Blog
告别冗长 PDF，一文读懂最新顶会与核心期刊的创新价值。
The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook
TAO: Tolerance-Aware Optimistic Verification for Floating-Point Neural Networks
Finding the Minimal Parameter Budget for Implicit Reasoning: A Data Complexity Driven Scaling Law for Language Models
14-43-1608941895979793
Reconstructing fine-scale 3D wind fields with terrain-informed machine learning.
The Last Human-Written Paper: Agent-Native Research Artifacts
SkillHone: A Harness for Continual Agent Skill Evolution Through Persistent Decision History
MatSciBench: Benchmarking the Reasoning Ability of Large Language Models in Materials Science
Fault Diagnosis and System Maintenance Based on Large Language Models and Knowledge Graphs
DIVERGE: Diversity-Enhanced RAG for Open-Ended Information Seeking
Chimera: Diagnosing Shortcut Learning in Visual-Language Understanding
Towards World Models in Biomedical Research
Treat-and-extend vs. pro re nata regimen of ranibizumab for diabetic macular edema—a two-year matched comparative study
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
RankCLIP: Ranking-Consistent Language-Image Pretraining
Image Super-Resolution Via Iterative Refinement
Deeper or Wider: A Perspective from Optimal Generalization Error with Sobolev Loss
Benchmark-Tight Approximation Ratio of Simple Mechanism for a Unit-Demand Buyer
ArchVelo: archetypal velocity modeling for single-cell multi-omic trajectories
Semaglutide attenuates neuroinflammation in male mice
DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training
A Comprehensive Survey of Direct Preference Optimization: Datasets, Theories, Variants, and Applications
GPO: Learning from Critical Steps to Improve LLM Reasoning
Geometry of Reason: Spectral Signatures of Valid Mathematical Reasoning
Publication Trend Analysis and Synthesis via Large Language Model: A Case Study of Engineering in PNAS
Open Materials Generation with Inference-Time Reinforcement Learning
LLMpedia: A Transparent Framework to Materialize an LLM's Encyclopedic Knowledge at Scale
ReasonAlloc: Hierarchical Decoding-Time KV Cache Budget Allocation for Reasoning Models
My Chemical Harness: Evolutionary Molecular Design over Synthetic Pathways with Large Language Model Agents
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models