cd/entity/AIME25· home› entities› AIME25
grep -l @aime25 /news/*.json | wc -l → 1

@AIME25

mentions 1 type Organization feed RSS
18:22
2026-05-16
research.nvidia.com
large-language-models

iGRPO: Self-Feedback-Driven LLM Reasoning

Researchers introduced Iterative Group Relative Policy Optimization (iGRPO), a two-stage reinforcement learning method that improves large language model reasoning by having the model generate and ref…

// co-occurs with top 7 entities
GRPO 1 iGRPO 1 Nemotron-H-8B-Base-8K 1 DeepSeek-R1 Distilled 1 OpenReasoning-Nemotron-7B 1 AceReason-Math 1 AIME24 1