8 papers
The Generalization Spectrum: A Chromatographic Approach to Evaluating Learning Algorithms
Jinghan Zhang, Zerui Cheng, Shiqi Chen +5
Traditional evaluations measure a learning algorithm's final performance on an i.i.d. test set, reducing learning to a single aggregate score. This approach obscures a fundamental…
The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL
Yingru Li, Jiawei Xu, Ziniu Li +10
Reinforcement Learning (RL) for Large Language Models (LLMs) often suffers from training collapse in long-horizon tasks due to exploding gradient variance. To mitigate this, a base…
In-Place Test-Time Training
Guhao Feng, Shengjie Luo, Kai Hua +4
The static ``train then deploy" paradigm fundamentally limits Large Language Models (LLMs) from dynamically adapting their weights in response to continuous streams of new informat…
Learn Hard Problems During RL with Reference Guided Fine-tuning
Yangzhen Wu, Shanda Li, Zixin Wen +5
Reinforcement learning (RL) for mathematical reasoning can suffer from reward sparsity: for challenging problems, LLM fails to sample any correct trajectories, preventing RL from r…
Dynamic Vocabulary Pruning: Stable LLM-RL by Taming the Tail
Yingru Li, Jiawei Xu, Jiacai Liu +6
Reinforcement Learning (RL) for Large Language Models (LLMs) faces a fundamental tension: the numerical divergence between high-throughput inference engines and numerically precise…
Mitigating LLM Hallucination via Behaviorally Calibrated Reinforcement Learning
Jiayun Wu, Jiashuo Liu, Zhiyuan Zeng +3
LLM deployment in critical domains is currently impeded by persistent hallucinations--generating plausible but factually incorrect assertions. While scaling laws drove significant…