at https://github.com/xuke225/PTMQ.

AAAI Conference 2024 Conference Paper

Recasting Regional Lighting for Shadow Removal

• Yuhao Liu
• Zhanghan Ke
• Ke Xu
• Fang Liu
• Zhenwei Wang
• Rynson W.H. Lau

Removing shadows requires an understanding of both lighting conditions and object textures in a scene. Existing methods typically learn pixel-level color mappings between shadow and non-shadow images, in which the joint modeling of lighting and object textures is implicit and inadequate. We observe that in a shadow region, the degradation degree of object textures depends on the local illumination, while simply enhancing the local illumination cannot fully recover the attenuated textures. Based on this observation, we propose to condition the restoration of attenuated textures on the corrected local lighting in the shadow region. Specifically, We first design a shadow-aware decomposition network to estimate the illumination and reflectance layers of shadow regions explicitly. We then propose a novel bilateral correction network to recast the lighting of shadow regions in the illumination layer via a novel local lighting correction module, and to restore the textures conditioned on the corrected illumination layer via a novel illumination-guided texture restoration module. We further annotate pixel-wise shadow masks for the public SRD dataset, which originally contains only image pairs. Experiments on three benchmarks show that our method outperforms existing state-of-the-art shadow removal methods. Project page in: yuhaoliu7456.github.io/RRL-Net.

NeurIPS Conference 2024 Conference Paper

RoleAgent: Building, Interacting, and Benchmarking High-quality Role-Playing Agents from Scripts

• Jiaheng Liu
• Zehao Ni
• Haoran Que
• Tao Sun
• Zekun Wang
• Jian Yang
• Jiakai Wang
• Hongcheng Guo

Believable agents can empower interactive applications ranging from immersive environments to rehearsal spaces for interpersonal communication. Recently, generative agents have been proposed to simulate believable human behavior by using Large Language Models. However, the existing method heavily relies on human-annotated agent profiles (e. g. , name, age, personality, relationships with others, and so on) for the initialization of each agent, which cannot be scaled up easily. In this paper, we propose a scalable RoleAgent framework to generate high-quality role-playing agents from raw scripts, which includes building and interacting stages. Specifically, in the building stage, we use a hierarchical memory system to extract and summarize the structure and high-level information of each agent for the raw script. In the interacting stage, we propose a novel innovative mechanism with four steps to achieve a high-quality interaction between agents. Finally, we introduce a systematic and comprehensive evaluation benchmark called RoleAgentBench to evaluate the effectiveness of our RoleAgent, which includes 100 and 28 roles for 20 English and 5 Chinese scripts, respectively. Extensive experimental results on RoleAgentBench demonstrate the effectiveness of RoleAgent.

AAAI Conference 2024 Conference Paper

ZOOM: Learning Video Mirror Detection with Extremely-Weak Supervision

• Ke Xu
• Tsun Wai Siu
• Rynson W.H. Lau

Mirror detection is an active research topic in computer vision. However, all existing mirror detectors learn mirror representations from large-scale pixel-wise datasets, which are tedious and expensive to obtain. Although weakly-supervised learning has been widely explored in related topics, we note that popular weak supervision signals (e.g., bounding boxes, scribbles, points) still require some efforts from the user to locate the target objects, with a strong assumption that the images to annotate always contain the target objects. Such an assumption may result in the over-segmentation of mirrors. Our key idea of this work is that the existence of mirrors over a time period may serve as a weak supervision to train a mirror detector, for two reasons. First, if a network can predict the existence of mirrors, it can essentially locate the mirrors. Second, we observe that the reflected contents of a mirror tend to be similar to those in adjacent frames, but exhibit considerable contrast to regions in far-away frames (e.g., non-mirror frames). To this end, in this paper, we propose ZOOM, the first method to learn robust mirror representations from extremely-weak annotations of per-frame ZerO-One Mirror indicators in videos. The key insight of ZOOM is to model the similarity and contrast (between mirror and non-mirror regions) in temporal variations to locate and segment the mirrors. To this end, we propose a novel fusion strategy to leverage temporal consistency information for mirror localization, and a novel temporal similarity-contrast modeling module for mirror segmentation. We construct a new video mirror dataset for training and evaluation. Experimental results under new and standard metrics show that ZOOM performs favorably against existing fully-supervised mirror detection methods.

JBHI Journal 2023 Journal Article

Deep Domain Adaptation Enhances Amplification Curve Analysis for Single-Channel Multiplexing in Real-Time PCR

• Ye Mao
• Ke Xu
• Luca Miglietta
• Louis Kreitmann
• Nicolas Moser
• Pantelis Georgiou
• Alison Holmes
• Jesus Rodriguez-Manzano

Data-driven approaches for molecular diagnostics are emerging as an alternative to perform an accurate and inexpensive multi-pathogen detection. A novel technique called Amplification Curve Analysis (ACA) has been recently developed by coupling machine learning and real-time Polymerase Chain Reaction (qPCR) to enable the simultaneous detection of multiple targets in a single reaction well. However, target classification purely relying on the amplification curve shapes faces several challenges, such as distribution discrepancies between different data sources (i. e. , training vs testing). Optimisation of computational models is required to achieve higher performance of ACA classification in multiplex qPCR through the reduction of those discrepancies. Here, we proposed a novel transformer-based conditional domain adversarial network (T-CDAN) to eliminate data distribution differences between the source domain (synthetic DNA data) and the target domain (clinical isolate data). The labelled training data from the source domain and unlabelled testing data from the target domain are fed into the T-CDAN, which learns both domains' information simultaneously. After mapping the inputs into a domain-irrelevant space, T-CDAN removes the feature distribution differences and provides a clearer decision boundary for the classifier, resulting in a more accurate pathogen identification. Evaluation of 198 clinical isolates containing three types of carbapenem-resistant genes ( bla NDM, bla IMP and bla OXA-48 ) illustrates a curve-level accuracy of 93. 1% and a sample-level accuracy of 97. 0% using T-CDAN, showing an accuracy improvement of 20. 9% and 4. 9% respectively. This research emphasises the importance of deep domain adaptation to enable high-level multiplexing in a single qPCR reaction, providing a solid approach to extend qPCR instruments' capabilities in real-world clinical applications.

