Author name cluster

Chenxi 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.

15 papers

2 author rows

EAAI Journal 2026 Journal Article

A novel variational feature decomposition framework for joint shadow detection and removal in complex visual scenes

Chenxi Wang
Yue Chi
Sheng Xu

Shadow detection and removal are vital for enhancing scene understanding and enabling visual interpretation of high-precision spatiotemporal remote sensing data. Existing methods predominantly rely on low-level features while neglecting the synergistic effects of illumination, light intensity, and scene geometry, often misclassifying warmer/brighter regions as non-shadows and causing false negatives in complex lighting. This study proposes an integrated deep learning framework for end-to-end shadow detection and removal, featuring a variational feature decomposition module to refine illumination features into intensity and color components, and variational inference to distinguish shadow types via photometric variations. Our method demonstrates remarkable performance in both visual perception and quantitative evaluation, outperforming the current state-of-the-art shadow removal methods.

Details DOI

AAAI Conference 2026 Conference Paper

ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models

Zirui Song
Guangxian Ouyang
Mingzhe Li
Yuheng Ji
Chenxi Wang
Zixiang Xu
Zeyu Zhang
Xiaoqing Zhang

Large Vision-Language Models (LVLMs) have recently advanced robotic manipulation by leveraging vision for scene perception and language for instruction following. However, existing methods rely heavily on costly human-annotated training datasets, which limits their generalization and causes them to struggle in out-of-domain (OOD) scenarios, reducing real-world adaptability. To address these challenges, we propose ManipLVM-R1, a novel reinforcement learning framework that replaces traditional supervision with Reinforcement Learning using Verifiable Rewards (RLVR). By directly optimizing for task-aligned outcomes, our method enhances generalization and physical reasoning while removing the dependence on costly annotations. Specifically, we design two rule-based reward functions targeting key robotic manipulation subtasks: an Affordance Perception Reward to enhance localization of interaction regions, and a Trajectory Match Reward to ensure the physical plausibility of action paths. These rewards provide immediate feedback and impose spatial-logical constraints, encouraging the model to go beyond shallow pattern matching and instead learn deeper, more systematic reasoning about physical interactions. Experimental results show that ManipLVM-R1 achieves substantial performance gains across multiple manipulation tasks, using only 50% of the training data while achieving strong generalization to OOD scenarios. We further analyze the benefits of our reward design and its impact on task success and efficiency.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Reflect Then Learn: Active Prompting for Information Extraction Guided by Introspective Confusion

Dong Zhao
Yadong Wang
Xiang Chen
Chenxi Wang
Hongliang Dai
Chuanxing Geng
Shengzhong Zhang
Shao-Yuan Li

Large Language Models (LLMs) show remarkable potential for few-shot information extraction (IE), yet their performance is highly sensitive to the choice of in-context examples. Conventional selection strategies often fail to provide informative guidance, as they overlook a key source of model fallibility: confusion stemming not just from semantic content, but also from the generation of well-structured formats required by IE tasks. To address this, we introduce Active Prompting for Information Extraction (APIE), a novel active prompting framework guided by a principle we term introspective confusion. Our method empowers an LLM to assess its own confusion through a dual-component uncertainty metric that uniquely quantifies both Format Uncertainty (difficulty in generating correct syntax) and Content Uncertainty (inconsistency in extracted semantics). By ranking unlabeled data with this comprehensive score, our framework actively selects the most challenging and informative samples to serve as few-shot exemplars. Extensive experiments on four benchmarks show that our approach consistently outperforms strong baselines, yielding significant improvements in both extraction accuracy and robustness. Our work highlights the critical importance of a fine-grained, dual-level view of model uncertainty when it comes to building effective and reliable structured generation systems.

