Matched-Context Group Policy Optimization for LLM Agent Training

cs.LG, cs.AI

Submitted: 2026-10-09

Updated: 2026-10-09

Comments: 29 pages, 9 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Reinforcement Learning for Long-Horizon Interactive LLM Agents
• Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States
• Progress- and Reliability-Oriented Group Policy Optimization for Agentic Reinforcement Learning
• Group-in-Group Policy Optimization for LLM Agent Training
• DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training
• REBEL: Reinforcement Learning via Regressing Relative Rewards
• Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
• Hierarchy-of-Groups Policy Optimization for Long-Horizon Agentic Tasks
• HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents
• VinePPO: Refining Credit Assignment in RL Training of LLMs
• Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs
• RubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable Rewards
• When Denser Credit Is Not Enough: Evidence-Calibrated Policy Optimization for Long-Horizon LLM Agent Training
• RTPO: Reverse-Turn Policy Optimization for Stabilizing Agentic RL Training
• SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learning
• Improve Mathematical Reasoning in Language Models by Automated Process Supervision
• Agent Lightning: Train ANY AI Agents with Reinforcement Learning
• Mitigating False Credit Propagation: Probabilistic Graphical Reward Aggregation for Rubric-Based Reinforcement Learning
• SimPO: Simple Preference Optimization with a Reference-Free Reward
• Dialogue Model Optimization via Agent Game and Adaptive Tree-based GRPO

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks