AI News

발행 2026년 9월 3일

이 목록을 나라·회사·기술별로 요약하는 프롬프트가 현재 언어로 복사됩니다. Claude·ChatGPT 등 아무 AI에 붙여넣으세요.

2026-09-02 ~ 2026-09-03 · AI 주요 소식 80건. 제목을 누르면 원문으로 이동합니다.

날짜출처키워드제목

2026-09-03🌐 Machine Learning Street TalkAIAI Internals Are Weird — Tom McGrath
2026-09-03The Register AI/MLAIAI-assisted mushroom hunting is a recipe for a bad trip
2026-09-03🇺🇸 MarkTechPostLLMMeet Switchyard: A Rust Proxy and Library That Routes and Translates LLM Traffic Across OpenAI and Anthropic APIs
2026-09-03🇺🇸 MarkTechPostGeminiGoogle DeepMind Releases Gemini 3.8 Flash and Gemini 3.8 Flash Cyber: One Core Model, Two Access Envelopes
2026-09-03🇺🇸 Google DeepMind BlogGeminiIntroducing Gemini 3.8 Flash and 3.8 Flash Cyber
2026-09-03🇺🇸 NVIDIA Developer BlogAICo-Designing AI Models Using Speculative Decoding for Faster LLM Inference
2026-09-02🇺🇸 MIT Technology ReviewAIFacilitating AI integration with simplicity at scale
2026-09-02🇺🇸 MIT Technology ReviewAIThe Download: AI puzzles and a path to our nearest star system
2026-09-02🇰🇷 CIO KoreaagentThe agent didn’t leak anything. It just figured something out
2026-09-02🇺🇸 Analytics VidhyaLLMA Complete Guide to Decoding LLM Model Names
2026-09-02🇰🇷 CIO KoreaAIRevenue is no longer a funnel. It’s an AI learning loop
2026-09-02🇰🇷 CIO KoreaAIEngineering AI into the product development lifecycle
2026-09-02🇰🇷 CIO KoreaAIAI agents need to learn when enough is enough
2026-09-02🇺🇸 MarkTechPostAnthropicAnthropic Introduces Enterprise Frontier Safeguards (EFS): Zero-Data-Retention Privacy Plus Cross-Session Misuse Detection
2026-09-02🌐 arXiv cs.AIAIOpenAgentFlow: Enabling System-Wide Safety Boundaries for Heterogeneous AI Agent Fleets
2026-09-02🌐 arXiv cs.AIagentEULER: Exploring Underused Links with Evidence-Checked Return for Multi-Agent Mathematical Discovery
2026-09-02🌐 arXiv cs.AIAIAI Morbidity and Mortality: A Framework for Clinical AI Failure Review
2026-09-02🌐 arXiv cs.AIAIRecursive Criticality of AI Self-Improvement
2026-09-02🌐 arXiv cs.AILLMLLM-Driven Autonomous Vehicles Inherit Human Driver Biases in Pedestrian Yielding: Results and Implications From A New Benchmark
2026-09-02🌐 arXiv cs.AIAIAI Should Not Only Be Helpful. It Should Be Contingent. Artificial Intimacy, Sycophancy, and the Future of Social Learning
2026-09-02🌐 arXiv cs.AILLMLearning What to Retain: Gated-Memory Routing for Efficient Collaboration in Multi-Agent LLM Systems
2026-09-02🌐 arXiv cs.AIagentInvalidation Contracts for Cross-Episode Agent Memory
2026-09-02🌐 arXiv cs.AIAIAutoresearch for Marketplace Catalogs: From Legacy Forms to AI-Native Matching
2026-09-02🌐 arXiv cs.AIagentThe Irreversibility Budget: Fleet-Level Risk Accounting and Admission Control for Agent Operating Systems
2026-09-02🌐 arXiv cs.AIAIHuman-AI Co-Interpretation for Responsible AI: A Hermeneutic Perspective
2026-09-02🌐 arXiv cs.AIAIDr. Claw: An AI Scientist Workspace for Vibe Research
2026-09-02🌐 arXiv cs.AIAIRestoreBench: Can AI Agents Restore Power Flow Convergence?
