11 papers
Mitigating the Length-Scaling Tax with Online Distillation
Xu Wan, Wenyue Xu, Shengjie Zhao +1
Length scaling during reinforcement-learning (RL) post-training is often viewed as a sign of improved reasoning ability, especially on difficult problems, but may also make respons…
Reformulate LLM Reinforcement Learning for Efficient Training under Black-box Discrepancy
Jiashun Liu, Runze Liu, Xu Wan +3
Reinforcement Learning (RL) has emerged as a pivotal post-training paradigm, yet it frequently suffers from unpredictable sub-optimum performance or even training collapses. Recent…
When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL Handoff
Runze Liu, Jiashun Liu, Xu Wan +2
Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) has become a standard pipeline for Large Language Model (LLM) post-training. SFT is expected to provide a usefu…
Fuz-RL: A Fuzzy-Guided Robust Framework for Safe Reinforcement Learning under Uncertainty
Xu Wan, Chao Yang, Cheng Yang +2
Safe Reinforcement Learning (RL) is crucial for achieving high performance while ensuring safety in real-world applications. However, the complex interplay of multiple uncertainty…
Buffer Matters: Unleashing the Power of Off-Policy Reinforcement Learning in Large Language Model Reasoning
Xu Wan, Yansheng Wang, Wenqi Huang +1
Traditional on-policy Reinforcement Learning with Verifiable Rewards (RLVR) frameworks suffer from experience waste and reward homogeneity, which directly hinders learning efficien…
LLM-Guided Safe Reinforcement Learning for Energy System Topology Reconfiguration
Zongyan Zhang, Chao Shen, Xu Wan +2
The increasing penetration of renewable generation and the growing variability of electrified demand introduce substantial operational uncertainty to modern power systems. Topology…
ProOPF: Benchmarking and Improving LLMs for Professional-Grade Power Systems Optimization Modeling
Chao Shen, Zihan Guo, Xu Wan +6
Growing renewable penetration introduces substantial uncertainty into power system operations, necessitating frequent adaptation of dispatch objectives and constraints and challeng…
AdapThink: Adaptive Thinking Preferences for Reasoning Language Model
Xu Wan, Wei Wang, Wenyue Xu +3
Reinforcement Learning (RL)-based post-training has significantly advanced the complex reasoning capabilities of language models, fostering sophisticated self-reflection processes.…
Think Twice, Act Once: A Co-Evolution Framework of LLM and RL for Large-Scale Decision Making
Xu Wan, Wenyue Xu, Chao Yang +1
Recent advancements in Large Language Models (LLMs) and Reinforcement Learning (RL) have shown significant promise in decision-making tasks. Nevertheless, for large-scale industria…
SrSv: Integrating Sequential Rollouts with Sequential Value Estimation for Multi-agent Reinforcement Learning
Xu Wan, Chao Yang, Cheng Yang +2
Although multi-agent reinforcement learning (MARL) has shown its success across diverse domains, extending its application to large-scale real-world systems still faces significant…
SAMG: Offline-to-Online Reinforcement Learning via State-Action-Conditional Offline Model Guidance
Liyu Zhang, Haochi Wu, Xu Wan +3
Offline-to-online (O2O) reinforcement learning (RL) pre-trains models on offline data and refines policies through online fine-tuning. However, existing O2O RL algorithms typically…