ERILS mixes length-controlled external rollouts from a stronger policy with on-policy rollouts and processes their rewards separately, substantially improving dLLM reinforcement learning on reasoning tasks.
degree in the Interdisciplinary Program in Artificial Intelli- gence at Seoul National University, Seoul, Korea
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
Beyond On-Policy Exploration: Integrating External Policy Rollouts for Reinforcement Learning in Diffusion Language Models
ERILS mixes length-controlled external rollouts from a stronger policy with on-policy rollouts and processes their rewards separately, substantially improving dLLM reinforcement learning on reasoning tasks.