AAAI Conference 2023 Conference Paper

Differentially Private Learning with Per-Sample Adaptive Clipping

• Tianyu Xia
• Shuheng Shen
• Su Yao
• Xinyi Fu
• Ke Xu
• Xiaolong Xu
• Xing Fu

Privacy in AI remains a topic that draws attention from researchers and the general public in recent years. As one way to implement privacy-preserving AI, differentially private learning is a framework that enables AI models to use differential privacy (DP). To achieve DP in the learning process, existing algorithms typically limit the magnitude of gradients with a constant clipping, which requires carefully tuned due to its significant impact on model performance. As a solution to this issue, latest works NSGD and Auto-S innovatively propose to use normalization instead of clipping to avoid hyperparameter tuning. However, normalization-based approaches like NSGD and Auto-S rely on a monotonic weight function, which imposes excessive weight on small gradient samples and introduces extra deviation to the update. In this paper, we propose a Differentially Private Per-Sample Adaptive Clipping (DP-PSAC) algorithm based on a non-monotonic adaptive weight function, which guarantees privacy without the typical hyperparameter tuning process of using a constant clipping while significantly reducing the deviation between the update and true batch-averaged gradient. We provide a rigorous theoretical convergence analysis and show that with convergence rate at the same order, the proposed algorithm achieves a lower non-vanishing bound, which is maintained over training iterations, compared with NSGD/Auto-S. In addition, through extensive experimental evaluation, we show that DP-PSAC outperforms or matches the state-of-the-art methods on multiple main-stream vision and language tasks.

JBHI Journal 2023 Journal Article

Dual-Stream Contrastive Learning for Channel State Information Based Human Activity Recognition

• Ke Xu
• Jiangtao Wang
• Le Zhang
• Hongyuan Zhu
• Dingchang Zheng

WiFi-based human activity recognition (HAR) has been extensively studied due to its far-reaching applications in health domains, including elderly monitoring, exercise supervision and rehabilitation monitoring, etc. Although existing supervised deep learning techniques have achieved remarkable performances for these tasks, they are however data-hungry and hence are notoriously difficult due to the privacy and incomprehensibility of WiFi-based HAR data. Existing contrastive learning models, mainly designed for computer vision, cannot guarantee their performance on channel state information (CSI) data. To this end, we propose a new dual-stream contrastive learning model that can process and learn the raw WiFi CSI data in a self-supervised manner. More specifically, our proposed method, coined as DualConFi, takes raw WiFI CSI data as input and incorporates channel and temporal streams to learn highly-discriminative spatiotemporal features under a mutual information constraint using unlabeled data. We exhibit the effectiveness of our model on three publicly available CSI data sets in various experiment settings, including linear evaluation, semi-supervised, and transfer learning. We show that DualConFi is able to perform favourably against challenging baselines in each setting. Moreover, by studying the effects of different transform functions on CSI data, we finally verify the effectiveness of highly-discriminative features.

ICLR Conference 2023 Conference Paper

Multimodal Federated Learning via Contrastive Representation Ensemble

• Qiying Yu
• Yang Liu 0165
• Yimu Wang
• Ke Xu
• Jingjing Liu

With the increasing amount of multimedia data on modern mobile systems and IoT infrastructures, harnessing these rich multimodal data without breaching user privacy becomes a critical issue. Federated learning (FL) serves as a privacy-conscious alternative to centralized machine learning. However, existing FL methods extended to multimodal data all rely on model aggregation on single modality level, which restrains the server and clients to have identical model architecture for each modality. This limits the global model in terms of both model complexity and data capacity, not to mention task diversity. In this work, we propose \textit{Contrastive Representation Ensemble and Aggregation for Multimodal FL (CreamFL)}, a multimodal federated learning framework that enables training larger server models from clients with heterogeneous model architectures and data modalities, while only communicating knowledge on public dataset. To achieve better multimodal representation fusion, we design a global-local cross-modal ensemble strategy to aggregate client representations. To mitigate local model drift caused by two unprecedented heterogeneous factors stemming from multimodal discrepancy (\textit{modality gap} and \textit{task gap}), we further propose two inter-modal and intra-modal contrasts to regularize local training, which complements information of the absent modality for uni-modal clients and regularizes local clients to head towards global consensus. Thorough evaluations and ablation studies on image-text retrieval and visual question answering tasks showcase the superiority of CreamFL over state-of-the-art FL methods and its practical value.

IJCAI Conference 2022 Conference Paper

A Simple yet Effective Method for Graph Classification

• Junran Wu
• Shangzhe Li
• Jianhao Li
• Yicheng Pan
• Ke Xu

In deep neural networks, better results can often be obtained by increasing the complexity of previously developed basic models. However, it is unclear whether there is a way to boost performance by decreasing the complexity of such models. Intuitively, given a problem, a simpler data structure comes with a simpler algorithm. Here, we investigate the feasibility of improving graph classification performance while simplifying the learning process. Inspired by structural entropy on graphs, we transform the data sample from graphs to coding trees, which is a simpler but essential structure for graph data. Furthermore, we propose a novel message passing scheme, termed hierarchical reporting, in which features are transferred from leaf nodes to root nodes by following the hierarchical structure of coding trees. We then present a tree kernel and a convolutional network to implement our scheme for graph classification. With the designed message passing scheme, the tree kernel and convolutional network have a lower runtime complexity of O(n) than Weisfeiler-Lehman subtree kernel and other graph neural networks of at least O(hm). We empirically validate our methods with several graph classification benchmarks and demonstrate that they achieve better performance and lower computational consumption than competing approaches.