PDF Details DOI

ICML Conference 2025 Conference Paper

A Non-isotropic Time Series Diffusion Model with Moving Average Transitions

Chenxi Wang
Linxiao Yang
Zhixian Wang
Liang Sun 0001
Yi Wang 0022

Diffusion models, known for their generative ability, have recently been adapted to time series analysis. Most pioneering works rely on the standard isotropic diffusion, treating each time step and the entire frequency spectrum identically. However, it may not be suitable for time series, which often have more informative low-frequency components. We empirically found that direct application of standard diffusion to time series may cause gradient contradiction during training, due to the rapid decrease of low-frequency information in the diffusion process. To this end, we proposed a novel time series diffusion model, MA-TSD, which utilizes the moving average, a natural low-frequency filter, as the forward transition. Its backward process is accelerable like DDIM and can be further considered a time series super-resolution. Our experiments on various datasets demonstrated MA-TSD’s superior performance in time series forecasting and super-resolution tasks.

Details

NeurIPS Conference 2025 Conference Paper

DyFlow: Dynamic Workflow Framework for Agentic Reasoning

Yanbo Wang
Zixiang Xu
Yue Huang
Xiangqi Wang
Zirui Song
Lang Gao
Chenxi Wang
Robert Tang

Agent systems based on large language models (LLMs) have shown great potential in complex reasoning tasks, but building efficient and generalizable workflows remains a major challenge. Most existing approaches rely on manually designed processes, which limits their adaptability across different tasks. While a few methods attempt automated workflow generation, they are often tied to specific datasets or query types and make limited use of intermediate feedback, reducing system robustness and reasoning depth. Moreover, their operations are typically predefined and inflexible. To address these limitations, we propose DyFlow, a dynamic workflow generation framework that adaptively constructs and adjusts reasoning procedures based on task requirements and real-time intermediate feedback, thereby enhancing cross-task generalization. DyFlow consists of two core components: a designer and an executor. The designer decomposes complex problems into a sequence of sub-goals defined by high-level objectives and dynamically plans the next steps based on intermediate outputs and feedback. These plans are then carried out by the executor, which executes each operation using dynamic operators with context-aware parameterization, enabling flexible and semantically grounded reasoning. We systematically evaluate DyFlow across diverse domains, including social reasoning, biomedical tasks, mathematical problem solving, and code generation. Results demonstrate that DyFlow significantly outperforms existing baselines, achieving substantial Pass@k improvements and exhibiting robust generalization across diverse domains.

PDF Details

YNIMG Journal 2025 Journal Article

Individual differences in resting alpha band power and changes in theta band power during sustained pain are correlated with the pain-relieving efficacy of alpha HD-tACS on SM1

Xingang Qi
Tianzhe Jia
Baijintao Sun
Jiahui Xia
Chenxi Wang
Zilong Hong
Yi Zhang
Hanfeng Yang

High-definition transcranial alternating current stimulation (HD-tACS) targeting alpha rhythms (8-13 Hz) shows promise as a pain-relieving intervention, but individual responses vary widely. Understanding the neurobiological mechanism behind this variability is crucial for optimizing HD-tACS parameters to enhance its efficacy in pain relief. In a double-blind, within-subject, sham-controlled experimental study, 34 healthy participants were recruited. We investigated how individual differences in brain oscillations during rest and capsaicin-induced sustained pain states influence the efficacy of alpha HD-tACS. Participants underwent EEG assessments at rest and during capsaicin-induced sustained pain. They then received either sham or active HD-tACS on the sensorimotor cortex (SM1) or dorsolateral prefrontal cortex (DLPFC). We found significant reductions in delta and theta band power at the C4 electrode during sustained pain correlated with individual pain intensity. Additionally, stimulating the SM1 and DLPFC significantly relieved sustained pain. Resting alpha band power and changes in theta band power during sustained pain (the difference in theta band power between sustained pain and rest) at the C4 electrode were both significantly correlated with the pain-relieving efficacy of alpha HD-tACS on SM1. Notably, changes in theta band power mediated the relationship between resting alpha band power and pain-relieving efficacy. These results were not found with alpha HD-tACS on DLPFC. Our results suggest that the variations in theta band power during sustained pain may be crucial for understanding the variability in the efficacy of alpha HD-tACS targeting SM1. The factors influencing the efficacy of alpha HD-tACS on the DLPFC might be multifaceted.

