Just now, MiniMax responded to the controversy about the license change on the M2.7 model.
@jiqizhixin
-
LLMs Fairness and Consistency in AI Model Evaluation
By
–
Are LLMs truly fair and consistent when judging other AI models? A collaborative team from Peking University, NUS, Institute of Science Tokyo, Nanjing University, Carnegie Mellon, Westlake, and Southeast University has the answer! They introduce TrustJudge, a probabilistic
-
Point-VLA: Robots Understanding Real-World Scenes Accurately
By
–
How can we make robots understand exactly what we mean, even in the most cluttered real-world scenes? Researchers from Tongji University, Shanghai Jiao Tong University, Spirit AI, and Tsinghua University introduce Point-VLA. This innovative system supercharges robot
-
InfoTok: Efficient Long Video Processing Using Information Theory
By
–
How do we process long videos efficiently without losing crucial information? NVIDIA, Stanford University, and National University of Singapore have an answer! They introduce InfoTok, a breakthrough method inspired by Shannon's information theory. It intelligently allocates
-
Goal-VLA: Zero-Shot Robot Manipulation from Images and Instructions
By
–
What if robots could perform complex manipulation tasks with zero prior examples, just from an image and instruction? Researchers from National University of Singapore, The University of Hong Kong, Peking University, and Tsinghua University present Goal-VLA! Their Goal-VLA uses
-
ADM-v2: AI Model for Reliable Future Prediction Offline Learning
By
–
Can your AI reliably predict the future for robust offline learning? Researchers from Nanjing University and the University of Montréal introduce ADM-v2. Unlike prior models that accumulate errors over long sequences, ADM-v2 directly forecasts full episodes, offering
-
OpenResearcher: Open Pipeline for AI-Powered Deep Web Research
By
–
How do we train AI agents to perform complex, multi-step research efficiently and reproducibly? Researchers from Texas A&M, University of Waterloo, UC San Diego, Verdent AI, NetMind AI, and Lambda introduce OpenResearcher, a fully open pipeline that simulates deep web research
-
AI Framework Decodes Real-Time Decision-Making for Critical Safety
By
–
How can AI truly understand your real-time decision-making for ultimate safety in critical situations? Researchers at the University of Michigan unveil a groundbreaking approach. They developed Self-Supervised Evolutionary Learning (SSEL), an AI framework that directly decodes
-
DVD Achieves Video Depth Estimation Without Geometric Errors
By
–
Can we finally achieve highly accurate video depth estimation without geometric errors or massive datasets? Researchers from HKUST(GZ), Princeton University, and a global team introduce DVD. DVD ingeniously transforms pre-trained video diffusion models into precise, single-pass
-
LLMs Enable Precision Multi-Robot Industrial Task Coordination
By
–
Can LLMs conquer the precision and complexity of industrial multi-robot tasks? Researchers from Shenzhen University, SpeedBot Robotics, Carleton University, and the Chinese Academy of Sciences present IMR-LLM! They combined LLMs with smart graph structures for efficient