Author name cluster

Wei Shi

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

12 papers

2 author rows

AAMAS Conference 2026 Conference Paper

Efficient Device-Cloud Collaborative Offline-to-Online Reinforcement Learning

Honglan Huang
Wei Shi
Chaoyue Niu

Federated reinforcement learning in device-cloud architectures suffers from high interaction costs, low sample efficiency, and slow convergence due to device constraints. We propose a data-centric device-cloud collaborative training method that pre-trains a global model on the cloud to provide a high-quality initial policy and selects high-value samples to guide safe and efficient local policy finetuning on devices. Experiments on public benchmarks show our method achieves significantly faster convergence, higher training efficiency, and improved policy stability, outperforming baselines.

PDF

AAAI Conference 2026 Conference Paper

HuiduRep: A Robust Self-Supervised Framework for Learning Neural Representations from Extracellular Recordings

Feng Cao
Zishuo Feng
Jicong Zhang
Wei Shi

Extracellular recordings are transient voltage fluctuations in the vicinity of neurons, serving as a fundamental modality in neuroscience for decoding brain activity at single-neuron resolution. Spike sorting, the process of attributing each detected spike to its corresponding neuron, is a pivotal step in brain sensing pipelines. However, it remains challenging under low signal-to-noise ratio (SNR), electrode drift, and cross-session variability. In this paper, we propose HuiduRep, a robust self-supervised representation learning framework that extracts discriminative and generalizable features from extracellular recordings. By integrating contrastive learning with a denoising autoencoder, HuiduRep learns latent representations robust to noise and drift. With HuiduRep, we develop a spike sorting pipeline that clusters spike representations without ground truth labels. Experiments on hybrid and real-world datasets demonstrate that HuiduRep achieves strong robustness. Furthermore, the pipeline significantly outperforms state-of-the-art tools such as KiloSort4 and MountainSort5 on accuracy and precision on diverse datasets. These findings demonstrate the potential of self-supervised spike representation learning as a foundational tool for robust and generalizable processing of extracellular recordings.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Interpretable Reward Model via Sparse Autoencoder

Shuyi Zhang
Wei Shi
Sihang Li
Jiayi Liao
Tao Liang
Hengxing Cai
Xiang Wang

Large language models (LLMs) have been widely deployed across numerous fields. Reinforcement Learning from Human Feedback (RLHF) leverages reward models (RMs) as proxies for human preferences to align LLM behaviors with human values, making the accuracy, reliability, and interpretability of RMs critical for effective alignment. However, traditional RMs lack interpretability, offer limited insight into the reasoning behind reward assignments, and are inflexible toward user preference shifts. While recent multidimensional RMs aim for improved interpretability, they often fail to provide feature-level attribution and require costly annotations. To overcome these limitations, we introduce the Sparse Autoencoder-Enhanced Reward Model (SARM), a novel architecture that integrates a pretrained Sparse Autoencoder (SAE) into a reward model. SARM maps the hidden activations of LLM-based RM into an interpretable, sparse, and monosemantic feature space, from which a scalar head aggregates feature activations to produce transparent and conceptually meaningful reward scores. Empirical evaluations demonstrate that SARM facilitates direct feature-level attribution of reward assignments, allows dynamic adjustment to preference shifts, and achieves superior alignment performance compared to conventional reward models.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

A Survey of Optimization Modeling Meets LLMs: Progress and Future Directions

Ziyang Xiao
Jingrong Xie
Lilin Xu
Shisi Guan
Jingyan Zhu
Xiongwei Han
Xiaojin Fu
WingYin Yu

By virtue of its great utility in solving real-world problems, optimization modeling has been widely employed for optimal decision-making across various sectors, but it requires substantial expertise from operations research professionals. With the advent of large language models (LLMs), new opportunities have emerged to automate the procedure of mathematical modeling. This survey presents a comprehensive and timely review of recent advancements that cover the entire technical stack, including data synthesis and fine-tuning for the base model, inference frameworks, benchmark datasets, and performance evaluation. In addition, we conducted an in-depth analysis on the quality of benchmark datasets, which was found to have a surprisingly high error rate. We cleaned the datasets and constructed a new leaderboard with fair performance evaluation in terms of base LLM model and datasets. We also build an online portal that integrates resources of cleaned datasets, code and paper repository to benefit the community. Finally, we identify limitations in current methodologies and outline future research opportunities.