Details DOI

ICLR Conference 2025 Conference Paper

MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation

Chenxi Wang
Xiang Chen 0016
Ningyu Zhang 0001
Bozhong Tian
Haoming Xu
Shumin Deng
Huajun Chen

Multimodal Large Language Models (MLLMs) frequently exhibit hallucination phenomena, but the underlying reasons remain poorly understood. In this paper, we present an empirical analysis and find that, although MLLMs incorrectly generate the objects in the final output, they are actually able to recognize visual objects in the preceding layers. We speculate that this may be due to the strong knowledge priors of the language model suppressing the visual information, leading to hallucinations. Motivated by this, we propose a novel dynamic correction decoding method for MLLMs DeCo, which adaptively selects the appropriate preceding layers and proportionally integrates knowledge into the final layer to adjust the output logits. Note that DeCo is model agnostic and can be seamlessly incorporated with various classic decoding strategies and applied to different MLLMs. We evaluate DeCo on widely-used benchmarks, demonstrating that it can reduce hallucination rates by a large margin compared to baselines, highlighting its potential to mitigate hallucinations. Code is available at https://github.com/zjunlp/DeCo.

Details

NeurIPS Conference 2025 Conference Paper

QiMeng-MuPa: Mutual-Supervised Learning for Sequential-to-Parallel Code Translation

Changxin Ke
Rui Zhang
Shuo Wang
Li Ding
Guangli Li
Yuanbo Wen
Shuoming Zhang
Ruiyuan Xu

The rise of GPU-based high-performance computing (HPC) has driven the widespread adoption of parallel programming models such as CUDA. Yet, the inherent complexity of parallel programming creates a demand for the automated sequential-to-parallel approaches. However, data scarcity poses a significant challenge for machine learning-based sequential-to-parallel code translation. Although recent back-translation methods show promise, they still fail to ensure functional equivalence in the translated code. In this paper, we propose \textbf{QiMeng-MuPa}, a novel \textbf{Mu}tual-Supervised Learning framework for Sequential-to-\textbf{Pa}rallel code translation, to address the functional equivalence issue. QiMeng-MuPa consists of two models, a Translator and a Tester. Through an iterative loop consisting of Co-verify and Co-evolve steps, the Translator and the Tester mutually generate data for each other and improve collectively. The Tester generates unit tests to verify and filter functionally equivalent translated code, thereby evolving the Translator, while the Translator generates translated code as augmented input to evolve the Tester. Experimental results demonstrate that QiMeng-MuPa significantly enhances the performance of the base models: when applied to Qwen2. 5-Coder, it not only improves Pass@1 by up to 28. 91\% and boosts Tester performance by 68. 90\%, but also outperforms the previous state-of-the-art method CodeRosetta by 1. 56 and 6. 92 in BLEU and CodeBLEU scores, while achieving performance comparable to DeepSeek-R1 and GPT-4. 1. Our code is available at \url{https: //github. com/kcxain/mupa}.

PDF Details

NeurIPS Conference 2025 Conference Paper

RSCC: A Large-Scale Remote Sensing Change Caption Dataset for Disaster Events

Zhenyuan Chen
Chenxi Wang
Ningyu Zhang
Feng Zhang

Remote sensing is critical for disaster monitoring, yet existing datasets lack temporal image pairs and detailed textual annotations. While single-snapshot imagery dominates current resources, it fails to capture dynamic disaster impacts over time. To address this gap, we introduce the Remote Sensing Change Caption (RSCC) dataset, a large-scale benchmark comprising 62, 351 pre-/post-disaster image pairs (spanning earthquakes, floods, wildfires, and more) paired with rich, human-like change captions. By bridging the temporal and semantic divide in remote sensing data, RSCC enables robust training and evaluation of vision-language models for disaster-aware bi-temporal understanding. Our results highlight RSCC’s ability to facilitate detailed disaster-related analysis, paving the way for more accurate, interpretable, and scalable vision-language applications in remote sensing. Code and dataset are available at https: //github. com/Bili-Sakura/RSCC.

