AIPO: Learning to Reason from Active Interaction
cs.CL, cs.AI
Submitted: 2026-05-08
Updated: 2026-09-27
Comments: Work in progress
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Program Synthesis with Large Language Models
- Reflect, Retry, Reward: Self-Improving LLMs via Reinforcement Learning
- Nudging the Boundaries of LLM Reasoning
- Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning
- Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models
- Multi-Agent Evolve: LLM Self-Improve through Co-evolution
- Reasoning with Exploration: An Entropy Perspective
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- Process Reinforcement through Implicit Rewards
- Weight Ensembling Improves Reasoning in Language Models
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning
- The Llama 3 Herd of Models
- SRFT: A Single-Stage Method with Supervised and Reinforcement Fine-Tuning for Reasoning
- FlowReasoner: Reinforcing Query-Level Meta-Agents
- OpenThoughts: Data Recipes for Reasoning Models
- DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning
- Scaling Laws for Autoregressive Generative Modeling
- Deep Learning Scaling is Predictable, Empirically
- REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering