Signals 4 · free daily AI digest

Sequential Beats Joint: On the Interplay between On-Policy Distillation and RLVR

Paper recorded by Signals 4 on 2026-09-03 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-03 on arXiv · recorded by Signals 4 on 2026-09-04

Category: cs.AI · 人工智能 · first seen 2026-09-04

Abstract

Reinforcement learning with verifiable rewards (RLVR) and on-policy distillation (OPD) have emerged as two dominant methods for post-training reasoning LLMs. Prior work uses OPD's dense token-level supervision to complement the sparse RL reward, fusing the two signals within a single step: either as a \emph{weighted-additive combination} or a \emph{teacher-modulated rescaling} of the RL advantage.

Read on arXiv →

#199 most recent of 300 cs.AI papers we have recorded · ↑ newer: Epistemic Warrant for LLM Recommendations: Characterizing the Basis fo · ↓ older: Why Gated DeltaNet Survives 4-Bit Quantization: NVFP4 W4A4 for the Rec
Cite this page: Sequential Beats Joint: On the Interplay between On-Policy Distillation and RLVR: the #199 most recent of 300 cs.AI papers we have recorded (as of 2026-09-03). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/sequential-beats-joint-on-the-interplay-between-on-policy-distillation-and-rlvr.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.AI 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