PDF Details

NeurIPS Conference 2024 Conference Paper

Autonomous Agents for Collaborative Task under Information Asymmetry

Wei Liu
Chenxi Wang
Yifei Wang
Zihao Xie
Rennai Qiu
Yufan Dang
Zhuoyun Du
Weize Chen

Large Language Model Multi-Agent Systems (LLM-MAS) have greatly progressed in solving complex tasks. It communicates among agents within the system to collaboratively solve tasks, under the premise of shared information. However, when agents' collaborations are leveraged to perform multi-person tasks, a new challenge arises due to information asymmetry, since each agent can only access the information of its human user. Previous MAS struggle to complete tasks under this condition. To address this, we propose a new MAS paradigm termed iAgents, which denotes Informative Multi-Agent Systems. In iAgents, the human social network is mirrored in the agent network, where agents proactively exchange human information necessary for task resolution, thereby overcoming information asymmetry. iAgents employs a novel agent reasoning mechanism, InfoNav, to navigate agents' communication towards effective information exchange. Together with InfoNav, iAgents organizes human information in a mixed memory to provide agents with accurate and comprehensive information for exchange. Additionally, we introduce InformativeBench, the first benchmark tailored for evaluating LLM agents' task-solving ability under information asymmetry. Experimental results show that iAgents can collaborate within a social network of 140 individuals and 588 relationships, autonomously communicate over 30 turns, and retrieve information from nearly 70, 000 messages to complete tasks within 3 minutes.

PDF Details DOI

EAAI Journal 2024 Journal Article

Differentiable sampling based efficient architecture search for automatic fault diagnosis

Xingwu Zhang
Rui Ma
Yu Zhao
Chenxi Wang
Zhibin Zhao
Xuefeng Chen

Intelligent diagnosis on rotating machinery has developed rapidly, but different methods have fluctuating performance and fussy design, causing poor effect in practical applications. Thus, it would be great to automatically generate the optimal method for given diagnosis tasks, as differentiable neural architecture search (DNAS) does. However, three challenges severely restrict DNAS methods in industrial scenarios: 1) vibration signals are multi-scale and non-stationary; 2) huge memory cost by supernet-based search is unsuitable to practical diagnosis; 3) manual architecture derivation causes performance collapse between architecture search and practical diagnosis. Thus, we propose Differentiable Sampling based Efficient Architecture Search (DS-EAS), which generates architecture by differentiable sampling. First, the operator involution is introduced to adaptively extract critical features from noisy signals. Second, Gumbel Max-Softmax is adopted to forward sample and backward propagate the gradient on single sub-architecture at one iteration, alleviating huge memory cost. Third, progressively pruning is proposed to eliminate manual discretization error, leading to the final architecture with zero operators. Based on the searched architecture, a deeper one is built to test its real performance. Traction motor experiment is performed to discuss the performance of DS-EAS on three different sample cases. Compared with other state-of-the-art methods, outperformance of DS-EAS is successfully verified.

Details DOI

IJCAI Conference 2024 Conference Paper

FactCHD: Benchmarking Fact-Conflicting Hallucination Detection

Xiang Chen
Duanzheng Song
Honghao Gui
Chenxi Wang
Ningyu Zhang
Yong Jiang
Fei Huang
Chengfei Lyu

Despite their impressive generative capabilities, LLMs are hindered by fact-conflicting hallucinations in real-world applications. The accurate identification of hallucinations in texts generated by LLMs, especially in complex inferential scenarios, is a relatively unexplored area. To address this gap, we present FactCHD, a dedicated benchmark designed for the detection of fact-conflicting hallucinations from LLMs. FactCHD features a diverse dataset that spans various factuality patterns, including vanilla, multi-hop, comparison, and set operation. A distinctive element of FactCHD is its integration of fact-based evidence chains, significantly enhancing the depth of evaluating the detectors' explanations. Experiments on different LLMs expose the shortcomings of current approaches in detecting factual errors accurately. Furthermore, we introduce TRUTH-TRIANGULATOR which synthesizes reflective considerations by tool-enhanced ChatGPT and LoRA-tuning based on Llama2, aiming to yield more credible detection through the amalgamation of predictive results and evidence.

