Plug-and-Play Verifier for Decoupling Process Reward in Stabilizing Long-Horizon Agentic Reasoning
cs.LG
Submitted: 2026-11-06
Updated: 2026-11-06
Code: https://github.com/momo1443/plug-and-play-verification
Terminology
Sources
- RLFactory: A Plug-and-Play Reinforcement Learning Post-Training Framework for LLM Multi-Turn Tool-Use
- Agent-R1: A Unified and Modular Framework for Agentic Reinforcement Learning
- Do Joint Audio-Video Generation Models Understand Physics?
- Agent-RLVR: Training Software Engineering Agents via Guidance and Environment Rewards
- Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
- TACO: Topics in Algorithmic COde generation dataset
- ABSeeker: Training Long-Horizon Search Agents via Answer-Backtracked Credit Assignment
- SearchAtlas: Analyzing Agentic Search Strategies via Evidential Query Graphs
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- ARLArena: A Unified Framework for Stable Agentic Reinforcement Learning
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning
- RAGEN-2: Reasoning Collapse in Agentic RL
- STAMP: Provenance-Guided Credit Assignment for Deep Search Agents
- Improving Reasoning Ability via Asynchronous On-Policy Self-Distillation under Positive Rollouts
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis