Author name cluster

Pei Wang

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.

19 papers

2 author rows

AAAI Conference 2026 Short Paper

DINOv3-Powered Multi-Task Foundation Model for Quantitative Remote Sensing Estimation (Student Abstract)

Zhenyu Yu
Mohd Yamani Idna Idris
Pei Wang
Rizwan Qureshi

Quantitative remote sensing estimation is critical for environmental monitoring, providing continuous measures of vegetation indices, canopy height, and carbon stock. Traditional radiative-transfer models and empirical regressions require expert knowledge and generalize poorly, while deep learning methods remain task-specific. We propose SatelliteCalculator+, a DINOv3-powered multi-task foundation model for continuous regression of spectral and structural variables. The framework combines prompt-driven cross-attentive adapters with lightweight MLP decoders, enabling efficient dense prediction from frozen features. To overcome limited supervision, we synthesize over one million paired samples from SPOT 6/7 imagery using physically defined formulas. On the Open-Canopy dataset, SatelliteCalculator+ achieves competitive accuracy across eight ecological variables while reducing inference cost, demonstrating the promise of self-supervised transformers and scalable multi-task learning for large-scale Earth observation.

PDF Details DOI

EAAI Journal 2025 Journal Article

ForgetMe: Benchmarking the selective forgetting capabilities of generative models

Zhenyu Yu
Mohd Yamani Idna Idris
Pei Wang
Yuelong Xia
Yong Xiang

The widespread adoption of diffusion models in image generation has increased the demand for privacy-compliant unlearning. However, due to the high-dimensional nature and complex feature representations of diffusion models, achieving selective unlearning remains challenging, as existing methods struggle to remove sensitive information while preserving the consistency of non-sensitive regions. To address this, we propose an Automatic Dataset Creation Framework based on prompt-based layered editing and training-free local feature removal, constructing the ForgetMe dataset and introducing the Entangled evaluation metric. The Entangled metric quantifies unlearning effectiveness by assessing the similarity and consistency between the target and background regions and supports both paired (Entangled-D) and unpaired (Entangled-S) image data, enabling unsupervised evaluation. The ForgetMe dataset encompasses a diverse set of real and synthetic scenarios, including CUB-200-2011 (Birds), Stanford-Dogs, ImageNet, and a synthetic cat dataset. We apply LoRA fine-tuning on Stable Diffusion to achieve selective unlearning on this dataset and validate the effectiveness of both the ForgetMe dataset and the Entangled metric, establishing them as benchmarks for selective unlearning. Our work provides a scalable and adaptable solution for advancing privacy-preserving generative AI. Code is available at: https: //github. com/YuZhenyuLindy/ForgetMe.

Details DOI

ICLR Conference 2025 Conference Paper

MTU-Bench: A Multi-granularity Tool-Use Benchmark for Large Language Models

Pei Wang
Yanan Wu
Noah Wang
Jiaheng Liu
Xiaoshuai Song
Z. Y. Peng
Ken Deng
Chenchen Zhang

Large Language Models (LLMs) have displayed massive improvements in reason- ing and decision-making skills and can hold natural conversations with users. Recently, many tool-use benchmark datasets have been proposed. However, existing datasets have the following limitations: (1). Insufficient evaluation scenarios (e.g., only cover limited tool-use scenes). (2). Extensive evaluation costs (e.g., GPT API costs). To address these limitations, in this work, we propose a multi-granularity tool-use benchmark for large language models called MTU-Bench. For the "multi-granularity" property, our MTU-Bench covers five tool usage scenes (i.e., single-turn and single-tool, single-turn and multiple-tool, multiple-turn and single-tool, multiple-turn and multiple-tool, and out-of-distribution tasks). Besides, all evaluation metrics of our MTU-Bench are based on the prediction results and the ground truth without using any GPT or human evaluation metrics. Moreover, our MTU-Bench is collected by transforming existing high-quality datasets to simulate real-world tool usage scenarios, and we also propose an instruction dataset called MTU-Instruct data to enhance the tool-use abilities of existing LLMs. Comprehensive experimental results demonstrate the effectiveness of our MTU-Bench.

