Institution

Tianjin University

A major public research university in Tianjin, China; the TJUNLP Lab in its College of Intelligence and Computing works on NLP and large language models.

Reinforcement Learning · Tianjin University

Why Multi-Domain RL Forgets, and How a Math Refresh Heals It

When you RL-tune an LLM across math, code, QA, and writing in sequence, math drops from 66.49 to 57.66 even though gradients look orthogonal. A short math refresh pulls it back to 66.04 without wrecking the other three.