10 papers
Act More, Decide Less: Skill-Guided Adaptive Action Chunking for Long-Horizon LLM Agents
Yanting Yang, Can Jin, Jinman Zhao +6
Large language model (LLM) agents for long-horizon interactive tasks typically follow a ReAct-style protocol, issuing one primitive action per LLM round. While this enables frequen…
ES-VP : Energy-Shaped Dynamic Visual Prompting for Efficient Model Adaptation
Can Jin, Ying Li, Jingchen Sun +5
Visual prompting (VP) has emerged as a parameter-efficient method for adapting pre-trained models to downstream tasks. However, existing approaches encounter a trade-off between fl…
Evidence Over Plans: Online Trajectory Verification for Skill Distillation
Yang Zhou, Zihan Dong, Zhenting Wang +7
Agent skills can remarkably improve task success rates by using human-written procedural documents, but their quality is difficult to assess without environment-grounded verificati…
Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs
Jiakang Li, Guanyu Zhu, Can Jin +8
Strong reasoning depends not only on model knowledge but also on how effectively cognitive behaviors are deployed during generation. Existing methods often rely on explicit behavio…
AgentGR: Semantic-aware Agentic Group Decision-Making Simulator for Group Recommendation
Yangtao Zhou, Wenhao You, Hua Chu +4
Group Recommendation (GR) aims to suggest items to a group of users, which has become a critical component of modern social platforms. Existing GR methods focus on aggregating indi…
DARE: Difficulty-Adaptive Reinforcement Learning with Co-Evolved Difficulty Estimation
Yang Zhou, Can Jin, Zihan Dong +7
Reinforcement learning improves the reasoning ability of large language models but remains costly and sample-inefficient, as many rollouts provide weak learning signals. Difficulty…
Unleashing Scalable Context Parallelism for Foundation Models Pre-Training via FCP
Yilong Zhao, Xiaonan Nie, Kan Zhu +6
Context parallelism (CP) has been widely adopted to support the growing context length in foundation model pretraining. However, existing designs fail to handle the large variation…
Jackpot: Optimal Budgeted Rejection Sampling for Extreme Actor-Policy Mismatch Reinforcement Learning
Zhuoming Chen, Hongyi Liu, Yang Zhou +2
Reinforcement learning (RL) for large language models (LLMs) remains expensive, particularly because the rollout is expensive. Decoupling rollout generation from policy optimizatio…
Kinetics: Rethinking Test-Time Scaling Laws
Ranajoy Sadhukhan, Zhuoming Chen, Haizhong Zheng +3
We rethink test-time scaling laws from a practical efficiency perspective, revealing that the effectiveness of smaller models is significantly overestimated. Prior work, grounded i…
Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts
Haizhong Zheng, Yang Zhou, Brian R. Bartoldson +4
Reinforcement learning, such as PPO and GRPO, has powered recent breakthroughs in LLM reasoning. Scaling rollout to sample more prompts enables models to selectively use higher-qua…