Skip to content
conftrace_

token-level exploration

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

reinforcement learning fine-tuning (1) large language model (1) kl penalty (1) critical token (1) reward signal (1) chain-of-thought reasoning (1) kl divergence (1) self-supervised learning (1) mathematical reasoning (1) policy optimization (1)

Papers

In-Token Rationality Optimization: Towards Accurate and Concise LLM Reasoning via Self-Feedback AAAI 2026 Ignore the KL Penalty! Boosting Exploration on Critical Tokens to Enhance RL Fine-Tuning NAACL 2025