13 papers
Hierarchical Credit Assignment for RLVR on Fused Gromov-Wasserstein Geometry
Qi Yu, Ruizhong Qiu, Zhichen Zeng +6
Reinforcement learning with verifiable rewards (RLVR) has been shown to improve the reasoning capability of large language models (LLMs) across diverse reasoning tasks. However, gr…
PolicyMem: Geometric Policy Memory for LLM Governance
Yuanchen Bei, Zhengzhang Chen, Yanjun Zhao +3
As large language models (LLMs) are increasingly deployed in real-world high-stakes applications, effective governance has become essential. Existing safeguards largely follow two…
Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agents
Tianxin Wei, Zhan Shi, Minhua Lin +14
Learning from experience is critical for developing capable, self-improving large language model (LLM) agents. Existing methods typically extract knowledge from accumulated traject…
From Inference to Adaptation: A Unified Optimal Transport View of Vision Language Model
Qi Yu, Zhichen Zeng, Katherine Tieu +8
Vision-language models (VLMs) have demonstrated remarkable zero-shot capabilities yet remain sensitive to real-world distribution shifts during inference. Although significant effo…
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
Yanjun Zhao, Ruizhong Qiu, Tianxin Wei +6
Understanding and reasoning over long contexts has become a key requirement for deploying large language models (LLMs) in realistic applications. Although recent LLMs support incre…
TAG-DLM: Diffusion Language Models for Text-Attributed Graph Learning
Lingjie Chen, Yuanchen Bei, Haobo Xu +3
Text-attributed graphs (TAGs), where each node carries a natural language description, require models to jointly reason over text and graph topology. Existing approaches often hand…
Harnessing Generalist Agents for Contextualized Time Series
Zihao Li, Kaifeng Jin, Yuanchen Bei +8
Time series are often embedded in rich contexts that are essential for holistic modeling. Moreover, real-world practitioners often require end-to-end workflows for analyzing tempor…
Code as Agent Harness
Xuying Ning, Katherine Tieu, Dongqi Fu +39
Recent large language models (LLMs) have demonstrated strong capabilities in understanding and generating code, from competitive programming to repository-level software engineerin…
Ramen: Robust Test-Time Adaptation of Vision-Language Models with Active Sample Selection
Wenxuan Bao, Yanjun Zhao, Xiyuan Yang +1
Pretrained vision-language models such as CLIP exhibit strong zero-shot generalization but remain sensitive to distribution shifts. Test-time adaptation adapts models during infere…
PaperMind: Benchmarking Agentic Reasoning and Critique over Scientific Papers in Multimodal LLMs
Yanjun Zhao, Tianxin Wei, Jiaru Zou +7
Understanding scientific papers requires more than answering isolated questions or summarizing content. It involves an integrated reasoning process that grounds textual and visual…
TRIMS: Trajectory-Ranked Instruction Masked Supervision for Diffusion Language Models
Lingjie Chen, Ruizhong Qiu, Yuyu Fan +2
Diffusion language models (DLMs) offer a promising path toward low-latency generation through parallel decoding, but their practical efficiency depends heavily on the decoding traj…
SABER: Switchable and Balanced Training for Efficient LLM Reasoning
Kai Zhao, Yanjun Zhao, Jiaming Song +4
Large language models (LLMs) empowered by chain-of-thought reasoning have achieved impressive accuracy on complex tasks but suffer from excessive inference costs and latency when a…
RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation
Tianjiao Li, Mengran Yu, Chenyu Shi +6
Large language models (LLMs) possess strong multilingual capabilities, and combining Reinforcement Learning from Human Feedback (RLHF) with translation tasks has shown great potent…