2026-09-02🌐 arXiv cs.AIagentSpecMind: Enabling Spectrum Intelligence via Multi-Agent Hybrid Retrieval-Augmented Generation
2026-09-02🌐 arXiv cs.AIAIConversation Coach: A Voice-enabled AI System that Helps Practice Difficult Workplace Conversations
2026-09-02🌐 arXiv cs.AIagentmimeo: Compiling Public Expert Corpora into Agent Skills and Testing What Transfers
2026-09-02🌐 arXiv cs.AILLMTowards a Belief-Based World Model for LLM Agents
2026-09-02🌐 arXiv cs.AIRAGISO-RAG: Isoperimetric Noise Control for Retrieval-Augmented Generation
2026-09-02🌐 arXiv cs.AIAISocrates went Nuclear: Comparing Interaction Strategies for AI systems in a Learning Context using Brain Sensing
2026-09-02🌐 arXiv cs.AIagentControl-Data Flow Separation: Stable Prompt Optimization in Multi-Agent LLMs
2026-09-02🌐 arXiv cs.AIagentREVISE: Validity-Guided Recovery for Online Revisions in Agent Workflows
2026-09-02🌐 arXiv cs.AILLMSelf-Reports Are Not Verification: Environment-Grounded Auditing of LLM Operators in Evolutionary Search
2026-09-02🌐 arXiv cs.AILLMDrift-Aware LLM Routing with Sparse Contexts and Shared Budgets
2026-09-02🌐 arXiv cs.AIAITriple-Bottom-Line Sustainability of Language Models for Edge AI: A Comparison Between SLMs and Quantized LLMs
2026-09-02🌐 arXiv cs.AIagentChatDev 2.0: A No-Code Multi-Agent Platform for Developing Everything
2026-09-02🌐 arXiv cs.AILLMSOVER: Formal Certification of Optimization Reformulations via LLM-Assisted SMT Verification
2026-09-02🌐 arXiv cs.AIinferenceEscaping Redundant Reasoning: Structure-Aware Search for Inference-Time LLMs
2026-09-02🌐 arXiv cs.AILLMReinforcement Learning Enhanced LLM Agents for Complex Vehicle Routing Problems
2026-09-02🌐 arXiv cs.AImultimodalTowards reliable multimodal disaster severity assessment through preference optimization and explainable vision-language reasoning
2026-09-02🌐 arXiv cs.AIgenerativeDenoising Diffusion Generative Models Secretly Calculate Attentions
2026-09-02🌐 arXiv cs.AILLMRPCBench: A Benchmark for Proactive Premise Critique in LLM-based Recommendation
2026-09-02🌐 arXiv cs.AILLMSpawn Freely, Act Sparingly: Progressive Risk Vesting for Recursive LLM-Agent Trees
2026-09-02🌐 arXiv cs.AIAISpace Generative AI with Solar Energy Harvesting
2026-09-02🌐 arXiv cs.AIagentJailbreaking Text-to-Image Models Through Cracks: Navigating Heterogeneous Safety Filters via Multi-Agent Debate
2026-09-02🌐 arXiv cs.AIagentAnalog-DB: An Agent-First Analog Integrated Circuit Database, From Blocks to Systems
2026-09-02🌐 arXiv cs.AILLMLEAP: Likelihood Elicitation and Aggregation for LLM-based Probabilistic Forecasting
2026-09-02🌐 arXiv cs.AILLMEDGE: Error Dependency Graph-Guided Multi-Error Attribution in Multi-Agent LLM Systems
2026-09-02🌐 arXiv cs.AILLMWhen Guardrails Look Effective: Construct Validity Failures in LLM Agent Commerce Evaluation
2026-09-02🌐 arXiv cs.AIagentSelective Agent Guidance via Entropy: Learning Autonomous Policies from Imperfect VLM Teachers
2026-09-02🌐 arXiv cs.AILLMtrajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories
