🔥 [CVPR2024] Official implementation of "Self-correcting LLM-controlled Diffusion Models (SLD)

Explore concepts like Self-Correct, Self-Refine, Self-Improve, Self-Contradict, Self-Play, and Self-Knowledge, alongside o1-like reasoning elevation🍓 and hallucination alleviation🍄.

Enhancing LLMs with LoRA

[ICLR 2025] SuperCorrect: Advancing Small LLM Reasoning with Thought Template Distillation and Self-Correction

List of papers on Self-Correction of LLMs.

A comrephensive collection of learning from rewards in the post-training and test-time scaling of LLMs, with a focus on both reward models and learning strategies across training, inference, and post-inference stages.

[ACL'24] Code and data of paper "When is Tree Search Useful for LLM Planning? It Depends on the Discriminator"

Scalable long read self-correction and assembly polishing with multiple sequence alignment

SCoRe: Training Language Models to Self-Correct via Reinforcement Learning

[ICLR 2025] Breaking Mental Set to Improve Reasoning through Diverse Multi-Agent Debate

[ICLR 2025 SSI-FM] Self-Taught Self-Correction for Small Language Models

Official Implementation of "ProgCo: Program Helps Self-Correction of Large Language Models" (Accpeted at ACL2025 Main)

A recursive alignment framework for cognitive coherence, belief tracking, and contradiction repair.

A recursive alignment framework for cognitive coherence, belief tracking, and contradiction repair.

A Python-based Agentic RAG system that uses local LLMs (Ollama) for intelligent documentation crawling, summarization, and vector embeddings (Supabase). Features a self-correcting agentic loop that dynamically refines answers through tool usage and validation, providing a robust, local solution for accurate information retrieval and generation.

A multi-agent cognitive architecture solving the LLM state-dependency problem with persistent memory and a mandatory self-correction loop. An architecture that is built on a more profound and biologically resonant principle: memory is an active component of intelligence itself.

Ground-Truth-Guided Evaluation (GTGE) framework using a multi-agent Crew AI system to improve the self-correction capabilities of Large Language Models (LLMs).

Library and tools to get validated structured JSON from LLMs (Pydantic v2, self-correction). Local-first (Ollama), Streamlit designer, FastAPI server.