Details

EAAI Journal 2024 Journal Article

An efficient frequency domain fusion network of infrared and visible images

Chenwu Wang
Junsheng Wu
Aiqing Fang
Zhixiang Zhu
Pei Wang
Hao Chen

Image fusion plays a crucial role in enhancing the quality and accuracy of semantic segmentation, which is essential for autonomous driving systems. By merging information from multiple imaging sensors or modalities, such as infrared and visible images, image fusion enriches the data and improves the perception capabilities of autonomous vehicles. However, current fusion methodologies often cannot balance model complexity, inference efficiency, and fusion accuracy simultaneously, making them difficult to implement in resource-constrained environments. In response to this, this paper presents a lightweight fusion network based on frequency transformation and deep learning techniques, leveraging wavelet transformation to fuse infrared and visible images. Concisely, the fusion model decomposes input images into different frequency sub-bands using wavelet transforms. It then efficiently fuses the multi-scale feature representations in the frequency domains with a specially designed fusion loss. Compared to traditional fusion approaches, our method not only achieves a better balance between subjective fusion quality and downstream vision tasks but also significantly improves model inference efficiency, paving the way for real-time autonomous driving systems. Extensive experiments on public datasets show that our method can achieve state-of-the-art performance while satisfying parameter efficiency in the context of image fusion and semantic segmentation tasks. Concisely, our approach is nearly 100 × faster while using a model 6000 × smaller in size compared to SegMIF.

Details DOI

AAAI Conference 2024 Conference Paper

Discriminatively Fuzzy Multi-View K-means Clustering with Local Structure Preserving

Jun Yin
Shiliang Sun
Lai Wei
Pei Wang

Multi-view K-means clustering successfully generalizes K-means from single-view to multi-view, and obtains excellent clustering performance. In every view, it makes each data point close to the center of the corresponding cluster. However, multi-view K-means only considers the compactness of each cluster, but ignores the separability of different clusters, which is of great importance to producing a good clustering result. In this paper, we propose Discriminatively Fuzzy Multi-view K-means clustering with Local Structure Preserving (DFMKLS). On the basis of minimizing the distance between each data point and the center of the corresponding cluster, DFMKLS separates clusters by maximizing the distance between the centers of pairwise clusters. DFMKLS also relaxes its objective by introducing the idea of fuzzy clustering, which calculates the probability that a data point belongs to each cluster. Considering multi-view K-means mainly focuses on the global information of the data, to efficiently use the local information, we integrate the local structure preserving into the framework of DFMKLS. The effectiveness of DFMKLS is evaluated on benchmark multi-view datasets. It obtains superior performances than state-of-the-art multi-view clustering methods, including multi-view K-means.

PDF Details DOI

JBHI Journal 2024 Journal Article

TSPLASSO: A Two-Stage Prior LASSO Algorithm for Gene Selection Using Omics Data

Sijia Yang
Shunjie Chen
Pei Wang
Aimin Chen
Tianhai Tian

Feature selection has been extensively applied to identify cancer genes using omics data. Although substantial studies have been conducted to search for cancer genes, the available rich knowledge on various cancers is seldom used as prior information in feature selection. This paper proposes a two-stage prior LASSO (TSPLASSO) method, which represents an early attempt in designing feature selection algorithms using prior information. The first stage performs gene selection via linear regression with LASSO. Candidate genes that are correlated with known cancer genes are retained for subsequent analysis. The second stage establishes a logistic regression model with LASSO to realize final cancer gene selection and sample classification. The key advantages of TSPLASSO include the successive consideration of prior cancer genes and binary sample types as response variables in stages one and two, respectively. In addition, the TSPLASSO performs sample classification and variable selection simultaneously. Compared with six state-of-the-art algorithms, numerical simulations in six real-world datasets show that TSPLASSO can improve the accuracy of variable selection by 5%–400% in the three bulk sequencing datasets and the scRNA-seq dataset; and the performance is robust against data noise and variations of prior cancer genes. The TSPLASSO provides an efficient, stable and practical algorithm for exploring biomedcial and health informatics from omics data.

