Signals 4 · free daily AI digest

Score Centering Stabilizes Off-policy Reinforcement Learning

Paper recorded by Signals 4 on 2026-09-17 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-17 on arXiv · recorded by Signals 4 on 2026-09-18

Category: cs.LG · 机器学习 · first seen 2026-09-18

Abstract

Reinforcement learning (RL) of large language models is notoriously sensitive to small differences between training and inference engines, often referred to as the training-inference mismatch (TIM). However, completely eliminating TIM is impractical, as it would come at a major cost to rollout efficiency. In this paper, we show that the instability of RL under TIM is primarily caused by drift: a p

Read on arXiv →

#3 most recent of 215 cs.LG papers we have recorded · ↑ newer: How Does Distribution Shift Shape Pretraining Gains in Neural PDE Surr · ↓ older: PosteriorBench: From Point Estimates to Posterior Matching in Evaluati
Cite this page: Score Centering Stabilizes Off-policy Reinforcement Learning: the #3 most recent of 215 cs.LG papers we have recorded (as of 2026-09-17). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/score-centering-stabilizes-off-policy-reinforcement-learning.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.LG papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions