What if your AI could tell you it's unsure, instead of confidently making things up? Researchers from ByteDance Seed, CMU, and Fudan University have a breakthrough. They've developed 'behavioral calibration,' a new reinforcement learning approach that trains LLMs to recognize
@jiqizhixin
-
Foundation Agent Memory Framework for Long Interactions
By
–
Tired of AI agents forgetting crucial context in long interactions? A groundbreaking survey from multiple universities introduces a unified framework for foundation agent memory. This work explores how AI can effectively store, manage, and retrieve vast amounts of information
-
Block Diffusion Language Models Enable Faster AI Reasoning
By
–
What if your AI could reason smarter AND faster, especially on complex tasks? A collaboration from Fudan University, Peking University, and Meituan LongCat Team just made it happen! They've developed a new framework for Block Diffusion Language Models (BDLMs) that lets the AI
-
LongHorizonUI: AI Agents Master Complex Interface Tasks
By
–
Why do AI agents stumble on long, complex interface tasks? A team from Tencent Turing Lab, Georgia Tech, Tsinghua, and Chinese Academy of Sciences presents LongHorizonUI. This unified framework uses a "Perceiver" for robust screen understanding, a "Decider" for iterative,
-
DeepImageSearch: New AI Agent Revolutionizes Visual Search
By
–
Are your current image searches missing the full story behind what you're looking for? Researchers from Renmin University of China and OPPO Research Institute are pioneering a new approach to visual search. They introduce DeepImageSearch, a novel AI agent that treats image
-
PlugMem: Universal Memory Module Breakthrough for LLM Agents
By
–
Could LLM agents finally gain truly adaptable, long-term memory? University of Illinois Urbana-Champaign, @Tsinghua_Uni , and Microsoft Research team up for a breakthrough! They introduce PlugMem, a universal plugin memory module that attaches to any LLM agent. Instead of
-
TextPecker: AI Solution for Clearer Text-to-Image Generation
By
–
Why does AI-generated text still look so distorted or blurry? Huazhong University of Science and Technology and ByteDance present TextPecker, a groundbreaking solution! TextPecker is a clever plug-and-play strategy that teaches text-to-image AI to "see" and correct structural
-
NeurIPS 2026 Handbook available official conference source
By
–
Source:
https://neurips.cc/Conferences/2026/MainTrackHandbook
… -
NeurIPS Restricts Submissions from US-Sanctioned Institutions
By
–
Breaking: Academic freedom no more. The NeurIPS Foundation has announced it will no longer accept submissions from US-sanctioned institutions.
-
DM0: Embodied-Native VLA Framework for Physical Understanding
By
–
How do we build AI that truly understands the physical world from the ground up, not as an afterthought? A team from Dexmal & @StepFun_ai just unveiled DM0, a groundbreaking Embodied-Native Vision-Language-Action (VLA) framework. Unlike adapting web-trained models, DM0 learns