Details DOI

EAAI Journal 2023 Journal Article

Contrastive fusion representation learning for foreground object detection

Pei Wang
Junsheng Wu
Aiqing Fang
Zhixiang Zhu
Chenwu Wang
Pengyuan Mu

This paper investigates the problem of an effective and robust fusion representation method for foreground moving object detection. Many fusion representation learning approaches pay attention to the similarity measurement between the fusion results and source images in texture details and pixel intensity, ignoring the harmful information, e. g. , noise, blur, and extreme illumination. Therefore, the aggregated features of infrared and visible images will introduce much harmful information, affecting the model performance of downstream visual tasks. This paper tackles these problems by proposing a contrastive fusion representation learning method for the foreground moving object detection task, which consists of two major modules: the upstream fusion representation module (FRM) and the downstream foreground moving object detection module (FODM). Unlike the traditional fusion optimization mechanism, the former aims to extract valuable features and reject harmful features via the maximum mutual information theory. The latter is a siamese convolutional neural network to detect foreground moving objects by aggregating the time-sequence images generated by FRM. Experimental results and comparisons with the state-of-the-art on three public datasets (i. e. , TNO, MF, and cross-modal FOD dataset of infrared and visible images), validate the effectiveness, robustness, and overall superiority of the proposed contrast fusion representation learning method. Concisely, our contrastive fusion representation learning has gained 53. 9 %, 43. 2 %, 46. 4 %, 52. 3 %, 2. 2 %, 87. 1 %, 3. 5 % on EI, SF, DF, AG, MI, and Nabf metrics compared with the best competitors.

Details DOI

NeurIPS Conference 2023 Conference Paper

Generalized Belief Transport

Junqi Wang
Pei Wang
Patrick Shafto

Human learners have ability to adopt appropriate learning approaches depending on constraints such as prior on the hypothesis, urgency of decision, and drift of the environment. However, existing learning models are typically considered individually rather than in relation to one and other. To build agents that have the ability to move between different modes of learning over time, it is important to understand how learning models are related as points in a broader space of possibilities. We introduce a mathematical framework, Generalized Belief Transport (GBT), that unifies and generalizes prior models, including Bayesian inference, cooperative communication and classification, as parameterizations of three learning constraints within Unbalanced Optimal Transport (UOT). We visualize the space of learning models encoded by GBT as a cube which includes classic learning models as special points. We derive critical properties of this parameterized space including proving continuity and differentiability which is the basis for model interpolation, and study limiting behavior of the parameters, which allows attaching learning models on the boundaries. Moreover, we investigate the long-run behavior of GBT, explore convergence properties of models in GBT mathematical and computationally, document the ability to learn in the presence of distribution drift, and formulate conjectures about general behavior. We conclude with open questions and implications for more unified models of learning.

PDF Details

EAAI Journal 2023 Journal Article

Operating performance assessment method for industrial process with slowness principle-based LSTM network

Fei Chu
Shuangshuang Liao
Lili Hao
Pei Wang
Yan Liu
Fuli Wang

Timely and accurate performance assessment and non-optimal regulation of industrial processes can effectively guarantee product quality. Most industrial processes are highly nonlinear and dynamic, so long short-term memory (LSTM) network is suitable for industrial performance assessment. However, in the network learning, the typical LSTM network focuses on the representation learning of input variables, lacks the representation of comprehensive economic indexes (CEI), and cannot selectively learn essential features, which increases the computational burden and easily mixes redundant information. Thus, a supervised slow feature analysis (SFA)-based LSTM (SSFALSTM) network is proposed for industrial operating performance assessment. By utilizing CEI information and SFA constraints, the network is guided to simultaneously learn features related to CEI and slow-changing features that reflect the inherent dynamics of the process. Further, cascade performance recognition model to construct the complete performance assessment framework. For the non-optimal performance, a reconstruction-based contribution plot method is proposed to identify the main cause variables and guide adjustment operations. Finally, the effectiveness of the proposed method is validated on the dense medium coal preparation process.