2026-09-02🌐 arXiv cs.AIagentRAPIDMap: Rapid Multi-Agent Pipeline for Interpretable Disaster Mapping from Satellite and Street-view Imagery
2026-09-02🌐 arXiv cs.AIagentGUI-CC: Benchmarking Contextual Consistency of GUI World Models as Agent Environments
2026-09-02🌐 arXiv cs.AILLMREAL-Q: E2E LLM Quantization via Dynamic Gradient Descent
2026-09-02🌐 arXiv cs.AIagentTowards Agentic Cloud Engineering: Graph and Loop Engineering with a Zero-Trust Agent Harness
2026-09-02🌐 arXiv cs.AILLMZero-Shot Respiratory Sound Classification through LLM-Augmented Audio-Text Alignment
2026-09-02🌐 arXiv cs.AIagentA Formal Analysis of Agent Payment Protocols
2026-09-02🌐 arXiv cs.AILLMRePro: Proof-Verified Benchmark Rewriting for Reliable Evaluation of LLM Mathematical Problem Solving
2026-09-02🌐 arXiv cs.AIfine-tuningAttention Sensitivity Is Not Enough: Dissociating Attention-Level and Behavioural In-Context Learning under Fine-Tuning
2026-09-02🌐 arXiv cs.AIagentScientific Agent Skills: A Library of Procedural Knowledge for Research Agents
2026-09-02🌐 arXiv cs.AImultimodalDo Multimodal LLMs See Before They Read? Diagnosing Contextual Sycophancy
2026-09-02🌐 arXiv cs.AILLMAutoXRD: Autonomous LLM Agents and Comprehensive Evaluation for Powder Diffraction Analysis
2026-09-02🌐 arXiv cs.AIfine-tuningRW-LoRA: Communication-Efficient Decentralized LoRA Fine-Tuning via Random Walks
2026-09-02🌐 arXiv cs.AILLMRetrieval, Scoring, and Decoding Shape Performance and Stability in LLM-based Conversational Recommendation
2026-09-02🌐 arXiv cs.AILLMCommit-first LLM judging inherits the judge's own errors
2026-09-02🌐 arXiv cs.AILLMDon't Let the Model Write the YAML: Deterministic, Minimal-Diff GitOps Remediation from LLM-Proposed Field Changes
2026-09-02🌐 arXiv cs.AIAICompanionSim: Synthetic Data for Evaluating Anthropomorphism in Human-AI Relationships
2026-09-02🌐 arXiv cs.AILLMDelegation Without Trust: An Empirical Gap Analysis of Identity, Authorization, and Runtime Governance in Multi-Agent LLM Systems
2026-09-02🌐 arXiv cs.AItransformerWiSDoM: Wireless Sparse Decision Transformer with Mixture-of-Experts for Multi-Task Mobile Network Optimization
2026-09-02🌐 arXiv cs.AIgenerativeGeometry-aware Latent Autoregressive Generative Model for PDEs in Complex Domains
2026-09-02🌐 arXiv cs.AIAIWorkload Identification with Physical Side Channels for AI Governance
2026-09-02🌐 arXiv cs.AIAIA Human-AI Theorem Connecting Spontaneous and Field-Induced Mechanisms of Collective Behavior in One Dimension
2026-09-02🌐 arXiv cs.AIgenerativeLatent-Space No-Arbitrage Geometry of Generative Models for Implied Volatility Surfaces
2026-09-02🌐 arXiv cs.AILLMHBQ: Hierarchical Scaling Block Quantization with Hardware-Efficiency-Aware Design for Accurate LLM Inference
2026-09-02🌐 arXiv cs.AIdeep learningHigher Structures in Deep Learning
2026-09-02🌐 arXiv cs.AIagentExploring Collaboration between a language and a non-language agent
2026-09-02🌐 arXiv cs.AILLMEvoFlint: An Evolutionary Atlas of Multi-Turn LLM Vulnerabilities