in Association for Computational Linguistics, 2026
To accurately model the intricate nature of length bias and facilitate more effective bias mitigation, it proposes FiMi-RM (Bias Fitting to Mitigate Length Bias of Reward Model in RLHF), a framework that autonomously learns and corrects underlying bias patterns.
Download here
in International Conference on Learning Representations, 2026
It introduces a Response-conditioned Bradley-Terry (Rc-BT) model that enhances the model’s capability in length bias mitigating and length instruction following, through training on the augmented dataset. Furthermore, it proposes the Rc-RM and Rc-DPO algorithm to leverage the Rc-BT model for reward modeling and direct policy optimization (DPO) of LLMs.
Download here
in Association for Computational Linguistics, 2026
The rise of reasoning models necessitates large-scale verifiable data, for which programming tasks serve as an ideal source. To address this, we propose a Feedback-Driven Iterative Framework for comprehensive test case construction and release CodeContests-O.
Download here
under review in Neural Information Processing Systems, 2025
The research identifies a critical oversight in existing techniques, which predominantly focus on comparing responses while neglecting valuable latent signals embedded within prompt inputs, and which only focus on preference disparities at the intra-sample level, while neglecting to account for the inter-sample level preference differentials that exist among preference data. To leverage these previously neglected indicators, it proposes a novel Multi-level Aware Preference Learning (MAPL) framework, capable of enhancing multi-instruction capabilities.
Download here
in Neural Information Processing Systems, 2025
It is the first to systematically investigate the effectiveness and underlying mechanisms of activation engineering for mitigating hallucinations in VideoLLMs. And it proposes a temporal-aware activation engineering framework for VideoLLMs, which adaptively identifies and manipulates hallucination-sensitive modules based on the temporal variation characteristic, substantially mitigating hallucinations without additional LLM fine-tuning.
Download here
in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023
This paper proposes a Heterogeneous Network based on Contrastive Learning (HCLNet). HCLNet aims to learn high-level representation from unlabeled PolSAR data for few-shot classification according to multi-features.
Download here
in Arxiv, 2023
This paper proposes the Flowmind2digital method and hdFlowmind dataset to address the convertion of hand-drawn flowchart/mindmap.
Download here