PDF Details DOI

ICLR Conference 2025 Conference Paper

OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling

Zhicheng Yang
Yiwei Wang 0001
Yinya Huang
Zhijiang Guo
Wei Shi
Xiongwei Han
Liang Feng 0001
Linqi Song

Large language models (LLMs) have exhibited their problem-solving abilities in mathematical reasoning. Solving realistic optimization (OPT) problems in application scenarios requires advanced and applied mathematics ability. However, current OPT benchmarks that merely solve linear programming are far from complex realistic situations. In this work, we propose **OptiBench**, a benchmark for End-to-end optimization problem-solving with human-readable inputs and outputs. **OptiBench** contains rich optimization problems, including linear and nonlinear programming with or without tabular data, which can comprehensively evaluate LLMs' solving ability. In our benchmark, LLMs are required to call a code solver to provide precise numerical answers. Furthermore, to alleviate the data scarcity for optimization problems, and to bridge the gap between open-source LLMs on a small scale (e.g., Llama-3-8b) and closed-source LLMs (e.g., GPT-4), we further propose a data synthesis method namely ***ReSocratic***. Unlike general data synthesis methods that proceed from questions to answers, \ReSocratic first incrementally synthesizes formatted optimization demonstration with mathematical formulations step by step and then back-translates the generated demonstrations into questions. Based on this, we synthesize the ***ReSocratic-29k*** dataset. We further conduct supervised fine-tuning with ***ReSocratic-29k*** on multiple open-source models. Experimental results show that ***ReSocratic-29k*** significantly improves the performance of open-source models.

Details

AAAI Conference 2025 Conference Paper

ParGo: Bridging Vision-Language with Partial and Global Views

An-Lan Wang
Bin Shan
Wei Shi
Kun-Yu Lin
Xiang Fei
Guozhi Tang
Lei Liao
Jingqun Tang

This work presents ParGo, a novel Partial-Global projector designed to connect the vision and language modalities for Multimodal Large Language Models (MLLMs). Unlike previous works that rely on global attention-based projectors, our ParGo bridges the representation gap between the separately pre-trained vision encoders and the LLMs by integrating global and partial views, which alleviates the overemphasis on prominent regions. To facilitate the effective training of ParGo, we collect a large-scale detail-captioned image-text dataset named ParGoCap-1M-PT, consisting of 1 million images paired with high-quality captions. Extensive experiments on several MLLM benchmarks demonstrate the effectiveness of our ParGo, highlighting its superiority in aligning vision and language modalities. Compared to conventional Q-Former projector, our ParGo achieves an improvement of 259.96 in MME benchmark. Furthermore, our experiments reveal that ParGo significantly outperforms other projectors, particularly in tasks that emphasize detail perception ability.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Towards Understanding Transformers in Learning Random Walks

Wei Shi
Yuan Cao

Transformers have proven highly effective across various applications, especially in handling sequential data such as natural languages and time series. However, transformer models often lack clear interpretability, and the success of transformers has not been well understood in theory. In this paper, we study the capability and interpretability of transformers in learning a family of classic statistical models, namely random walks on circles. We theoretically demonstrate that, after training with gradient descent, a one-layer transformer model can achieve optimal accuracy in predicting random walks. Importantly, our analysis reveals that the trained model is interpretable: the trained softmax attention serves as a token selector, focusing on the direct parent state; subsequently, the value matrix executes a one-step probability transition to predict the location of the next state based on this parent state. We also show that certain edge cases not covered by our theory are indeed failure cases, demonstrating that our theoretical conditions are tight. By investigating these success and failure cases, it is revealed that gradient descent with small initialization may fail or struggle to converge to a good solution in certain simple tasks even beyond random walks. Experiments are conducted to support our theoretical findings.

PDF Details

TMLR Journal 2024 Journal Article

From Persona to Personalization: A Survey on Role-Playing Language Agents

Jiangjie Chen
Xintao Wang
Rui Xu
Siyu Yuan
Yikai Zhang
Wei Shi
Jian Xie
Shuang Li

Recent advancements in large language models (LLMs) have significantly boosted the rise of Role-Playing Language Agents (RPLAs), i.e., specialized AI systems designed to simulate assigned personas. By harnessing multiple advanced abilities of LLMs, including in-context learning, instruction following, and social intelligence, RPLAs achieve a remarkable sense of human likeness and vivid role-playing performance. RPLAs can mimic a wide range of personas, ranging from historical figures and fictional characters to real-life individuals. Consequently, they have catalyzed numerous AI applications, such as emotional companions, interactive video games, personalized assistants and copilots, and digital clones. In this paper, we conduct a comprehensive survey of this field, illustrating the evolution and recent progress in RPLAs integrating with cutting-edge LLM technologies. We categorize personas into three types: 1) Demographic Persona, which leverages statistical stereotypes; 2) Character Persona, focused on well-established figures; and 3) Individualized Persona, customized through ongoing user interactions for personalized services. We begin by presenting a comprehensive overview of current methodologies for RPLAs, followed by the details for each persona type, covering corresponding data sourcing, agent construction, and evaluation. Afterward, we discuss the fundamental risks, existing limitations, and prospects of RPLAs. Additionally, we provide a brief review of RPLAs in AI products in the market, which reflects practical user demands that shape and drive RPLA research. Through this survey, we aim to establish a clear taxonomy of RPLA research and applications, facilitate future research in this critical and ever-evolving field, and pave the way for a future where humans and RPLAs coexist in harmony.

