AgentFlow: In-the-Flow Agentic System Optimization
-
Updated
Feb 8, 2026 - Python
AgentFlow: In-the-Flow Agentic System Optimization
[arxiv: 2503.23895] Dynamic Parametric Retrieval Augmented Generation for Test-time Knowledge Enhancement
Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache
The data and implementation for the experiments in the paper "Flows: Building Blocks of Reasoning and Collaborating AI".
Belief-Bias evaluation of local LLMs
An open-source attempt at training a variant of LeCun's energy-based models (EBM) to reason in latent space and solve Sudoku.
An advanced, fully local, and GPU-accelerated RAG pipeline. Features a sophisticated LLM-based preprocessing engine, state-of-the-art Parent Document Retriever with RAG Fusion, and a modular, Hydra-configurable architecture. Built with LangChain, Ollama, and ChromaDB for 100% private, high-performance document Q&A.
Measure of estimated confidence for non-hallucinative nature of outputs generated by Transformer-based Language Models.
Self-Improving LLMs Through Iterative Refinement
Latent-Explorer is the Python implementation of the framework proposed in the paper "Unveiling LLMs: The Evolution of Latent Representations in a Dynamic Knowledge Graph".
A new training framework for Trustworthy Large Reasoning Models
Memory AutoResearch Agent that tests different hypothesis to create better memory inspired by @karpathy
A curated dataset of 31,822 annotated passages from Italian political speeches, featuring expert-validated explanations of implicatures and presuppositions. Derived from the IMPAQTS corpus, this resource supports research on implicit content, pragmatic inference, and political language understanding.
Multi-Agent System Process Verification (MAS-ProVe)
Causal reasoning evaluation of LLMs with human benchmark
Análise de Intervenção em Ansiedade com Descoberta Causal
A neural network calculates your chance of finding love.
AI agents and Agentic AI projects - ReAct agents , Reflection Agents(Generator + Reflector) etc.
Can LLMs really 'Think'? This repo contains implementation of different types of memory for LLMs.
Quantifying how close is a coding AI to AGI
Add a description, image, and links to the llms-reasoning topic page so that developers can more easily learn about it.
To associate your repository with the llms-reasoning topic, visit your repo's landing page and select "manage topics."