Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 Trajectories
Zhepei Wei, Xinyu Zhu, Wei-Lin Chen +3
Reinforcement learning with verifiable rewards (RLVR) has become a dominant paradigm for improving reasoning in large language models (LLMs), yet the underlying geometry of the res…
cs.LG2026
G-Zero: Self-Play for Open-Ended Generation from Zero Data
Chengsong Huang, Haolin Liu, Tong Zheng +7
Self-evolving LLMs excel in verifiable domains but struggle in open-ended tasks, where reliance on proxy LLM judges introduces capability bottlenecks and reward hacking. To overcom…