11 papers
Tmax: A simple recipe for terminal agents
Hamish Ivison, Junjie Oscar Yin, Rulin Shao +3
Terminal-using agents have quickly become the most popular downstream application of language models (LMs). Despite their prevalence, relatively little academic work has examined R…
EvoLM: Self-Evolving Language Models through Co-Evolved Discriminative Rubrics
Shuyue Stella Li, Rui Xin, Teng Xiao +8
Language models encode substantial evaluative knowledge from pretraining, yet current post-training methods rely on external supervision (human annotations, proprietary models, or…
Small Reward Models via Backward Inference
Yike Wang, Faeze Brahman, Shangbin Feng +3
Reward models (RMs) play a central role throughout the language model (LM) pipeline, particularly in non-verifiable domains. However, the dominant LLM-as-a-Judge paradigm relies on…
W2S-AlignTree: Weak-to-Strong Inference-Time Alignment for Large Language Models via Monte Carlo Tree Search
Zhenyu Ding, Yuhao Wang, Tengyue Xiao +3
Large Language Models (LLMs) demonstrate impressive capabilities, yet their outputs often suffer from misalignment with human preferences due to the inadequacy of weak supervision…
Scaling and Transferability of Annealing Strategies in Large Language Model Training
Siqi Wang, Zhengyu Chen, Teng Xiao +5
Learning rate scheduling is crucial for training large language models, yet understanding the optimal annealing strategies across different model configurations remains challenging…
Why Do LLM Agents Fail in Exploring New Environments? A World-Modeling Perspective
Shiqi Chen, Tongyao Zhu, Zian Wang +8
Large Language Models (LLMs) as agents often fail to improve in new environments. We identify and characterize a failure mode we call exploration collapse: under reinforcement lear…
Reinforcement Learning for Tool-Integrated Interleaved Thinking towards Cross-Domain Generalization
Zhengyu Chen, Jinluan Yang, Teng Xiao +6
Recent advances in large language models (LLMs) have demonstrated remarkable capabilities in reasoning and tool utilization. However, the generalization of tool-augmented reinforce…
Sub-Scaling Laws: On the Role of Data Density and Training Strategies in LLMs
Zhengyu Chen, Siqi Wang, Teng Xiao +5
Traditional scaling laws in natural language processing suggest that increasing model size and training data enhances performance. However, recent studies reveal deviations, partic…
From Mathematical Reasoning to Code: Generalization of Process Reward Models in Test-Time Scaling
Zhengyu Chen, Yudong Wang, Teng Xiao +5
Recent advancements in improving the reasoning capabilities of Large Language Models have underscored the efficacy of Process Reward Models (PRMs) in addressing intermediate errors…
InfoPO: On Mutual Information Maximization for Large Language Model Alignment
Teng Xiao, Zhen Ge, Sujay Sanghavi +5
We study the post-training of large language models (LLMs) with human preference data. Recently, direct preference optimization and its variants have shown considerable promise in…
In-Context Sharpness as Alerts: An Inner Representation Perspective for Hallucination Mitigation
Shiqi Chen, Miao Xiong, Junteng Liu +4
Large language models (LLMs) frequently hallucinate and produce factual errors, yet our understanding of why they make these errors remains limited. In this study, we delve into th…