Self-Improving Training & Thinking
cs.CL, cs.AI, cs.LG
Submitted: 2026-01-29
Updated: 2026-11-10
Code: https://github.com/huggingface/Math-Verify
Terminology
Sources
- GPT-4 Technical Report
- SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
- Physics of Language Models: Part 3.1, Knowledge Storage and Extraction
- Constitutional AI: Harmlessness from AI Feedback
- Train for Truth, Keep the Skills: Binary Retrieval-Augmented Reward Mitigates Hallucinations
- Learning to Reason for Factuality
- RM-R1: Reward Modeling as Reasoning
- Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models
- BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
- 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
- Build it Break it Fix it for Dialogue Safety: Robustness from Adversarial Human Attack
- Reinforcement Pre-Training
- Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
- Rewriting Pre-Training Data Boosts LLM Performance in Math and Code
- RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
- Training Large Language Models to Reason in a Continuous Latent Space
- Reformulation for Pretraining Data Augmentation
- ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data