Blog
告别冗长 PDF，一文读懂最新顶会与核心期刊的创新价值。
NuMuon: Nuclear-Norm-Constrained Muon for Compressible LLM Training
Geodesic flows on a black-hole background
Benchmarking Quantum Computers via Protocols, Comparing IBM's Heron vs IBM's Eagle
Tell2Adapt: A Unified Framework for Source Free Unsupervised Domain Adaptation via Vision Foundation Model
Fault-tolerant execution of error-corrected quantum algorithms
QGPU: Parallel logic in quantum LDPC codes
Measurement Induced Asymmetric Entanglement in Deconfined Quantum Critical Ground State
Beyond Scattered Acceptance: Fast and Coherent Inference for DLMs via Longest Stable Prefixes
Mathematicians in the age of AI
Reasoning Models Struggle to Control their Chains of Thought
Building AI Coding Agents for the Terminal: Scaffolding, Harness, Context Engineering, and Lessons Learned
Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders
Omni-Diffusion: Unified Multimodal Understanding and Generation with Masked Discrete Diffusion
Physical Simulator In-the-Loop Video Generation
Dynamic Chunking Diffusion Transformer
FlashPrefill: Instantaneous Pattern Discovery and Thresholding for Ultra-Fast Long-Context Prefilling
OVGGT: O(1) Constant-Cost Streaming Visual Geometry Transformer
WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching
Multimodal Large Language Models as Image Classifiers
What if? Emulative Simulation with World Models for Situated Reasoning
Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models
BEVLM: Distilling Semantic Knowledge from LLMs into Bird's-Eye View Representations
Thinking with Spatial Code for Physical-World Video Reasoning
SUREON: A Benchmark and Vision-Language-Model for Surgical Reasoning
Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
Gradient Flow Polarizes Softmax Outputs towards Low-Entropy Solutions
MoE Lens -- An Expert Is All You Need
Revisiting unitarity of single scalar field with non-minimal coupling
InfoGatherer: Principled Information Seeking via Evidence Retrieval and Strategic Questioning
Hierarchical Latent Action Model