Details DOI

ICML Conference 2022 Conference Paper

Discrete Probabilistic Inverse Optimal Transport

Wei-Ting Chiu
Pei Wang
Patrick Shafto

Inverse Optimal Transport (IOT) studies the problem of inferring the underlying cost that gives rise to an observation on coupling two probability measures. Couplings appear as the outcome of matching sets (e. g. dating) and moving distributions (e. g. transportation). Compared to Optimal transport (OT), the mathematical theory of IOT is undeveloped. We formalize and systematically analyze the properties of IOT using tools from the study of entropy-regularized OT. Theoretical contributions include characterization of the manifold of cross-ratio equivalent costs, the implications of model priors, and derivation of an MCMC sampler. Empirical contributions include visualizations of cross-ratio equivalent effect on basic examples, simulations validating theoretical results and experiments on real world data.

Details

NeurIPS Conference 2020 Conference Paper

A mathematical theory of cooperative communication

Pei Wang
Junqi Wang
Pushpi Paranamana
Patrick Shafto

Cooperative communication plays a central role in theories of human cognition, language, development, culture, and human-robot interaction. Prior models of cooperative communication are algorithmic in nature and do not shed light on why cooperation may yield effective belief transmission and what limitations may arise due to differences between beliefs of agents. Through a connection to the theory of optimal transport, we establishing a mathematical framework for cooperative communication. We derive prior models as special cases, statistical interpretations of belief transfer plans, and proofs of robustness and instability. Computational simulations support and elaborate our theoretical results, and demonstrate fit to human behavior. The results show that cooperative communication provably enables effective, robust belief transmission which is required to explain feats of human learning and improve human-machine interaction.

PDF Details

AAAI Conference 2020 Conference Paper

Dynamic Reward-Based Dueling Deep Dyna-Q: Robust Policy Learning in Noisy Environments

Yangyang Zhao
Zhenyu Wang
Kai Yin
Rui Zhang
Zhenhua Huang
Pei Wang

Task-oriented dialogue systems provide a convenient interface to help users complete tasks. An important consideration for task-oriented dialogue systems is the ability to against the noise commonly existed in the real-world conversation. Both rule-based strategies and statistical modeling techniques can solve noise problems, but they are costly. In this paper, we propose a new approach, called Dynamic Reward-based Dueling Deep Dyna-Q (DR-D3Q). The DR-D3Q can learn policies in noise robustly, and it is easy to implement by combining dynamic reward and the Dueling Deep Q-Network (Dueling DQN) into Deep Dyna-Q (DDQ) framework. The Dueling DQN can mitigate the negative impact of noise on learning policies, but it is inapplicable to dialogue domain due to different reward mechanisms. Unlike typical dialogue reward function, we integrate dynamic reward that provides reward in real-time for agent to make Dueling DQN adapt to dialogue domain. For the purpose of supplementing the limited amount of real user experiences, we take the DDQ framework as the basic framework. Experiments using simulation and human evaluation show that the DR-D3Q signiﬁcantly improve the performance of policy learning tasks in noisy environments. 1

PDF Details

ICML Conference 2020 Conference Paper

Sequential Cooperative Bayesian Inference

Junqi Wang 0002
Pei Wang
Patrick Shafto

Cooperation is often implicitly assumed when learning from other agents. Cooperation implies that the agent selecting the data, and the agent learning from the data, have the same goal, that the learner infer the intended hypothesis. Recent models in human and machine learning have demonstrated the possibility of cooperation. We seek foundational theoretical results for cooperative inference by Bayesian agents through sequential data. We develop novel approaches analyzing consistency, rate of convergence and stability of Sequential Cooperative Bayesian Inference (SCBI). Our analysis of the effectiveness, sample efficiency and robustness show that cooperation is not only possible but theoretically well-founded. We discuss implications for human-human and human-machine cooperation.

