Blog
告别冗长 PDF，一文读懂最新顶会与核心期刊的创新价值。
Risk-Aware World Model Predictive Control for Generalizable End-to-End Autonomous Driving
UniVBench: Towards Unified Evaluation for Video Foundation Models
Partial recovery of meter-scale surface weather
AdaEvolve: Adaptive LLM Driven Zeroth-Order Optimization
ViCLIP-OT: The First Foundation Vision-Language Model for Vietnamese Image-Text Retrieval with Optimal Transport
Semantic Tube Prediction: Beating LLM Data Efficiency with JEPA
A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring
TorchLean: Formalizing Neural Networks in Lean
SPATIALALIGN: Aligning Dynamic Spatial Relationships in Video Generation
NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning
SO3UFormer: Learning Intrinsic Spherical Features for Rotation-Robust Panoramic Segmentation
dLLM: Simple Diffusion Language Modeling
Toward Personalized LLM-Powered Agents: Foundations, Evaluation, and Future Directions
Denoising as Path Planning: Training-Free Acceleration of Diffusion Models with DPCache
AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications
General Agent Evaluation
SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning
HulluEdit: Single-Pass Evidence-Consistent Subspace Editing for Mitigating Hallucinations in Large Vision-Language Models
VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous Driving
When to Act, Ask, or Learn: Uncertainty-Aware Policy Steering
EgoAVFlow: Robot Policy Learning with Active Vision from Human Egocentric Videos via 3D Flow
DHP: Efficient Scaling of MLLM Training with Dynamic Hybrid Parallelism
Understanding Usage and Engagement in AI-Powered Scientific Research Tools: The Asta Interaction Dataset
Tuning the magnetic properties of Kitaev materials via the antiferromagnetic proximity effect: Novel phases and application to an $α$-RuCl$_3$/MnPS$_3$ bilayer
Efficient Encoder-Free Fourier-based 3D Large Multimodal Model
Self-Correcting VLA: Online Action Refinement via Sparse World Imagination
CLIP Is Shortsighted: Paying Attention Beyond the First Sentence
A Mixture-of-Experts Model for Multimodal Emotion Recognition in Conversations
Towards Autonomous Memory Agents
SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generation