About
Research & Development
Self-Evolution Agent: A New Approach to Computer Control
TLDR: SEA is a 7-billion parameter AI agent designed for computer use that achieves high performance by introducing a novel closed-loop data generation pipeline...
Research & Development
Enhancing Mathematical Problem-Solving with Knowledge Graphs and Executable Code
TLDR: KGA-ECoT is a new framework that significantly improves Large Language Models' (LLMs) ability to solve complex mathematical problems and generate accurate code. It...
Research & Development
AgREE: A New Approach to Keeping Knowledge Graphs Current with Emerging Data
TLDR: AgREE is a novel agent-based framework that uses iterative retrieval and multi-step reasoning to dynamically build knowledge graph triplets, especially for new and...
Research & Development
AI Agents Master Collaboration: A Hybrid Approach to Ad Hoc Teamwork
TLDR: This research paper introduces a new AI architecture for Ad Hoc Teamwork (AHT), where AI agents collaborate with others without prior coordination. The...
Research & Development
New Framework for Designing Memory Challenges in Reinforcement Learning Environments
TLDR: This research introduces a novel framework for creating synthetic Partially Observable Markov Decision Process (POMDP) environments to rigorously test memory-augmented Reinforcement Learning (RL)...
Research & Development
Deliberative Reasoning Networks: A New Path to Logical AI
TLDR: The Deliberative Reasoning Network (DRN) is a novel AI paradigm that improves large language models' logical reasoning by focusing on minimizing uncertainty rather...