SARL: Label-Free Reinforcement Learning by Rewarding Reasoning Topology
cs.AI
Submitted: 2026-03-30
Updated: 2026-11-15
Code: https://github.com/huggingface/trl
Terminology
Sources
- The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning
- Training Verifiers to Solve Math Word Problems
- What Characterizes Effective Reasoning? Revisiting Length, Review, and Structure of CoT
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Measuring Mathematical Problem Solving With the MATH Dataset
- Process Reward Models That Think
- WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild
- OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignment
- Let's reward step by step: Step-Level reward model as the Navigators for Reasoning
- Topology of Reasoning: Understanding Large Reasoning Models through Reasoning Graph Properties
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- HybridFlow: A Flexible and Efficient RLHF Framework
- The Shape of Reasoning: Topological Analysis of Reasoning Traces in Large Language Models
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Surrogate Signals from Format and Length: Reinforcement Learning for Solving Mathematical Problems without Ground Truth Answers
- Qwen3 Technical Report
- Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization
- Learning to Reason without External Rewards
- LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models
- The Geometry of Reasoning: Flowing Logics in Representation Space
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- 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
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents