KnowAgent: Knowledge-Augmented Planning for LLM-Based Agents

LLMs as Method Actors: A Model for Prompt Engineering and Architecture

Analyzing and scoring reasoning traces of LLMs

It shows case studies of the LangGraph agent.

More compute and less training. Lets reason with CoT & RL!

ZoomEye: Enhancing Multimodal LLMs with Human-Like Zooming Capabilities through Tree-Based Image Exploration

Code and dataset for the ICLR 2024 paper "Thought Propagation: An analogical Approach to Complex Reasoning with Large Language Models."

Codebase and tutorial of ContPhy dataset generation for ICML 2024 paper "ContPhy: Continuum Physical Concept Learning and Reasoning from Videos"

A Multi-Agent Reasoning Problem Solver. You build teams and they work together to solve the problems you give them.

Autonomous conversational AI agent to work as your assistant coach, strategist, psycologist and personal life ally

Hierarchical Spatial Proximity Reasoning for Vision-and-Language Navigation

Code for paper: Assessing Logical Puzzle Solving in Large Language Models: Insights from a Minesweeper Case Study

BlueMO: A comprehensive collection of challenging mathematical olympiad problems and solutions from the renowned 'Little Blue Book' (小蓝书) series, tailored for evaluating the reasoning capabilities of language models.

easy Augmented Generative Intelligence for LLM

Project involving Human-Robot Interaction and Reasoning Agents modules. Pepper Robot is put into a restaurant environment to accomplish waiter tasks.

Reasoning agent project: Policy Networks for Non-Markovian Reinforcement Learning Rewards

Official repository for the paper 'Meta-Reasoning Improves Tool Use in Large Language Models'.

This architecture mimics subconscious-like processing, enabling AI systems to generate novel solutions through dynamic memory and controlled chaos. ETF unlocks new levels of creativity, decision-making, and problem-solving, moving us closer to true AGI.

AI Forecasting tools to help humans forecast the future. Additionally a framework for building a Metaculus AI Benchmarking Tournament Bot

Code for PS-BEAR (Physics Solving - Basic Expert system and Automated Reasoner)