AGENT KB Leveraging Cross-Domain Experience for Agentic Problem Solving

A Survey of Latent Reasoning Nice overview on the emerging field of latent reasoning. Great read for AI devs. (bookmark it)

Thanks for sharing! We'll be continue working on the direction of building agent systems that can effectively learn from experience and self-evolve!

🚀 Thrilled to announce our new work: FR3E (First Return, Entropy-Eliciting Explore)! LLM reasoning with Reinforcement Learning often struggles with unstable and inefficient exploration. We propose FR3E, a structured framework to make it more robust & efficient.

A Personal AI Assistant that Sees What You See and Understands You - Powered by Our Most Advanced Memory System 🧠 Imagine having a personal AI that truly remembers you. Not just another chatbot. Not just a search box. A real assistant that sees what you see, understands what

Come and build your own truly **Personal** AI assistant powered by the most advanced memory system MIRIX! Full Free and Open-Source. Download for Mac and experience AI that never forgets: mirix.io

🔥 M+ is at #ICML2025 now! We combine long-term memory (on CPU) with short-term memory (on GPU) for LLMs, pushing efficient long-context modeling to 160k+ tokens. 📍 My co-author Yuanzhe Hu will present M+ in person tomorrow 4:30pm. 👋 Come chat about scaling memory for LLMs!

Paper sharing: Computer scientists have mathematically proven what many have long suspected: for certain types of Boolean logic puzzles, there simply is no clever shortcut. Title：SAT requires exhaustive search Paper: rdcu.be/evDkS Review: scienceblog.com/computer-scien…

感谢认可哈哈哈 会继续持续产出agent领域有意思并且有用的研究的！🫡

感谢推荐！

thanks a lot for posting our recent work! our next step is to train efficient agent models to advance the pareto-front of agentic AI via model-system co-design!

thanks AK!

Chain-of-Agents End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL

it actually sounds nice!

sounds really interesting! will certainly test our multiagent system and agent model on this benchmark! It reminds me of "psychohistory" in the Foundation series🤣

「 Chain-of-Agents, OPPO」 来自OPPO Agent team，太好了。 CoA的核心思想是，让单个LLM却可以完成 Multi-Agent 的工作流， 如果来一个新词，那就是 Multi-Agentic LLM。 今日分享论文： [ Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL ]

Is text-only information enough for LLM/VLM Web Agents? 🤔 Clearly not. 🙅‍♂️ The modern web is a rich tapestry of text, images 🖼️, and videos 🎥. To truly assist us, agents need to understand it all. That's why we built MM-BrowseComp. 🌐 We're introducing MM-BrowseComp 🚀, a new

Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL The paper introduces the Chain-of-Agents paradigm, which enhances large language models’ complex problem-solving capabilities through dynamic multi-agent collaboration and agentic

TreePO Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling