Papers
Important papers in plain language: the takeaway, the method, and what it means for you.
Latest · Papers
EfficiencyMamba Explained: A New Paradigm for Efficient Long-Sequence Modeling Based on State Space Models
SurveysPaper: "First-Language Bias" in LLM Essay Scoring — Robust Across Prompts, Yet Scores Differ by Test-Taker's L1
Efficiency105,000 Nano-Oscillators Synchronized in 45ns: The Largest Spintronic Sync Network Scouts New AI Compute
LLMPaper VISTA: Let Tool Agents "See Their Own Context" — Self-Managed Memory Without Training
EfficiencyPaper: When Machine Unlearning Meets RAG — Simulate Forgetting by Editing the Knowledge Base, Not the Model
SurveysStudy Warns the EU's AI "Guardrails" Are Too Rigid — a "Rigidity Trap" That Loses to the US "Leash" Approach
SurveysWhat Makes a 'Good Benchmark'? An arXiv Paper Distills Five Design Principles
LLMOpenAI's Entity Disambiguation: Not Memorizing Entities, but Composing Meaning from ~100 'Types'
LLMLLMs Know Your Project, Not You: NameRank Measures Parametric Recognition
LLMTRACE: A Typed Schema for Auditable Agent Reasoning Records
SurveysIsolation as a First-Class Principle for LLM-Agent Safety: A Boundary-Centric Survey
LLMOpenAI's Interpretable ML via Teaching: Picking Examples Humans Also Understand
SurveysPreparing for malicious uses of AI
LLMRemovable Defects: The Economics and Limits of Deliberate Deficiency
EfficiencyDeepSeek Introduces SPCT: A Novel Inference Scaling Method for General Reward Models (GRMs)
LLMThe One-Word Census: Answer-Choice Conformity Across 44 Language Models
LLMMAGE: Understanding Stability-Performance Trade-offs in Multi-component Prompt Optimization
Multimodal