Blog
告别冗长 PDF，一文读懂最新顶会与核心期刊的创新价值。
A Semiclassical Diagnostic for Spacetime Emergence
RepoZero: Can LLMs Generate a Code Repository from Scratch?
SCOPE: Structured Decomposition and Conditional Skill Orchestration for Complex Image Generation
A$^2$RD: Agentic Autoregressive Diffusion for Long Video Consistency
Unified Value Alignment for Generative Recommendation in Industrial Advertising
Decoupling Semantics and Fingerprints: A Universal Representation for AI-Generated Image Detection
Layer Collapse in Diffusion Language Models
Hard to Read, Easy to Jailbreak: How Visual Degradation Bypasses MLLM Safety Alignment
DVD: Discrete Voxel Diffusion for 3D Generation and Editing
Tracing the Arrow of Time: Diagnosing Temporal Information Flow in Video-LLMs
Measuring and Mitigating the Distributional Gap Between Real and Simulated User Behaviors
RCoT-Seg: Reinforced Chain-of-Thought for Video Reasoning and Segmentation
PathPainter: Transferring the Generalization Ability of Image Generation Models to Embodied Navigation
MIND: Monge Inception Distance for Generative Models Evaluation
Topology-Enhanced Alignment for Large Language Models: Trajectory Topology Loss and Topological Preference Optimization
A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications
Finite-time memory detectors and fully constraining Faddeev-Kulish dressings in QED and gravity
IntentGrasp: A Comprehensive Benchmark for Intent Understanding
Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs
Toward Better Geometric Representations for Molecule Generative Models
Syndrome resampling enhances quantum error correction thresholds
LiteGUI: Distilling Compact GUI Agents with Reinforcement Learning
Loop Composition in Quantum Algorithms
Direct Experimental Test of Conformal Invariance via Grazing Scattering: A Proposal for X-ray and Neutron Experiments
Implicit Compression Regularization: Concise Reasoning via Internal Shorter Distributions in RL Post-Training
LookWhen? Fast Video Recognition by Learning When, Where, and What to Compute