Blog
告别冗长 PDF，一文读懂最新顶会与核心期刊的创新价值。
There Will Be a Scientific Theory of Deep Learning
OpenGame: Open Agentic Coding for Games
Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation
MusicalSCORE Benchmark: Musical Structural and Semantic Comprehension and Reasoning Benchmark
Does Reinforcement Fine-Tuning Improve Generalization of LLM Agents? An Empirical Study
Agent-GWO: Collaborative Agents for Dynamic Prompt Optimization in Large Language Models
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
MathCritique: Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervision
Thinking with Visual Primitives
SoK: On the Role and Future of AIGC Watermarking in the Era of Gen-AI
tacl-review-assignment-10639-Original+Submission+-+first+time+submission-23995 (1)
AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery
Agentic Search in the Wild: Intents and Trajectory Dynamics from 14M+ Real Search Requests
A survey on llm-as-a-judge
Geometry of the cumulant series in diffusion MRI
LLM-based NLG Evaluation: Current Status and Challenges
Large Language and Foundation Models_ Taxonomy of Architectures, Evaluation, and Domain Applications-survey
A Survey of Model Architectures in Information Retrieval
Safety of Semaglutide
Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
A Survey on AIGC Watermarking: Challenges and Future Directions
Parameter-Efficient Sparsity for Large Language Models Fine-Tuning
MAGICAGENT TOWARDS GENERALIZED AGENT PLANNING
Q-Sparse_ All Large Language Models can be Fully Sparsely-Activated
STREET-LEVEL BUREAUCRACY AND PUBLIC POLICIES: ANALYZING EDUCATIONAL POLICY IMPLEMENTATION FROM THE PERSPECTIVE OF SCHOOLS AND TEACHERS
Guardians of Academic Integrity: Multilingual Detection of AI-Generated Essays
DeTeCtive: Detecting AI-generated Text via Multi-Level Contrastive Learning
The New England Journal of Medicine
SoK: On the Role and Future of AIGC Watermarking in the Era of Gen-AI
Human-LLM Coevolution: Evidence from Academic Writing