RL Post-Training Builds Compositional Reasoning Strategies
cs.AI, cs.CL
Submitted: 2026-07-08
Updated: 2026-11-17
Terminology
Sources
- Reasoning with Exploration: An Entropy Perspective
- SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
- Training Verifiers to Solve Math Word Problems
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Let's Verify Step by Step
- ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
- Understanding R1-Zero-Like Training: A Critical Perspective
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Rethinking Sample Polarity in Reinforcement Learning with Verifiable Rewards
- Solving math word problems with process- and outcome-based feedback
- Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning
- The Invisible Leash: Why RLVR May or May Not Escape Its Origin
- A Minimalist Approach to LLM Reasoning: from Rejection Sampling to Reinforce
- From $f(x)$ and $g(x)$ to $f(g(x))$: LLMs Learn New Skills in RL by Composing Old Ones
- Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
- STaR: Bootstrapping Reasoning With Reasoning
- Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining
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