T 5: Twin-Critic Training for Token-Level Thoughts in Reinforcement Mid-Training
cs.AI
Submitted: 2026-09-26
Updated: 2026-09-26
Terminology
Sources
- PIQA: Reasoning about Physical Commonsense in Natural Language
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Training Verifiers to Solve Math Word Problems
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Reinforcement Pre-Training
- RLP: Reinforcement as a Pretraining Objective
- Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning
- REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization
- Fast Quiet-STaR: Thinking Without Thought Tokens
- Reinforcement Learning on Pre-Training Data
- Understanding R1-Zero-Like Training: A Critical Perspective
- Beyond Uniform Token-Level Trust Region in LLM Reinforcement Learning
- Score Centering Stabilizes Off-policy Reinforcement Learning
- OpenWebMath: An Open Dataset of High-Quality Mathematical Web Text
- The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale
- Less is More: Clustered Cross-Covariance Control for Offline RL
- High-Dimensional Continuous Control Using Generalized Advantage Estimation
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Reinforcement Mid-Training
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection