11 papers
Mechanistic Insights into Functional Sparsity in Multimodal LLMs via CoRe Heads
Ruoxi Sun, Quantong Qiu, Juntao Li +3
While Multimodal Large Language Models (MLLMs) demonstrate remarkable proficiency on complex vision-language tasks, the mechanisms by which they extract query-relevant visual featu…
LongRM: Revealing and Unlocking the Context Boundary of Reward Modeling
Zecheng Tang, Baibei Ji, Quantong Qiu +4
Reward model (RM) plays a pivotal role in aligning large language model (LLM) with human preferences. As real-world applications increasingly involve long history trajectories, e.g…
Revisiting Long-context Modeling from Context Denoising Perspective
Zecheng Tang, Baibei Ji, Juntao Li +3
Long-context models (LCMs) have demonstrated great potential in processing long sequences, facilitating many real-world applications. The success of LCMs can be attributed to their…
SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning
Yuyang Ding, Xinyu Shi, Juntao Li +3
Process reward models (PRMs) offer fine-grained, step-level evaluations that facilitate deeper reasoning processes in large language models (LLMs), proving effective in complex tas…
Improving Rationality in the Reasoning Process of Language Models through Self-playing Game
Pinzheng Wang, Juntao Li, Zecheng Tang +2
Large language models (LLMs) have demonstrated considerable reasoning abilities in various tasks such as mathematics and coding. However, recent studies indicate that even the best…
Unlocking Recursive Thinking of LLMs: Alignment via Refinement
Haoke Zhang, Xiaobo Liang, Cunxiang Wang +2
The OpenAI o1-series models have demonstrated that leveraging long-form Chain of Thought (CoT) can substantially enhance performance. However, the recursive thinking capabilities o…
Revealing and Mitigating Over-Attention in Knowledge Editing
Pinzheng Wang, Zecheng Tang, Keyan Zhou +3
Large Language Models have demonstrated superior performance across a wide range of tasks, but they still exhibit undesirable errors due to incorrect knowledge learned from the tra…
LOGO -- Long cOntext aliGnment via efficient preference Optimization
Zecheng Tang, Zechen Sun, Juntao Li +2
Long-context models(LCMs) have shown great potential in processing long input sequences(even more than 100M tokens) conveniently and effectively. With significant progress, recent…
Unleashing LLM Reasoning Capability via Scalable Question Synthesis from Scratch
Yuyang Ding, Xinyu Shi, Xiaobo Liang +4
Improving the mathematical reasoning capabilities of Large Language Models (LLMs) is critical for advancing artificial intelligence. However, access to extensive, diverse, and high…
Fennec: Fine-grained Language Model Evaluation and Correction Extended through Branching and Bridging
Xiaobo Liang, Haoke Zhang, Helan hu +3
The rapid advancement of large language models has given rise to a plethora of applications across a myriad of real-world tasks, mainly centered on aligning with human intent. Howe…
Rethinking Negative Instances for Generative Named Entity Recognition
Yuyang Ding, Juntao Li, Pinzheng Wang +3
Large Language Models (LLMs) have demonstrated impressive capabilities for generalizing in unseen tasks. In the Named Entity Recognition (NER) task, recent advancements have seen t…