IJCAI Conference 2022 Conference Paper

Document-level Relation Extraction via Subgraph Reasoning

• Xingyu Peng
• Chong Zhang
• Ke Xu

Document-level relation extraction aims to extract relations between entities in a document. In contrast to sentence-level relation extraction, it deals with longer texts and more complex entity interactions, which requires reasoning over multiple sentences with rich reasoning skills. Most current researches construct a document-level graph first, and then focus on the overall graph structure or the paths between the target entity pair in the graph. In this paper, we propose a novel subgraph reasoning (SGR) framework for document-level relation extraction. SGR combines the advantages of both graph-based models and path-based models, integrating various paths between the target entity pair into a much simpler subgraph structure to perform relational reasoning. Moreover, the paths generated by our designed heuristic strategy explicitly model the requisite reasoning skills and roughly cover the supporting sentences for each relation instance. Experimental results on DocRED show that SGR outperforms existing models, and further analyses demonstrate that our method is both effective and explainable. Our code is available at https: //github. com/Crysta1ovo/SGR.

ICLR Conference 2022 Conference Paper

Exploring Memorization in Adversarial Training

• Yinpeng Dong
• Ke Xu
• Xiao Yang 0028
• Tianyu Pang
• Zhijie Deng
• Hang Su 0006
• Jun Zhu 0001

Deep learning models have a propensity for fitting the entire training set even with random labels, which requires memorization of every training sample. In this paper, we explore the memorization effect in adversarial training (AT) for promoting a deeper understanding of model capacity, convergence, generalization, and especially robust overfitting of the adversarially trained models. We first demonstrate that deep networks have sufficient capacity to memorize adversarial examples of training data with completely random labels, but not all AT algorithms can converge under the extreme circumstance. Our study of AT with random labels motivates further analyses on the convergence and generalization of AT. We find that some AT approaches suffer from a gradient instability issue and the recently suggested complexity measures cannot explain robust generalization by considering models trained on random labels. Furthermore, we identify a significant drawback of memorization in AT that it could result in robust overfitting. We then propose a new mitigation algorithm motivated by detailed memorization analyses. Extensive experiments on various datasets validate the effectiveness of the proposed method.

NeurIPS Conference 2022 Conference Paper

Geometry-aware Two-scale PIFu Representation for Human Reconstruction

• Zheng Dong
• Ke Xu
• Ziheng Duan
• Hujun Bao
• Weiwei Xu
• Rynson Lau

Although PIFu-based 3D human reconstruction methods are popular, the quality of recovered details is still unsatisfactory. In a sparse (e. g. , 3 RGBD sensors) capture setting, the depth noise is typically amplified in the PIFu representation, resulting in flat facial surfaces and geometry-fallible bodies. In this paper, we propose a novel geometry-aware two-scale PIFu for 3D human reconstruction from sparse, noisy inputs. Our key idea is to exploit the complementary properties of depth denoising and 3D reconstruction, for learning a two-scale PIFu representation to reconstruct high-frequency facial details and consistent bodies separately. To this end, we first formulate depth denoising and 3D reconstruction as a multi-task learning problem. The depth denoising process enriches the local geometry information of the reconstruction features, while the reconstruction process enhances depth denoising with global topology information. We then propose to learn the two-scale PIFu representation using two MLPs based on the denoised depth and geometry-aware features. Extensive experiments demonstrate the effectiveness of our approach in reconstructing facial details and bodies of different poses and its superiority over state-of-the-art methods.

YNIMG Journal 2022 Journal Article

Informed MEG/EEG source imaging reveals the locations of interictal spikes missed by SEEG

• Su Shu
• Shen Luo
• Miao Cao
• Ke Xu
• Lang Qin
• Li Zheng
• Jing Xu
• Xiongfei Wang

Determining the accurate locations of interictal spikes has been fundamental in the presurgical evaluation of epilepsy surgery. Stereo-electroencephalography (SEEG) is able to directly record cortical activity and localize interictal spikes. However, the main caveat of SEEG techniques is that they have limited spatial sampling (covering <5% of the whole brain), which may lead to missed spikes originating from brain regions that were not covered by SEEG. To address this problem, we propose a SEEG-informed minimum-norm estimates (SIMNE) method by combining SEEG with magnetoencephalography (MEG) or EEG. Specifically, the spike locations determined by SEEG offer as a priori information to guide MEG source reconstruction. Both computer simulations and experiments using data from five epilepsy patients were conducted to evaluate the performance of SIMNE. Our results demonstrate that SIMNE generates more accurate source estimation than a traditional minimum-norm estimates method and reveals the locations of spikes missed by SEEG, which would improve presurgical evaluation of the epileptogenic zone.

IJCAI Conference 2022 Conference Paper

MultiQuant: Training Once for Multi-bit Quantization of Neural Networks

• Ke Xu
• Qiantai Feng
• Xingyi Zhang
• Dong Wang

Quantization has become a popular technique to compress deep neural networks (DNNs) and reduce computational costs, but most prior work focuses on training DNNs at each individual fixed bit-width and accuracy trade-off point. How to produce a model with flexible precision is largely unexplored. This work proposes a multi-bit quantization framework (MultiQuant) to make the learned DNNs robust for different precision configuration during inference by adopting Lowest-Random-Highest bit-width co-training method. Meanwhile, we propose an online adaptive label generation strategy to alleviate the problem of vicious competition under different precision caused by one-hot labels in the supernet training. The trained supernet model can be flexibly set to different bit widths to support dynamic speed and accuracy trade-off. Furthermore, we adopt the Monte Carlo sampling-based genetic algorithm search strategy with quantization-aware accuracy predictor as evaluation criterion to incorporate the mixed precision technology in our framework. Experiment results on ImageNet datasets demonstrate MultiQuant method can attain the quantization results under different bit-widths comparable with quantization-aware training without retraining.

