Chu-Cheng Lin

Papers on Lattice

Total citations

Topics

h-index

Publication activitypapers/week, last 8 weeks

Research focus

Reasoning & Chain-of-Thought (1)Training Efficiency & Optimization (1)

Frequent co-authors

Eugene Ie (1)

Papers (1)

Apr 28, 2026

Chu-Cheng Lin +13w ago

How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum

Stuck training your reasoning model with RLVR due to a low initial success rate? This paper shows how a Tsallis q-logarithm loss can jumpstart learning by adaptively amplifying gradients, achieving a +14.4 point boost over GRPO on HotPotQA.

Chu-Cheng Lin, Eugene Ie

Reasoning & Chain-of-Thought Training Efficiency & Optimization

Search

Chu-Cheng Lin

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (1)