6 papers · 1 filter
Tail-Likelihood Reinforcement Learning
Shrinivas Ramasubramanian, Daman Arora, Fahim Tajwar +11
Reinforcement learning typically optimizes average reward. For generative policies, the average can hide an important distinction: two policies can achieve the same mean reward whi…
Expanding the Capabilities of Reinforcement Learning via Text Feedback
Yuda Song, Lili Chen, Fahim Tajwar +5
The success of RL for LLM post-training stems from an unreasonably uninformative source: a single bit of information per rollout as binary reward or preference label. At the other…
On the Emergence of Implicit Curriculum in RLVR Learning Dynamics
Yu Huang, Zixin Wen, Yuejie Chi +4
Reinforcement learning with verifiable rewards (RLVR) has been a main driver of recent breakthroughs in large reasoning models. Yet it remains a mystery how rewards based solely on…
To Distill or Decide? Understanding the Algorithmic Trade-off in Partially Observable Reinforcement Learning
Yuda Song, Dhruv Rohatgi, Aarti Singh +1
Partial observability is a notorious challenge in reinforcement learning (RL), due to the need to learn complex, history-dependent policies. Recent empirical successes have used pr…
The Importance of Online Data: Understanding Preference Fine-tuning via Coverage
Yuda Song, Gokul Swamy, Aarti Singh +2
Learning from human preference data has emerged as the dominant paradigm for fine-tuning large language models (LLMs). The two most common families of techniques -- online reinforc…
Hybrid Reinforcement Learning from Offline Observation Alone
Yuda Song, J. Andrew Bagnell, Aarti Singh
We consider the hybrid reinforcement learning setting where the agent has access to both offline data and online interactive access. While Reinforcement Learning (RL) research typi…