IJCAI Conference 2022 Conference Paper

SatFormer: Saliency-Guided Abnormality-Aware Transformer for Retinal Disease Classification in Fundus Image

• Yankai Jiang
• Ke Xu
• Xinyue Wang
• Yuan Li
• Hongguang Cui
• Yubo Tao
• Hai Lin

Automatic and accurate retinal disease diagnosis is critical to guide proper therapy and prevent potential vision loss. Previous works simply exploit the most discriminative features while ignoring the pathological visual clues of scattered subtle lesions. Therefore, without a comprehensive understanding of features from different lesion regions, they are vulnerable to noise from complex backgrounds and suffer from misclassification failures. In this paper, we address these limitations with a novel saliency-guided abnormality-aware transformer which explicitly captures the correlation between different lesion features from a global perspective with enhanced pathological semantics. The model has several merits. First, we propose a saliency enhancement module (SEM) which adaptively integrates disease related semantics and highlights potentially salient lesion regions. Second, to the best of our knowledge, this is the first work to explore comprehensive lesion feature dependencies via a tailored efficient self-attention. Third, with the saliency enhancement module and abnormality-aware attention, we propose a new variant of Vision Transformer models, called SatFormer, which outperforms the state-of-the-art methods on two public retinal disease classification benchmarks. Ablation study shows that the proposed components can be easily embedded into any Vision Transformers via a plug-and-play manner and effectively boost the performance.

JBHI Journal 2021 Journal Article

Cancelable HD-sEMG-Based Biometrics for Cross-Application Discrepant Personal Identification

• Xinyu Jiang
• Ke Xu
• Xiangyu Liu
• Chenyun Dai
• David A. Clifton
• Edward A. Clancy
• Metin Akay
• Wei Chen

With the soaring development of body sensor network (BSN)-based health informatics, information security in such medical devices has attracted increasing attention in recent years. Employing the biosignals acquired directly by the BSN as biometrics for personal identification is an effective approach. Noncancelability and cross-application invariance are two natural flaws of most traditional biometric modalities. Once the biometric template is exposed, it is compromised forever. Even worse, because the same biometrics may be employed as tokens for different accounts in multiple applications, the exposed template can be used to compromise other accounts. In this work, we propose a cancelable and cross-application discrepant biometric approach based on high-density surface electromyogram (HD-sEMG) for personal identification. We enrolled two accounts for each user. HD-sEMG signals from the right dorsal hand under isometric contractions of different finger muscles were employed as biometric tokens. Since isometric contraction, in contrast to dynamic contraction, requires no actual movement, the users’ choice to login to different accounts is greatly protected against impostors. We realized a promising identification accuracy of 85. 8% for 44 identities (22 subjects × 2 accounts) with training and testing data acquired 9 days apart. The high identification accuracy of different accounts for the same user demonstrates the promising cancelability and cross-application discrepancy of the proposed HD-sEMG-based biometrics. To the best of our knowledge, this is the first study to employ HD-sEMG in personal identification applications, with signal variation across days considered.

AAAI Conference 2021 Short Paper

Incorporating Curiosity into Personalized Ranking for Collaborative Filtering (Student Abstract)

• Qiqi Ding
• Yi Cai
• Ke Xu
• Huakui Zhang

Curiosity affects the users’ selections of the items, and it motivates them to explore the items regardless of their preferences. This phenomenon is particularly common in the social networks. However, the existing social-based recommendation methods neglect users’ curiosity in the social networks, and it may cause the accuracy decease in recommendation. What’s more, only focusing on simulating the users’ preferences can lead to information cocoons. In order to tackle the problems, we propose a Curiosity Enhanced Bayesian Personalized Ranking (CBPR) model. Our model makes full use of the theories of psychology to model the users’ curiosity aroused when facing different opinions. The experimental results on two public datasets demonstrate the advantages of our CBPR model over the existing models.

JBHI Journal 2021 Journal Article

Quantifying Spatial Activation Patterns of Motor Units in Finger Extensor Muscles

• Xinyu Jiang
• Haoran Ren
• Ke Xu
• Xinming Ye
• Chenyun Dai
• Edward A. Clancy
• Yuan-Ting Zhang
• Wei Chen

The ability to expertly control different fingers contributes to hand dexterity during object manipulation in daily life activities. The macroscopic spatial patterns of muscle activations during finger movements using global surface electromyography (sEMG) have been widely researched. However, the spatial activation patterns of microscopic motor units (MUs) under different finger movements have not been well investigated. The present work aims to quantify MU spatial activation patterns during movement of distinct fingers (index, middle, ring and little finger). Specifically, we focused on extensor muscles during extension contractions. Motor unit action potentials (MUAPs) during movement of each finger were obtained through decomposition of high-density sEMG (HD-sEMG). First, we quantified the spatial activation patterns of MUs for each finger based on 2-dimension (2-D) root-mean-square (RMS) maps of MUAP grids after spike-triggered averaging. We found that these activation patterns under different finger movements are distinct along the distal-proximal direction, but with partial overlap. Second, to further evaluate MU separability, we classified the spatial activation pattern of each individual MU under distinct finger movement and associated each MU with its corresponding finger with Regularized Uncorrelated Multilinear Discriminant Analysis (RUMLDA). A high accuracy of MU-finger classification tested on 12 subjects with a mean of 88. 98% was achieved. The quantification of MU spatial activation patterns could be beneficial to studies of neural mechanisms of the hand. To the best of our knowledge, this is the first work which manages to quantify MU behaviors under different finger movements.

