Bootstrapping Exploration with Group-Level Natural Language Feedback in Reinforcement Learning
Paper • 2603.04597 • Published • 210
DailyPapers
-
-
SII-Enigma/Llama3.2-8B-Ins-AMPO
Text Generation • 8B • Updated • 76 -
Understanding R1-Zero-Like Training: A Critical Perspective
Paper • 2503.20783 • Published • 59 -
Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs
Paper • 2509.25779 • Published • 19 -
On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
Paper • 2306.13649 • Published • 31 -
Unfamiliar Finetuning Examples Control How Language Models Hallucinate
Paper • 2403.05612 • Published • 3 -
Reinforcement Learning Finetunes Small Subnetworks in Large Language Models
Paper • 2505.11711 • Published • 11 -
Reinforcement Learning via Self-Distillation
Paper • 2601.20802 • Published • 43 -
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Paper • 2509.02547 • Published • 237 -
AI Can Learn Scientific Taste
Paper • 2603.14473 • Published • 413 -
REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models
Paper • 2501.03262 • Published • 104 -
GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks
Paper • 2510.04374 • Published -
SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?
Paper • 2502.12115 • Published • 46 -
Persona Features Control Emergent Misalignment
Paper • 2506.19823 • Published -
π_{0.5}: a Vision-Language-Action Model with Open-World Generalization
Paper • 2504.16054 • Published • 4