PDF Details

IJCAI Conference 2024 Conference Paper

Meta-Learning via PAC-Bayesian with Data-Dependent Prior: Generalization Bounds from Local Entropy

Shiyu Liu
Wei Shi
Zenglin Xu
Shaogao Lv
Yehong Zhang
Hui Wang

Meta-learning accelerates the learning process on unseen learning tasks by acquiring prior knowledge through previous related tasks. The PAC-Bayesian theory provides a theoretical framework to analyze the generalization of meta-learning to unseen tasks. However, previous works still encounter two notable limitations: (1) they merely focus on the data-free priors, which often result in inappropriate regularization and loose generalization bounds; (2) more importantly, their optimization process usually involves nested optimization problems, incurring significant computational costs. To address these issues, we derive new generalization bounds and introduce a novel PAC-Bayesian framework for meta-learning that integrates data-dependent priors. This framework enables the extraction of optimal posteriors for each task in closed form, thereby allowing us to minimize generalization bounds incorporated data-dependent priors with only a simple local entropy. The resulting algorithm, which employs SGLD for sampling from the optimal posteriors, is stable, efficient, and computationally lightweight, eliminating the need for nested optimization. Extensive experimental results demonstrate that our proposed method outperforms the other baselines.

PDF Details DOI

AAAI Conference 2022 Conference Paper

Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer

Tao Wang
Hong Liu
Pinhao Song
Tianyu Guo
Wei Shi

Occluded person re-identification is a challenging task as human body parts could be occluded by some obstacles (e. g. trees, cars, and pedestrians) in certain scenes. Some existing pose-guided methods solve this problem by aligning body parts according to graph matching, but these graph-based methods are not intuitive and complicated. Therefore, we propose a transformer-based Pose-guided Feature Disentangling (PFD) method by utilizing pose information to clearly disentangle semantic components (e. g. human body or joint parts) and selectively match non-occluded parts correspondingly. First, Vision Transformer (ViT) is used to extract the patch features with its strong capability. Second, to preliminarily disentangle the pose information from patch information, the matching and distributing mechanism is leveraged in Pose-guided Feature Aggregation (PFA) module. Third, a set of learnable semantic views are introduced in transformer decoder to implicitly enhance the disentangled body part features. However, those semantic views are not guaranteed to be related to the body without additional supervision. Therefore, Pose-View Matching (PVM) module is proposed to explicitly match visible body parts and automatically separate occlusion features. Fourth, to better prevent the interference of occlusions, we design a Pose-guided Push Loss to emphasize the features of visible body parts. Extensive experiments over five challenging datasets for two tasks (occluded and holistic Re-ID) demonstrate that our proposed PFD is superior promising, which performs favorably against state-of-the-art methods. Code is available at https: //github. com/WangTaoAs/PFD Net

PDF Details

IJCAI Conference 2021 Conference Paper

Adversarial Feature Disentanglement for Long-Term Person Re-identification

Wanlu Xu
Hong Liu
Wei Shi
Ziling Miao
Zhisheng Lu
Feihu Chen

Most existing person re-identification methods are effective in short-term scenarios because of their appearance dependencies. However, these methods may fail in long-term scenarios where people might change their clothes. To this end, we propose an adversarial feature disentanglement network (AFD-Net) which contains intra-class reconstruction and inter-class adversary to disentangle the identity-related and identity-unrelated (clothing) features. For intra-class reconstruction, the person images with the same identity are represented and disentangled into identity and clothing features by two separate encoders, and further reconstructed into original images to reduce intra-class feature variations. For inter-class adversary, the disentangled features across different identities are exchanged and recombined to generate adversarial clothes-changing images for training, which makes the identity and clothing features more independent. Especially, to supervise these new generated clothes-changing images, a re-feeding strategy is designed to re-disentangle and reconstruct these new images for image-level self-supervision in the original image space and feature-level soft-supervision in the disentangled feature space. Moreover, we collect a challenging Market-Clothes dataset and a real-world PKU-Market-Reid dataset for evaluation. The results on one large-scale short-term dataset (Market-1501) and five long-term datasets (three public and two we proposed) confirm the superiority of our method against other state-of-the-art methods.

PDF Details DOI

IJCAI Conference 2021 Conference Paper

Modality-aware Style Adaptation for RGB-Infrared Person Re-Identification

Ziling Miao
Hong Liu
Wei Shi
Wanlu Xu
Hanrong Ye

RGB-infrared (IR) person re-identification is a challenging task due to the large modality gap between RGB and IR images. Many existing methods bridge the modality gap by style conversion, requiring high-similarity images exchanged by complex CNN structures, like GAN. In this paper, we propose a highly compact modality-aware style adaptation (MSA) framework, which aims to explore more potential relations between RGB and IR modalities by introducing new related modalities. Therefore, the attention is shifted from bridging to filling the modality gap with no requirement on high-quality generated images. To this end, we firstly propose a concise feature-free image generation structure to adapt the original modalities to two new styles that are compatible with both inputs by patch-based pixel redistribution. Secondly, we devise two image style quantification metrics to discriminate styles in image space using luminance and contrast. Thirdly, we design two image-level losses based on the quantified results to guide the style adaptation during an end-to-end four-modality collaborative learning process. Experimental results on two datasets SYSU-MM01 and RegDB show that MSA achieves significant improvements with little extra computation cost and outperforms the state-of-the-art methods.

PDF Details DOI