Author name cluster

Jiawen Li

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.

7 papers

1 author row

AAAI Conference 2026 Conference Paper

CLIP2Pose: Frozen CLIP as Semantic Guide for Domain Adaptive Pose Estimation

Jiawen Li
Fei Jiang
Dandan Zhu
Jinxin Shi
Aimin Zhou

Unsupervised domain adaptive pose estimation is a fundamental yet challenging task due to the need to transfer from labeled synthetic data to unlabeled real data. Nevertheless, the underlying pose semantics, which are governed by spatial structure, remain largely consistent across domains. This observation motivates the use of vision-language models, which provide domain-invariant representations that align well with high-level semantic concepts. Motivated by this, we propose CLIP2Pose, a novel framework that leverages the semantic robustness of frozen CLIP encoders to facilitate cross-domain generalization. We first introduce a semantic-driven prompt mechanism that encodes structural priors, domain-specific appearance, and instance-level context into the image representation. This guides the model to focus on semantically meaningful and structurally relevant features. Next, we propose a semantic modulation module that adaptively refines visual features by conditioning them on prompt-derived embeddings, enhancing alignment between semantics and visual patterns. To further bridge the modality and domain gaps, we design a directional alignment loss that encourages consistent structural reasoning across both vision and language representations. Extensive experiments on domain adaptive human body and hand pose benchmarks show that CLIP2Pose achieves state-of-the-art performance.

PDF Details DOI

JBHI Journal 2026 Journal Article

DiagR1: A Vision-Language Model Trained via Reinforcement Learning for Digestive Pathology Diagnosis

Minxi Ouyang
Lianghui Zhu
Yaqing Bao
Qiang Huang
Jingli Ouyang
Tian Guan
Xitong Ling
Jiawen Li

Multimodal large models have shown great potential in automating pathology image analysis. However, current multimodal models for gastrointestinal pathology are constrained by both data quality and reasoning transparency: pervasive noise and incomplete annotations in public datasets predispose vision-language models to factual hallucinations when generating diagnostic text, while the absence of explicit intermediate reasoning chains renders the outputs difficult to audit and thus less trustworthy in clinical practice. To address these issues, we construct a large-scale gastrointestinal pathology dataset containing both microscopic descriptions and diagnostic conclusions, and propose a prompt augmentation strategy that incorporates lesion classification and anatomical site information. This design guides the model to better capture image-specific features and maintain semantic consistency in generation. Furthermore, we employ a post-training pipeline that combines supervised fine-tuning with Group Relative Policy Optimization (GRPO) to improve reasoning quality and output structure. Experimental results on real-world pathology report generation tasks demonstrate that our approach significantly outperforms state-of-the-art open-source and proprietary baselines in terms of generation quality, structural completeness, and clinical relevance. Our solution outperforms state-of-the-art models with 18. 7% higher clinical relevance, 32. 4% improved structural completeness, and 41. 2% fewer diagnostic errors, demonstrating superior accuracy and clinical utility compared to existing solutions.

Details DOI

NeurIPS Conference 2025 Conference Paper

ACT as Human: Multimodal Large Language Model Data Annotation with Critical Thinking

Lequan Lin
Dai Shi
Andi Han
Feng Chen
Qiuzheng Chen
Jiawen Li
Zhaoyang Li
Jiyuan Zhang

Supervised learning relies on high-quality labeled data, but obtaining such data through human annotation is both expensive and time-consuming. Recent work explores using large language models (LLMs) for annotation, but LLM-generated labels still fall short of human-level quality. To address this problem, we propose the Annotation with Critical Thinking (ACT) data pipeline, where LLMs serve not only as annotators but also as judges to critically identify potential errors. Human effort is then directed towards reviewing only the most "suspicious" cases, significantly improving the human annotation efficiency. Our major contributions are as follows: (1) ACT is applicable to a wide range of domains, including natural language processing (NLP), computer vision (CV), and multimodal understanding, by leveraging multimodal-LLMs (MLLMs). (2) Through empirical studies, we derive 7 insights on how to enhance annotation quality while efficiently reducing the human cost, and then translate these findings into user-friendly guidelines. (3) We theoretically analyze how to modify the loss function so that models trained on ACT data achieve similar performance to those trained on fully human-annotated data. Our experiments show that the performance gap can be reduced to less than 2% on most benchmark datasets while saving up to 90% of human costs.

