Blog
No more endless PDFs. Discover the core value of the latest top-tier research in one article.
Any Resolution Any Geometry: From Multi-View To Multi-Patch
RelaxFlow: Text-Driven Amodal 3D Generation
Generalization Properties of Score-matching Diffusion Models for Intrinsically Low-dimensional Data
Interactive Benchmarks
Act-Observe-Rewrite: Multimodal Coding Agents as In-Context Policy Learners for Robot Manipulation
Think, Then Verify: A Hypothesis-Verification Multi-Agent Framework for Long Video Understanding
Observation of Superfluidity and Meissner Effect of Composite Bosons in GaAs Quantum Hall System
SAIL: Similarity-Aware Guidance and Inter-Caption Augmentation-based Learning for Weakly-Supervised Dense Video Captioning
Measuring AI R&D Automation
AI+HW 2035: Shaping the Next Decade
Distribution-Conditioned Transport
Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation
PinPoint: Evaluation of Composed Image Retrieval with Explicit Negatives, Multi-Image Queries, and Paraphrase Testing
Reward-Conditioned Reinforcement Learning
IntroductionDMD-augmented Unpaired Neural Schrödinger Bridge for Ultra-Low Field MRI Enhancement
POET-X: Memory-efficient LLM Training by Scaling Orthogonal Transformation
Mario: Multimodal Graph Reasoning with Large Language Models
$\nabla$-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Latent Space
Bidirectional Curriculum Generation: A Multi-Agent Framework for Data-Efficient Mathematical Reasoning
Probabilistic Dreaming for World Models
SurvHTE-Bench: A Benchmark for Heterogeneous Treatment Effect Estimation in Survival Analysis
The Bayesian view of DESI DR2: Evidence and tension in a combined analysis with CMB and supernovae across cosmological models
Discovering mathematical concepts through a multi-agent system
The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology
WebChain: A Large-Scale Human-Annotated Dataset of Real-World Web Interaction Traces
MASQuant: Modality-Aware Smoothing Quantization for Multimodal Large Language Models
3D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understanding
EvoTool: Self-Evolving Tool-Use Policy Optimization in LLM Agents via Blame-Aware Mutation and Diversity-Aware Selection
Latent-Mark: An Audio Watermark Robust to Neural Resynthesis
Observational Indistinguishability and the Beginning of the Universe