Authors

Yuhan Dong

Publications 2

Preprint Open access

Learn Here, Move Less Elsewhere: Input-Conditioned Plasticity from Retained-Domain Activation Atlases

Jiangtao Lin, Bangyang Wei, Yihang Ding et al. · 2026

Task-specific fine-tuning can rewrite a language model's answers beyond the training task, complicating updates that must preserve existing behavior. We introduce ATLAS, which turns retained-domain representations into an input-dependent rule for task adaptation. An activation atlas supplies local reference centers and …

Preprint Open access

From Soft Targets to Reward Signals: How Assignment and Reward Objectives Interact

Jiangtao Lin, Bangyang Wei, Siyi Liu et al. · 2026

Soft preference targets specify supervision strength, and reward objectives convert that strength into learned reward signals. A central design question remains: how does assigning a fixed set of preference strengths to different response pairs change the rewards produced by different objectives? We introduce assignmen …

Co-authors