Keywords whose mentions rose sharply across collected items in the last 7 days.
Top-ranked papers by significance.
Repositories with the strongest weekly star growth.
News items with strong community response.
llm agent is showing slower recent signals.
e.g. CoinRAG: Contextualized Information Nugget KV Cache Reuse for Long-Context RAG
e.g. MultiGlobeQA: A Multilingual and Globally Diverse Benchmark for Geospatial Reasoning
e.g. Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations
e.g. SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent
e.g. Argus: A General-Purpose Agentic Runtime for Long-Horizon Reasoning
e.g. Beyond Modern Asymptotics for Log-Likelihood Ratios in Logistic Regression
e.g. Winning by Peeking: Unenforced Budgets and Test-Set Selection Inflate Short-Budget AutoML Comparisons
e.g. Interaction Creates Dynamical AI Behavior Absent in Isolation
e.g. Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations
e.g. MirrorWorld: Taming Video Diffusion Models for Mirror Reflection Generation