AAAI Conference 2021 Conference Paper

Self-Attention Attribution: Interpreting Information Interactions Inside Transformer

• Yaru Hao
• Li Dong
• Furu Wei
• Ke Xu

The great success of Transformer-based models benefits from the powerful multi-head self-attention mechanism, which learns token dependencies and encodes contextual information from the input. Prior work strives to attribute model decisions to individual input features with different saliency measures, but they fail to explain how these input features interact with each other to reach predictions. In this paper, we propose a self-attention attribution method to interpret the information interactions inside Transformer. We take BERT as an example to conduct extensive studies. Firstly, we apply selfattention attribution to identify the important attention heads, while others can be pruned with marginal performance degradation. Furthermore, we extract the most salient dependencies in each layer to construct an attribution tree, which reveals the hierarchical interactions inside Transformer. Finally, we show that the attribution results can be used as adversarial patterns to implement non-targeted attacks towards BERT.

NeurIPS Conference 2020 Conference Paper

BERT Loses Patience: Fast and Robust Inference with Early Exit

• Wangchunshu Zhou
• Canwen Xu
• Tao Ge
• Julian McAuley
• Ke Xu
• Furu Wei

In this paper, we propose Patience-based Early Exit, a straightforward yet effective inference method that can be used as a plug-and-play technique to simultaneously improve the efficiency and robustness of a pretrained language model (PLM). To achieve this, our approach couples an internal-classifier with each layer of a PLM and dynamically stops inference when the intermediate predictions of the internal classifiers do not change for a pre-defined number of steps. Our approach improves inference efficiency as it allows the model to make a prediction with fewer layers. Meanwhile, experimental results with an ALBERT model show that our method can improve the accuracy and robustness of the model by preventing it from overthinking and exploiting multiple classifiers for prediction, yielding a better accuracy-speed trade-off compared to existing early exit methods.

AAAI Conference 2020 Conference Paper

Learning to Compare for Better Training and Evaluation of Open Domain Natural Language Generation Models

• Wangchunshu Zhou
• Ke Xu

Automated evaluation of open domain natural language generation (NLG) models remains a challenge and widely used metrics such as BLEU and Perplexity can be misleading in some cases. In our paper, we propose to evaluate natural language generation models by learning to compare a pair of generated sentences by fine-tuning BERT, which has been shown to have good natural language understanding ability. We also propose to evaluate the model-level quality of NLG models with sample-level comparison results with skill rating system. While able to be trained in a fully self-supervised fashion, our model can be further fine-tuned with a little amount of human preference annotation to better imitate human judgment. In addition to evaluating trained models, we propose to apply our model as a performance indicator during training for better hyperparameter tuning and early-stopping. We evaluate our approach on both story generation and chitchat dialogue response generation. Experimental results show that our model correlates better with human preference compared with previous automated evaluation approaches. Training with the proposed metric yields better performance in human evaluation, which further demonstrates the effectiveness of the proposed model.

IJCAI Conference 2020 Conference Paper

Relation-Aware Transformer for Portfolio Policy Learning

• Ke Xu
• Yifan Zhang
• Deheng Ye
• Peilin Zhao
• Mingkui Tan

Portfolio selection is an important yet challenging task in AI for FinTech. One of the key issues is how to represent the non-stationary price series of assets in a portfolio, which is important for portfolio decisions. The existing methods, however, fall short of capturing: 1) the complicated sequential patterns for asset price series and 2) the price correlations among multiple assets. In this paper, under a deep reinforcement learning paradigm for portfolio selection, we propose a novel Relation-aware Transformer (RAT) to handle these aspects. Specifically, being equipped with our newly developed attention modules, RAT is structurally innovated to capture both sequential patterns and asset correlations for portfolio selection. Based on the extracted sequential features, RAT is able to make profitable portfolio decisions regarding each asset via a newly devised leverage operation. Extensive experiments on real-world crypto-currency and stock datasets verify the state-of-the-art performance of RAT.

YNIMG Journal 2019 Journal Article

Reproducibility of functional brain alterations in major depressive disorder: Evidence from a multisite resting-state functional MRI study with 1,434 individuals

• Mingrui Xia
• Tianmei Si
• Xiaoyi Sun
• Qing Ma
• Bangshan Liu
• Li Wang
• Jie Meng
• Miao Chang

Resting-state functional MRI (R-fMRI) studies have demonstrated widespread alterations in brain function in patients with major depressive disorder (MDD). However, a clear and consistent conclusion regarding a repeatable pattern of MDD-relevant alterations is still limited due to the scarcity of large-sample, multisite datasets. Here, we address this issue by including a large R-fMRI dataset with 1434 participants (709 patients with MDD and 725 healthy controls) from five centers in China. Individual functional activity maps that represent very local to long-range connections are computed using the amplitude of low-frequency fluctuations, regional homogeneity and distance-related functional connectivity strength. The reproducibility analyses involve different statistical strategies, global signal regression, across-center consistency, clinical variables, and sample size. We observed significant hypoactivity in the orbitofrontal, sensorimotor, and visual cortices and hyperactivity in the frontoparietal cortices in MDD patients compared to the controls. These alterations are not affected by different statistical analysis strategies, global signal regression and medication status and are generally reproducible across centers. However, these between-group differences are partially influenced by the episode status and the age of disease onset in patients, and the brain-clinical variable relationship exhibits poor cross-center reproducibility. Bootstrap analyses reveal that at least 400 subjects in each group are required to replicate significant alterations (an extent threshold of P <. 05 and a height threshold of P <. 001) at 50% reproducibility. Together, these results highlight reproducible patterns of functional alterations in MDD and relevant influencing factors, which provides crucial guidance for future neuroimaging studies of this disorder.