PDF Details

EAAI Journal 2025 Journal Article

Hierarchical cascaded networks with multi-task balanced loss for fine-grained hashing

Shun Liu
Yanjun Zheng
Xianxian Zeng
Jun Yuan
Jiawen Li
Rongjun Chen

Fine-grained image retrieval has seen significant advancements with the rise of deep hashing methods. However, these methods often prioritize high-level features, which may lead to the loss of important low-level details in hash code representations. Additionally, balancing the classification and hashing tasks remains a challenge. To address these issues, we propose a Hierarchical Cascaded Network (HCN) with a multi-task balanced loss function tailored for fine-grained hashing. Our model captures detailed information from different feature levels through a hierarchical backbone network and utilizes a cascaded representation learning module to enhance and fuse features using attention mechanisms. An adaptive loss function ensures a balanced contribution from both classification and hashing tasks during training. Extensive experiments on benchmark datasets demonstrate that HCN outperforms state-of-the-art methods, achieving a promising improvement across five datasets and multiple hash code lengths. These results highlight the effectiveness of HCN in enhancing fine-grained image retrieval, with potential applications in areas requiring both high accuracy and efficient retrieval.

Details DOI

NeurIPS Conference 2024 Conference Paper

Unveiling the Tapestry of Consistency in Large Vision-Language Models

Yuan Zhang
Fei Xiao
Tao Huang
Chun-Kai Fan
Hongyuan Dong
Jiawen Li
Jiacong Wang
Kuan Cheng

Large vision-language models (LVLMs) have recently achieved rapid progress, exhibiting great perception and reasoning abilities concerning visual information. However, when faced with prompts in different sizes of solution spaces, LVLMs fail to always give consistent answers regarding the same knowledge point. This inconsistency of answers between different solution spaces is prevalent in LVLMs and erodes trust. To this end, we provide a multi-modal benchmark ConBench, to intuitively analyze how LVLMs perform when the solution space of a prompt revolves around a knowledge point. Based on the ConBench tool, we are the first to reveal the tapestry and get the following findings: (1) In the discriminate realm, the larger the solution space of the prompt, the lower the accuracy of the answers. (2) Establish the relationship between the discriminative and generative realms: the accuracy of the discriminative question type exhibits a strong positive correlation with its Consistency with the caption. (3) Compared to open-source models, closed-source models exhibit a pronounced bias advantage in terms of Consistency. Eventually, we ameliorate the consistency of LVLMs by trigger-based diagnostic refinement, indirectly improving the performance of their caption. We hope this paper will accelerate the research community in better evaluating their models and encourage future advancements in the consistency domain.

PDF Details DOI

EAAI Journal 2023 Journal Article

Distributed deep reinforcement learning-based gas supply system coordination management method for solid oxide fuel cell

Jiawen Li
Haoyang Cui
Wei Jiang

In order to sustain solid oxide fuel cell (SOFC) net output power and prevent violation of oxygen excess ratio (OER) constraint and fuel utilization (FU) constraint, a data-driven gas supply system coordination management method is proposed. Accordingly, a population evolution-based multi-agent double delay deep deterministic policy gradient (PE-MA4DPG) algorithm is introduced. The artificial intelligence design of the algorithm is guided by the concepts of imitation learning and curriculum learning, whereby different agents of different combinations are trained in different environments, thus improving the robustness of the coordination strategy. In this algorithm, the hydrogen controller and the air controller are treated as two agents. The centralized training enables agents with different objectives to coordinate with each other. The effectiveness of the proposed algorithm is demonstrated in three experiments, wherein the proposed algorithm is compared with a group of existing algorithms.

Details DOI

EAAI Journal 2021 Journal Article

Emergency fault affected wide-area automatic generation control via large-scale deep reinforcement learning

Jiawen Li
Tao Yu
Xiaoshun Zhang

When a complex large power system is in an emergency, the conventional discrete emergency control strategy implemented will cause excess load or derivative accidents like line power overload, thereby raising the operation risk. To overcome the problems of excessive/insufficient regulation and subsequent accidents in the above context, this paper proposes a wide area automatic generation control (WA-AGC) framework, which integrates the emergency control strategy and a performance-based frequency regulation market mechanism. According to the frequency status of the power system, WA-AGC divides the AGC into four intervals, emergency AGC (EAGC), conventional AGC (CAGC), AGC transition and optimal power flow (OPF). These four together realize a comprehensive optimization of frequency and system stability as well as economy Based on the above framework, a swarm agent exploration distributed multiple delayed deep policy gradient algorithm (SAE-MD3) is developed, which uses multiple explorers with different exploration strategies for distributed optimization. In addition, several technologies are introduced to prevent Q value overestimation and generate a more robust optimal AGC strategy. Afterward, the effectiveness and feasibility of WA-AGC are verified through the simulations of an IEEE-9 two-area system and an IEEE-118 two-area system. Compared to conventional AGC strategies, the WA-AGC algorithm reduces the constraint violation time of the power line by 92. 06% and the power generation cost by 0. 27% as well as improves the CPS1 index by 0. 04%.

Details DOI