PDF Details DOI

YNIMG Journal 2024 Journal Article

Individual differences of white matter characteristic along the anterior insula-based fiber tract circuit for pain empathy in healthy women and women with primary dysmenorrhea

Junya Mu
Leiming Wu
Chenxi Wang
Wanghuan Dun
Zilong Hong
Xinyue Feng
Ming Zhang
Jixin Liu

Pain empathy, defined as the ability of one person to understand another person's pain, shows large individual variations. The anterior insula is the core region of the pain empathy network. However, the relationship between white matter (WM) properties of the fiber tracts connecting the anterior insula with other cortical regions and an individual's ability to modulate pain empathy remains largely unclear. In this study, we outline an automatic seed-based fiber streamline (sFS) analysis method and multivariate pattern analysis (MVPA) to predict the levels of pain empathy in healthy women and women with primary dysmenorrhoea (PDM). Using the sFS method, the anterior insula-based fiber tract network was divided into five fiber cluster groups. In healthy women, interindividual differences in pain empathy were predicted only by the WM properties of the five fiber cluster groups, suggesting that interindividual differences in pain empathy may rely on the connectivity of the anterior insula-based fiber tract network. In women with PDM, pain empathy could be predicted by a single cluster group. The mean WM properties along the anterior insular-rostroventral area of the inferior parietal lobule further mediated the effect of pain on empathy in patients with PDM. Our results suggest that chronic periodic pain may lead to maladaptive plastic changes, which could further impair empathy by making women with PDM feel more pain when they see other people experiencing pain. Our study also addresses an important gap in the analysis of the microstructural characteristics of seed-based fiber tract network.

Details DOI

IROS Conference 2022 Conference Paper

A Composable Framework for Policy Design, Learning, and Transfer Toward Safe and Efficient Industrial Insertion

Rui Chen 0030
Chenxi Wang
Tianhao Wei
Changliu Liu

Delicate industrial insertion tasks (e. g. , PC board assembly) remain challenging for industrial robots. The chal-lenges include low error tolerance, delicacy of the components, and large task variations with respect to the components to be inserted. To deliver a feasible robotic solution for these insertion tasks, we also need to account for hardware limits of existing robotic systems and minimize the integration effort. This paper proposes a composable framework for efficient integration of a safe insertion policy on existing robotic platforms to accomplish these insertion tasks. The policy has an interpretable modularized design and can be learned efficiently on hardware and transferred to new tasks easily. In particular, the policy includes a safe insertion agent as a baseline policy for insertion, an optimal configurable Cartesian tracker as an interface to robot hardware, a probabilistic inference module to handle component variety and insertion errors, and a safe learning module to optimize the parameters in the aforementioned modules to achieve the best performance on designated hard-ware. The experiment results on a URIO robot show that the proposed framework achieves safety (for the delicacy of components), accuracy (for low tolerance), robustness (against perception error and component defection), adaptability and transferability (for task variations), as well as task efficiency during execution plus data and time efficiency during learning.

Details

IJCAI Conference 2019 Conference Paper

Deep Reinforcement Learning for Ride-sharing Dispatching and Repositioning

Zhiwei (Tony) Qin
Xiaocheng Tang
Yan Jiao
Fan Zhang
Chenxi Wang
Qun (Tracy) Li

In this demo, we will present a simulation-based human-computer interaction of deep reinforcement learning in action on order dispatching and driver repositioning for ride-sharing. Specifically, we will demonstrate through several specially designed domains how we use deep reinforcement learning to train agents (drivers) to have longer optimization horizon and to cooperate to achieve higher objective values collectively.

PDF Details