NeurIPS Conference 2018 Conference Paper

Active Matting

• Xin Yang
• Ke Xu
• Shaozhe Chen
• Shengfeng He
• Baocai Yin Yin
• Rynson Lau

Image matting is an ill-posed problem. It requires a user input trimap or some strokes to obtain an alpha matte of the foreground object. A fine user input is essential to obtain a good result, which is either time consuming or suitable for experienced users who know where to place the strokes. In this paper, we explore the intrinsic relationship between the user input and the matting algorithm to address the problem of where and when the user should provide the input. Our aim is to discover the most informative sequence of regions for user input in order to produce a good alpha matte with minimum labeling efforts. To this end, we propose an active matting method with recurrent reinforcement learning. The proposed framework involves human in the loop by sequentially detecting informative regions for trivial human judgement. Comparing to traditional matting algorithms, the proposed framework requires much less efforts, and can produce satisfactory results with just 10 regions. Through extensive experiments, we show that the proposed model reduces user efforts significantly and achieves comparable performance to dense trimaps in a user-friendly manner. We further show that the learned informative knowledge can be generalized across different matting algorithms.

AAAI Conference 2018 Short Paper

Solving Generalized Column Subset Selection With Heuristic Search

• Swair Shah
• Baokun He
• Ke Xu
• Crystal Maung
• Haim Schweitzer

We address the problem of approximating a matrix by the linear combination of a column sparse matrix and a low rank matrix. Two variants of a heuristic search algorithm are described. The first produces an optimal solution but may be slow, as these problems are believed to be NP-hard. The second is much faster, but only guarantees a suboptimal solution. The quality of the approximation and the optimality criterion can be specified in terms of unitarily invariant norms.

AAAI Conference 2017 Conference Paper

Cleaning the Null Space: A Privacy Mechanism for Predictors

• Ke Xu
• Tongyi Cao
• Swair Shah
• Crystal Maung
• Haim Schweitzer

In standard machine learning and regression setting feature values are used to predict some desired information. The privacy challenge considered here is to prevent an adversary from using available feature values to predict confidential information that one wishes to keep secret. We show that this can sometimes be achieved with almost no effect on the quality of predicting desired information. We describe two algorithms aimed at providing such privacy when the predictors have a linear operator in the first stage. The desired effect can be achieved by zeroing out feature components in the approximate null space of the linear operator.

AAAI Conference 2017 Short Paper

Enhancing the Privacy of Predictors

• Ke Xu
• Swair Shah
• Tongyi Cao
• Crystal Maung
• Haim Schweitzer

The privacy challenge considered here is to prevent an adversary from using available feature values to predict confi- dential information. We propose an algorithm providing such privacy for predictors that have a linear operator in the first stage. Privacy is achieved by zeroing out feature components in the approximate null space of the linear operator. We show that this has little effect on predicting desired information.

JAIR Journal 2016 Journal Article

An Exact Algorithm Based on MaxSAT Reasoning for the Maximum Weight Clique Problem

• Zhiwen Fang
• Chu-Min Li
• Ke Xu

Recently, MaxSAT reasoning is shown very effective in computing a tight upper bound for a Maximum Clique (MC) of a (unweighted) graph. In this paper, we apply MaxSAT reasoning to compute a tight upper bound for a Maximum Weight Clique (MWC) of a wighted graph. We first study three usual encodings of MWC into weighted partial MaxSAT dealing with hard clauses, which must be satisfied in all solutions, and soft clauses, which are weighted and can be falsified. The drawbacks of these encodings motivate us to propose an encoding of MWC into a special weighted partial MaxSAT formalism, called LW (Literal-Weighted) encoding and dedicated for upper bounding an MWC, in which both soft clauses and literals in soft clauses are weighted. An optimal solution of the LW MaxSAT instance gives an upper bound for an MWC, instead of an optimal solution for MWC. We then introduce two notions called the Top-k literal failed clause and the Top-k empty clause to extend classical MaxSAT reasoning techniques, as well as two sound transformation rules to transform an LW MaxSAT instance. Successive transformations of an LW MaxSAT instance driven by MaxSAT reasoning give a tight upper bound for the encoded MWC. The approach is implemented in a branch-and-bound algorithm called MWCLQ. Experimental evaluations on the broadly used DIMACS benchmark, BHOSLIB benchmark, random graphs and the benchmark from the winner determination problem show that our approach allows MWCLQ to reduce the search space significantly and to solve MWC instances effectively. Consequently, MWCLQ outperforms state-of-the-art exact algorithms on the vast majority of instances. Moreover, it is surprisingly effective in solving hard and dense instances.

NeurIPS Conference 2016 Conference Paper

Learning Influence Functions from Incomplete Observations

• Xinran He
• Ke Xu
• David Kempe
• Yan Liu

We study the problem of learning influence functions under incomplete observations of node activations. Incomplete observations are a major concern as most (online and real-world) social networks are not fully observable. We establish both proper and improper PAC learnability of influence functions under randomly missing observations. Proper PAC learnability under the Discrete-Time Linear Threshold (DLT) and Discrete-Time Independent Cascade (DIC) models is established by reducing incomplete observations to complete observations in a modified graph. Our improper PAC learnability result applies for the DLT and DIC models as well as the Continuous-Time Independent Cascade (CIC) model. It is based on a parametrization in terms of reachability features, and also gives rise to an efficient and practical heuristic. Experiments on synthetic and real-world datasets demonstrate the ability of our method to compensate even for a fairly large fraction of missing observations.