Details

NeurIPS Conference 2019 Conference Paper

Deliberative Explanations: visualizing network insecurities

Pei Wang
Nuno Nvasconcelos

A new approach to explainable AI, denoted {\it deliberative explanations, \/} is proposed. Deliberative explanations are a visualization technique that aims to go beyond the simple visualization of the image regions (or, more generally, input variables) responsible for a network prediction. Instead, they aim to expose the deliberations carried by the network to arrive at that prediction, by uncovering the insecurities of the network about the latter. The explanation consists of a list of insecurities, each composed of 1) an image region (more generally, a set of input variables), and 2) an ambiguity formed by the pair of classes responsible for the network uncertainty about the region. Since insecurity detection requires quantifying the difficulty of network predictions, deliberative explanations combine ideas from the literatures on visual explanations and assessment of classification difficulty. More specifically, the proposed implementation combines attributions with respect to both class predictions and a difficulty score. An evaluation protocol that leverages object recognition (CUB200) and scene classification (ADE20K) datasets that combine part and attribute annotations is also introduced to evaluate the accuracy of deliberative explanations. Finally, an experimental evaluation shows that the most accurate explanations are achieved by combining non self-referential difficulty scores and second-order attributions. The resulting insecurities are shown to correlate with regions of attributes that are shared by different classes. Since these regions are also ambiguous for humans, deliberative explanations are intuitive, suggesting that the deliberative process of modern networks correlates with human reasoning.

PDF Details

EAAI Journal 2015 Journal Article

A framework with reasoning capabilities for crisis response decision–support systems

Nady Slam
Wenjun Wang
Guixiang Xue
Pei Wang

This paper reviews the methods in decision–support systems for crisis management. While much research has been conducted in this field, little emphasis has been placed on the uncertainty representation, reasoning, learning and real time decision-making capabilities of system. The purpose of this paper is to explore the basic assumptions of constructing an intelligent decision–support system for crisis response management. A novel framework for crisis response decision-making system under the assumption of openness to various kinds of uncertainties, reasoning and learning with real-time response is proposed. We applied the Non-Axiomatic Logic in representing and reasoning the uncertainty knowledge in the framework and demonstrated the reasoning and learning mechanisms of the framework through an application in a case study in the field of urban firefighting. The results show that the framework provides a suitable model for intelligent crisis response decision support systems.

Details DOI

AIJ Journal 2004 Journal Article

The limitation of Bayesianism

Pei Wang

In the current discussion about the capacity of Bayesianism in reasoning under uncertainty, there is a conceptual and notational confusion between the explicit condition and the implicit condition of a probability evaluation. Consequently, the limitation of Bayesianism is often seriously underestimated. To represent the uncertainty of a belief system where revision is needed, it is not enough to assign a probability value to each belief.

Details DOI

AAAI Conference 2000 Conference Paper

Non-Axiomatic Reasoning System (Version 4.1)

Pei Wang

NARS (Non-Axiomatic Reasoning System) is an intelligent reasoning system. It can answer questions according to the knowledge originally provided by its user. What makes it different from conventional reasoning systems is its ability to learn from its experience and to work with insufficient knowledge and resources. The NARS 4.1 demo is a Java applet. It comes with help information and simple examples to show how the system does deduction, induction, abduction, analogy, belief revision, membership evaluation, relational inference, backward inference, new concept formation, and so on, in a unified manner. The demo also allows the user to create new examples to test the system, as well as to see the internal structure and process when the system is running. The on-line help information contains links to relevant publications.

PDF Details

UAI Conference 1994 Conference Paper

A Defect in Dempster-Shafer Theory

Pei Wang

By analyzing the relationships among chance, weight of evidence and degree of beliefwe show that the assertion "probability functions are special cases of belief functions" and the assertion "Dempster's rule can be used to combine belief functions based on distinct bodies of evidence" together lead to an inconsistency in Dempster-Shafer theory. To solve this problem, we must reject some fundamental postulates of the theory. We introduce a new approach for uncertainty management that shares many intuitive ideas with D-S theory, while avoiding this problem.

Details

UAI Conference 1993 Conference Paper

Belief Revision in Probability Theory

Pei Wang

In a probability-based reasoning system, Bayes' theorem and its variations are often used to revise the system's beliefs. However, if the explicit conditions and the implicit conditions of probability assignments `me properly distinguished, it follows that Bayes' theorem is not a generally applicable revision rule. Upon properly distinguishing belief revision from belief updating, we see that Jeffrey's rule and its variations are not revision rules, either. Without these distinctions, the limitation of the Bayesian approach is often ignored or underestimated. Revision, in its general form, cannot be done in the Bayesian approach, because a probability distribution function alone does not contain the information needed by the operation.

Details