cd/entity/SDAE· home› entities› SDAE
grep -l @sdae /news/*.json | wc -l → 1

SDAE

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

04:00
2026-09-24
arxiv.org
artificial-intelligence

Reinforcement Learning with Decomposed Subtasks

A new arXiv paper (arXiv:2609.27035v1) introduces Reinforcement Learning with Decomposed Subtasks (RLDS), a method that replaces the scalar GRPO advantage with Subtask-Decomposed Advantage Estimation …

// co-occurs with top 7 entities
RLDS 1 GRPO 1 ScienceWorld 1 FrozenLake 1 HotpotQA 1 DeepResearch 1 arXiv 1
// topics top 5 topics
artificial intelligence 1 machine learning 1 large language models 1 ai research 1 ai agents 1