AAAI Conference 2016 Conference Paper

Unsupervised Feature Selection by Heuristic Search with Provable Bounds on Suboptimality

• Hiromasa Arai
• Crystal Maung
• Ke Xu
• Haim Schweitzer

Identifying a small number of features that can represent the data is a known problem that comes up in areas such as machine learning, knowledge representation, data mining, and numerical linear algebra. Computing an optimal solution is believed to be NP-hard, and there is extensive work on approximation algorithms. Classic approaches exploit the algebraic structure of the underlying matrix, while more recent approaches use randomization. An entirely different approach that uses the A* heuristic search algorithm to find an optimal solution was recently proposed. Not surprisingly it is limited to effectively selecting only a small number of features. We propose a similar approach related to the Weighted A* algorithm. This gives algorithms that are not guaranteed to find an optimal solution but run much faster than the A* approach, enabling effective selection of many features from large datasets. We demonstrate experimentally that these new algorithms are more accurate than the current state-of-the-art while still being practical. Furthermore, they come with an adjustable guarantee on how different their error may be from the smallest possible (optimal) error. Their accuracy can always be increased at the expense of a longer running time.

AAAI Conference 2016 Conference Paper

Weighted A* Algorithms for Unsupervised Feature Selection with Provable Bounds on Suboptimality

• Hiromasa Arai
• Ke Xu
• Crystal Maung
• Haim Schweitzer

IJCAI Conference 2015 Conference Paper

A Hybrid Neural Model for Type Classification of Entity Mentions

• Li Dong
• Furu Wei
• Hong Sun
• Ming Zhou
• Ke Xu

The semantic class (i. e. , type) of an entity plays a vital role in many natural language processing tasks, such as question answering. However, most of existing type classification systems extensively rely on hand-crafted features. This paper introduces a hybrid neural model which classifies entity mentions to a wide-coverage set of 22 types derived from DBpedia. It consists of two parts. The mention model uses recurrent neural networks to recursively obtain the vector representation of an entity mention from the words it contains. The context model, on the other hand, employs multilayer perceptrons to obtain the hidden representation for contextual information of a mention. Representations obtained by the two parts are used together to predict the type distribution. Using automatically generated data, these two parts are jointly learned. Experimental studies illustrate that the proposed approach outperforms baseline methods. Moreover, when type information provided by our method is used in a question answering system, we observe a 14. 7% relative improvement for the top-1 accuracy of answers.

AAAI Conference 2014 Conference Paper

Adaptive Multi-Compositionality for Recursive Neural Models with Applications to Sentiment Analysis

• Li Dong
• Furu Wei
• Ming Zhou
• Ke Xu

Recursive neural models have achieved promising results in many natural language processing tasks. The main difference among these models lies in the composition function, i. e. , how to obtain the vector representation for a phrase or sentence using the representations of words it contains. This paper introduces a novel Adaptive Multi-Compositionality (AdaMC) layer to recursive neural models. The basic idea is to use more than one composition functions and adaptively select them depending on the input vectors. We present a general framework to model each semantic composition as a distribution over these composition functions. The composition functions and parameters used for adaptive selection are learned jointly from data. We integrate AdaMC into existing recursive neural models and conduct extensive experiments on the Stanford Sentiment Treebank. The results illustrate that AdaMC significantly outperforms state-of-the-art sentiment classification methods. It helps push the best accuracy of sentence-level negative/positive classification from 85. 4% up to 88. 5%.

TCS Journal 2014 Journal Article

Circular convex bipartite graphs: Feedback vertex sets

• Tian Liu
• Min Lu
• Zhao Lu
• Ke Xu

A feedback vertex set is a subset of vertices, such that the removal of this subset renders the remaining graph cycle-free. The weight of a feedback vertex set is the sum of weights of its vertices. Finding a minimum weighted feedback vertex set is tractable for convex bipartite graphs, but NP -complete even for unweighted bipartite graphs. In a circular convex (convex, respectively) bipartite graph, there is a circular (linear, respectively) ordering defined on one class of vertices, such that for every vertex in another class, the neighborhood of this vertex is a circular arc (an interval, respectively). The minimum weighted feedback vertex set problem is shown tractable for circular convex bipartite graphs in this paper, by making a Cook reduction (i. e. polynomial time Turing reduction) for this problem from circular convex bipartite graphs to convex bipartite graphs.

TCS Journal 2013 Journal Article

Feedback vertex sets on restricted bipartite graphs

• Wei Jiang
• Tian Liu
• Chaoyi Wang
• Ke Xu

A feedback vertex set (FVS) in a graph is a subset of vertices whose complement induces a forest. Finding a minimum FVS is NP -complete on bipartite graphs, but tractable on convex bipartite graphs and on chordal bipartite graphs. A bipartite graph is called tree convex, if a tree is defined on one part of the vertices, such that for every vertex in the other part, its neighborhood induces a subtree. When the tree is a path, a triad or a star, the bipartite graph is called convex bipartite, triad convex bipartite or star convex bipartite, respectively. We show that: (1) FVS is tractable on triad convex bipartite graphs; (2) FVS is NP -complete on star convex bipartite graphs and on tree convex bipartite graphs where the maximum degree of vertices on the tree is at most three.

AAAI Conference 2013 Conference Paper

The Automated Acquisition of Suggestions from Tweets

• Li Dong
• Furu Wei
• Yajuan Duan
• Xiaohua Liu
• Ming Zhou
• Ke Xu

This paper targets at automatically detecting and classifying user’s suggestions from tweets. The short and informal nature of tweets, along with the imbalanced characteristics of suggestion tweets, makes the task extremely challenging. To this end, we develop a classification framework on Factorization Machines, which is effective and efficient especially in classification tasks with feature sparsity settings. Moreover, we tackle the imbalance problem by introducing cost-sensitive learning techniques in Factorization Machines. Extensively experimental studies on a manually annotated real-life data set show that the proposed approach significantly improves the baseline approach, and yields the precision of 71. 06% and recall of 67. 86%. We also investigate the reason why Factorization Machines perform better. Finally, we introduce the first manually annotated dataset for suggestion classification.

