self-correction
Self-correction in AI refers to a model's ability to recognize and adjust its predictions or decisions based on feedback or new information. This capability enables continuous learning and improvement over time, enhancing the model's effectiveness in dynamic environments.
- Any Large Language Model Can Be a Reliable Judge: Debiasing with a Reasoning-based Bias Detector
- CURE: Co-Evolving Coders and Unit Testers via Reinforcement Learning
- Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
- ConfTuner: Training Large Language Models to Express Their Confidence Verbally
- Latent Principle Discovery for Language Model Self-Improvement
- Position: Machine Learning Conferences Should Establish a "Refutations and Critiques" Track
- ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning
- SRPO: Enhancing Multimodal LLM Reasoning via Reflection-Aware Reinforcement Learning
- Sherlock: Self-Correcting Reasoning in Vision-Language Models
- SiriuS: Self-improving Multi-agent Systems via Bootstrapped Reasoning
- Thinker: Learning to Think Fast and Slow