About
Research & Development
Scalable Option Learning: Unlocking Large-Scale Training for Hierarchical AI
TLDR: Scalable Option Learning (SOL) is a new hierarchical reinforcement learning algorithm that achieves 25x higher training throughput than existing methods. Developed by FAIR...
Research & Development
InfoSeek: A New Framework for Training Language Models in Deep Research
TLDR: InfoSeek is a novel, open-source framework that synthesizes large-scale, complex Deep Research tasks, formalizing them as Hierarchical Constraint Satisfaction Problems (HCSPs). It uses...
Research & Development
Enhancing Graph Neural Network Resilience with PerturbEmbedding
TLDR: The paper introduces PerturbEmbedding, a novel method to enhance Graph Neural Networks (GNNs) against adversarial attacks and improve their generalization. Unlike traditional methods...
Research & Development
Building Robots with Spatial Awareness: A Deep Dive into Scene Understanding and Reasoning
TLDR: This research explores how to equip robots with "Embodied Spatial Intelligence," enabling them to understand and interact with the 3D world using natural...
Research & Development
Unveiling the Strengths and Weaknesses of AI Models in Understanding Video Responses
TLDR: Researchers introduce VideoRewardBench, a new comprehensive benchmark for evaluating multimodal reward models (MRMs) in video understanding. It addresses limitations of previous benchmarks by...
Research & Development
Advanced Hashing for Intelligent Video Group Activity Retrieval
TLDR: This research introduces STVH (Spatiotemporal Video Hashing) and its enhanced version, M-STVH (Multi-Focused Spatiotemporal Video Hashing), to address the challenge of rapidly retrieving...