AIJ Journal 2012 Journal Article

A general model and thresholds for random constraint satisfaction problems

• Yun Fan
• Jing Shen
• Ke Xu

In this paper, we study the relation among the parameters in their most general setting that define a large class of random CSP models d-k-CSP where d is the domain size and k is the length of the constraint scopes. The model d-k-CSP unifies several related models such as the model RB and the model k-CSP. We prove that the model d-k-CSP exhibits exact phase transitions if k ln d increases no slower than the logarithm of the number of variables. A series of experimental studies with interesting observations are carried out to illustrate the solubility phase transition and the hardness of instances around phase transitions.

AAAI Conference 2011 Conference Paper

Exact Phase Transitions and Approximate Algorithm of #CSP

• Ping Huang
• Minghao Yin
• Ke Xu

IJCAI Conference 2011 Conference Paper

Large Hinge Width on Sparse Random Hypergraphs

• Tian Liu
• Xiaxiang Lin
• Chaoyi Wang
• Kaile Su
• Ke Xu

Consider random hypergraphs on n vertices, where each k-element subset of vertices is selected with probability p independently and randomly as a hyperedge. By sparse we mean that the total number of hyperedges is O(n) or O(n ln n). When k = 2, these are exactly the classical Erd&ouml; s-R&eacute; nyi random graphs G(n, p). We prove that with high probability, hinge width on these sparse random hypergraphs can grow linearly with the expected number of hyperedges. Some random constraint satisfaction problems such as Model RB and Model RD have satisfiability thresholds on these sparse constraint hypergraphs, thus the large hinge width results provide some theoretical evidence for random instances around satisfiability thresholds to be hard for a standard hinge-decomposition based algorithm. We also conduct experiments on these and other kinds of random graphs with several hundreds vertices, including regular random graphs and power law random graphs. The experimental results also show that hinge width can grow linearly with the number of edges on these different random graphs. These results may be of further interests.

AIJ Journal 2007 Journal Article

Random constraint satisfaction: Easy generation of hard (satisfiable) instances

• Ke Xu
• Frédéric Boussemart
• Fred Hemery
• Christophe Lecoutre

In this paper, we show that the models of random CSP instances proposed by Xu and Li [K. Xu, W. Li, Exact phase transitions in random constraint satisfaction problems, Journal of Artificial Intelligence Research 12 (2000) 93–103; K. Xu, W. Li, Many hard examples in exact phase transitions with application to generating hard satisfiable instances, Technical report, CoRR Report cs. CC/0302001, Revised version in Theoretical Computer Science 355 (2006) 291–302] are of theoretical and practical interest. Indeed, these models, called RB and RD, present several nice features. First, it is quite easy to generate random instances of any arity since no particular structure has to be integrated, or property enforced, in such instances. Then, the existence of an asymptotic phase transition can be guaranteed while applying a limited restriction on domain size and on constraint tightness. In that case, a threshold point can be precisely located and all instances have the guarantee to be hard at the threshold, i. e. , to have an exponential tree-resolution complexity. Next, a formal analysis shows that it is possible to generate forced satisfiable instances whose hardness is similar to unforced satisfiable ones. This analysis is supported by some representative results taken from an intensive experimentation that we have carried out, using complete and incomplete search methods.

TCS Journal 2006 Journal Article

Many hard examples in exact phase transitions

• Ke Xu
• Wei Li

This paper analyzes the resolution complexity of two random constraint satisfaction problem (CSP) models (i. e. Model RB/RD) for which we can establish the existence of phase transitions and identify the threshold points exactly. By encoding CSPs into CNF formulas, it is proved that almost all instances of Model RB/RD have no tree-like resolution proofs of less than exponential size. Thus, we not only introduce new families of CSPs and CNF formulas hard to solve, which can be useful in the experimental evaluation of CSP and SAT algorithms, but also propose models with both many hard instances and exact phase transitions. Finally, conclusions are presented, as well as a detailed comparison of Model RB/RD with the Hamiltonian cycle problem and random 3-SAT, which, respectively, exhibit three different kinds of phase transition behavior in NP-complete problems.

AAAI Conference 2005 Conference Paper

A Domain-Independent System for Case-Based Task Decomposition without Domain Theories

• Ke Xu

We propose using domain-independent task decomposition techniques for situations in which cases are the sole or the main source for domain knowledge. Our work is motivated by project planning domains, where hierarchical cases are readily available, but neither a planning domain theory nor case adaptation knowledge is available. We present DInCaD (Domain-Independent System for Case-Based Task Decomposition), a system that encompasses case retrieval, refinement, and reuse, following from the idea of reusing generalized cases to solve new problems. DInCaD consists of a case refinement procedure that reduces case overgeneralization, and a similarity criterion that takes advantage of the refinement to improve case retrieval precision. We will analyze the properties of the system, and present an empirical evaluation.

IJCAI Conference 2005 Conference Paper

A Simple Model to Generate Hard Satisfiable Instances

• Ke Xu
• Frédéric Boussemart
• Fred Hemery
• Christophe

In this paper, we try to further demonstrate that the models of random CSP instances proposed by [Xu and Li, 2000; 2003] are of theoretical and practical interest. Indeed, these models, called RB and RD, present several nice features. First, it is quite easy to generate random instances of any arity since no particular structure has to be integrated, or property enforced, in such instances. Then, the existence of an asymptotic phase transition can be guaranteed while applying a limited restriction on domain size and on constraint tightness. In that case, a threshold point can be precisely located and all instances have the guarantee to be hard at the threshold, i. e. , to have an exponential tree-resolution complexity. Next, a formal analysis shows that it is possible to generate forced satisfiable instances whose hardness is similar to unforced satisfiable ones. This analysis is supported by some representative results taken from an intensive experimentation that we have carried out, using complete and incomplete search methods.