让大模型真正「记住你」｜LightMem：轻量高效的记忆增强生成框架
xzwnlp
• • 2NEW Articles from Team or Enterprise organizations will get promoted to the main section.
Exploring New Frontiers of LLMs: Adaptive Dual-Search Distillation (ADS) and the 30B Model Open Beta
NoesisLab
• • 2Benchmarks + Report: Optimized Cosmos-Reason2 (Qwen3-VL) for on-device inference on 8GB RAM (Jetson Orin Nano Super)
JonnaMat
• Making LLMs Truly Remember You | LightMem: Lightweight and Efficient Memory-Augmented Generation
xzwnlp
• • 5构建、评估与连接 AI 技能 | SkillNet：大规模智能体“技能图谱”知识库
xzwnlp
• • 10Create, Evaluate, and Connect AI Skills | SkillNet: A Large-Scale Agentic "Skill Graph" Knowledge Base
xzwnlp
• • 13Prompt Engineering Is Fading — The Era of Skills, Tools, and Frameworks Has Arrived
sadpig70
• • 1A framework and leaderboard for Retrieval Pipelines evaluation on ViDoRe v3
antoineedy
• • 13Synthetic-XRay
chimbiwide
• • 2**New Trackio Integration: From LLM Observability to Rapid Experimentation and Real-Time Control**
rapidfire-ai-inc
• AI Coding Assistants Keep Shipping Vulnerable Code -- Here's What We're Doing About It
scthornton
• • 3How I Landed Multiple Senior LLM Engineer Offers (And The Brutal Reality of AI Interviews Right Now)
herooooooooo
• • 3Bringing Autonomous Driving RL to OpenEnv and TRL
sergiopaniego
• • 22PicoClaw, PicoLM, and the $10-$15 hardware based Edge AI
Javedalam
• • 1Your MoE Model Does Not Have to Select Fixed Number of Experts
Spico
• • 8GRPO: Building Intuition Through Ablation Studies
garg-aayush
• • 6Shattering the Memory Wall: O(1) Inference and Causal Monoid State Compression in Spartacus-1B
OzTianlu
• • 2Visual Aesthetic Benchmark: Can Frontier Models Judge Beauty?
zhangchenxu
• • 14Transformer-Based SSL Embeddings vs ECAPA-TDNN for Speaker Recognition
norwooodsystems
• SECourses Upscaler Pro Beating Topaz AI by Far With Specalized FlashVSR+ & SeedVR2.5 - Local Windows
MonsterMMORPG
• • 2