7 papers
VI-CuRL: Stabilizing Verifier-Independent RL Reasoning via Confidence-Guided Variance Reduction
Xin-Qiang Cai, Masashi Sugiyama
Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a dominant paradigm for enhancing Large Language Models (LLMs) reasoning, yet its reliance on external verifier…
Mahjax: A GPU-Accelerated Mahjong Simulator for Reinforcement Learning in JAX
Soichiro Nishimori, Shinri Okano, Keigo Habara +3
Riichi Mahjong is a multi-player, imperfect-information game characterized by stochasticity and high-dimensional state spaces. These attributes present a unique combination of chal…
Positive-Unlabeled Reinforcement Learning Distillation for On-Premise Small Models
Zhiqiang Kou, Junyang Chen, Xin-Qiang Cai +8
Due to constraints on privacy, cost, and latency, on-premise deployment of small models is increasingly common. However, most practical pipelines stop at supervised fine-tuning (SF…
Offline Reinforcement Learning with Domain-Unlabeled Data
Soichiro Nishimori, Xin-Qiang Cai, Johannes Ackermann +1
Offline reinforcement learning (RL) is vital in areas where active data collection is expensive or infeasible, such as robotics or healthcare. In the real world, offline datasets o…
Beyond Simple Sum of Delayed Rewards: Non-Markovian Reward Modeling for Reinforcement Learning
Yuting Tang, Xin-Qiang Cai, Jing-Cheng Pang +3
Reinforcement Learning (RL) empowers agents to acquire various skills by learning from reward signals. Unfortunately, designing high-quality instance-level rewards often demands si…
Reinforcement Learning from Bagged Reward
Yuting Tang, Xin-Qiang Cai, Yao-Xiang Ding +3
In Reinforcement Learning (RL), it is commonly assumed that an immediate reward signal is generated for each action taken by the agent, helping the agent maximize cumulative reward…