Huaxiu Yao
@HuaxiuYaoML • 6,676 subscribers
Assistant Professor @unccs @uncsdss | Postdoc @StanfordAILab | Author of MetaClaw, SkillRL, ClawArena, SimpleMem, AutoResearchClaw, Agent World Model, Agent0
Shorts
Videos
🧠 Can agent memory scale without losing reasoning? 🔥 We’re excited to share our latest work, SimpleMem, a principled memory framework for LLM agents built around semantic lossless compression. 📉 30× fewer inference tokens 📈 +26.4% avg F1 (vs Mem0) ⚡ 50.2% faster retrieval (vs Mem0) Instead of storing raw interaction history 🗂️ or relying on costly iterative reasoning loops 🔁, SimpleMem treats memory as a structured, evolving representation whose primary objective is 🎯 maximizing information density per token. 📄 Paper: 🔗 Code: 📦 Website： Nice work @JiaqiLiu835914, Yaofeng Su, Peng (Richard) Xia, Siwei Han, and great collab. w/ Cihang Xie, Zeyu Zheng, Mingyu Ding
Huaxiu Yao119,426 görüntüleme • 5 ay önce
🚀 SimpleMem now supports Claude Skills! Use SimpleMem in to remember long-term information and project history across conversations. Register at 👉 🔗 Detailed Code: Try it out and let us know what you think! @JiaqiLiu835914, Yaofeng Su, Peng (Richard) Xia, Siwei Han, Cihang Xie, Zeyu Zheng, Mingyu Ding
Huaxiu Yao22,912 görüntüleme • 4 ay önce
🚀 SimpleMem MCP is now LIVE & Open Source! Experience SimpleMem as a cloud-hosted memory service with fast retrieval, fewer tokens, and persistent memory - all exposed via MCP. Easily integrate with Cursor, Claude Desktop, LM Studio, Cherry Studio, and more. 👉 Try it now: 🔗 Code: Nice work again @JiaqiLiu835914, Yaofeng Su, Peng (Richard) Xia, Siwei Han, and great collab. w/ Cihang Xie, Zeyu Zheng, Mingyu Ding
Huaxiu Yao16,002 görüntüleme • 5 ay önce
Can vision-language-action (VLA) models generalize to diverse OOD tasks and align with customized objectives? 🤔 🚀 We introduce GRAPE, a plug-and-play algorithm to generalize robot policies via preference alignment. GRAPE unfolds three benefits to boost the generalizability of VLAs: 👉1. GRAPE aligns VLAs on a trajectory level and endows the model with the ability for global decision-making, instead of merely cloning behavior; 👉2. GRAPE implicitly models reward from both successful and failed trials to boost generalizability to diverse tasks; 👉3. GRAPE adopts a scalable preference synthesis algorithm to rank trajectories with preferences that align with arbitrary objectives. Our experiments on a diverse array of real-world and simulated robotic tasks reveal: 1⃣GRAPE enhances the performance of state-of-the-art VLA models, increasing success rates on in-domain and unseen manipulation tasks by 51.79% and 60.36%; 2⃣GRAPE is versatile to be aligned with diverse objectives and reduce collision rates by 44.31% or rollout length by 11.15% when aligning towards safer or more efficient manipulation policy, respectively. Check out our full project for more details: 🔥 Paper: 🔥 Project: 🔥 Code:
Huaxiu Yao19,988 görüntüleme • 1 yıl önce
Daha fazla içerik yok.