Thickening-to-Thinning: Reward Shaping via Human-Inspired Learning Dynamics for LLM Reasoning

Wenze Lin, Zhen Yang, Xitai Jiang, Pony Ma, Gao Huang · 2026 · arXiv 2602.04265

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

open full Pith review browse 1 citing papers arXiv PDF

representative citing papers

Rethinking the Comparison Unit in Sequence-Level Reinforcement Learning: An Equal-Length Paired Training Framework from Loss Correction to Sample Construction

cs.LG · 2026-04-19

citing papers explorer

Showing 1 of 1 citing paper.

Rethinking the Comparison Unit in Sequence-Level Reinforcement Learning: An Equal-Length Paired Training Framework from Loss Correction to Sample Construction cs.LG · 2026-04-19 · unreviewed · ref 23 · internal anchor