What’s Happening in My Field

Reading

What's happening in my field

Recent arXiv papers in RL post-training, reasoning, self-improvement and multimodal LLMs.

Updated weekly · last update 2026-10-26
RL Training
NgramRL: Accelerating RL Post-Training for Structured Generation via Rollout-Aware N-gram Speculative Decoding
Nilesh Prasad Pandey, Jangseon Park, Jason Kong et al.
2026-10-23
 Self-Improvement
On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models
Leon Sick, Lukas Hoyer, Yannick Strümpler et al.
2026-10-23
 Self-Improvement
The Predictive--Optimization Gap in GuanDan Self-Play
Qicheng Zou, Biao Yuan, Siyu Song et al.
2026-10-21
 Self-Improvement
Recursive Self-Improvement on a Budget: Scoring Code Edits Before They Run
Erfan Darzi, Sree Bhargavi Balija, Sheng He
2026-10-21
 Self-Improvement
Auditing the Product, Not the Recursion: The Evaluation Evidence Behind Recursive Self-Improvement Claims
Zihan Chen, Di Zhu, Lei Nico Zheng
2026-10-20
 Self-Improvement
HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs
Zhuoshi Pan, ZhiYuan Feng, Yukuo Cen et al.
2026-10-19
 Audio Reasoning
Text Dominance in Paralinguistic Conflict: Tracing Suppression Through Audio LLM Circuits
Ankur Banerjee, Vishnu Guntakala, Danity Pike et al.
2026-10-19
 RL Training
Where Groups Split: A Probe Prior for Nonzero Within-Group Variance in RL Post-Training
Jonah Ellison, Mateo Ruiz, Owen Pell et al.
2026-10-18
 Reasoning
Process Reward Modeling with Reliability- and Conservatism-Aware Bayesian Belief
Zhongjian Qiao, Zhixiang Zhou, Jiafei Lyu et al.
2026-10-16

← All posts