π€ AI Agent12 items
Latest AI agent research papers, analysis, and improvement insights
AgentDrift: A Step-Labeled Benchmark of Injection-Hijacked LLM Agent Trajectories
LLM μμ΄μ νΈμ κ°μ ν둬ννΈ μ£Όμ 곡격μ λ¨κ³λ³λ‘ μΈ‘μ νλ λ²€μΉλ§ν¬
BenchShield: Formal Model-Backed Instrumentation for Reward Integrity in LLM-Agent Evaluation Infrastructure
μμ΄μ νΈ λ²€μΉλ§ν¬μμ 보μ ν΄νΉμ λ°©μ§νλ νμ λͺ¨λΈ κΈ°λ° νκ° μΈνλΌ
AgentLeak: Cloning Stronger LLM Agent Capabilities onto Weaker Agents Beyond Skill Stealing
λΈλλ°μ€ μνΈμμ©μΌλ‘ κ°λ ₯ν μμ΄μ νΈμ μ묡μ μ§μμ μ½ν μμ΄μ νΈλ‘ μ΄μ
AURA-Eval: Evaluation Framework for Acting Under Risk Awareness in LLM Agent Trajectories
μμ ν μμ¬κ²°μ κ³Ό μν μΈμμ μΈλΆννμ¬ νκ°νλ μμ΄μ νΈ μμ μ± νλ μμν¬
What Does an LLM-Agent Leaderboard Rank Actually Compare?
μμ΄μ νΈ λ¦¬λ보λ μ μμ ν΅κ³μ νλΉμ±κ³Ό λΉκ΅ 쑰건 λΆμ
Personalizing LLM Agent Memory Using Biometrics
μ¬μ©μ μ μ μΈμ¦ κΈ°λ° λ©ν°μ¬μ©μ νκ²½μμ κ°μΈνλ μμ΄μ νΈ λ©λͺ¨λ¦¬ κ΄λ¦¬
Agentic Share-of-Search: A Multi-Agent AI System for Competitive Decision-Making in LLM-Mediated E-Commerce
λ©ν°μμ΄μ νΈλ‘ AI μΌν κ²½νμμ ν맀μμ κ²½μλ ₯κ³Ό κ°μμ±μ λΆμΒ·μ΅μ ν
Robust Conformal Consensus: Multi-Agent LLM-as-a-Judge Interval Evaluation with Conformal Prediction
μ¬λ¬ LLM μμ΄μ νΈ νμ¬μ νκ° κ²°κ³Όμ λν λΆνμ€μ±μ μ λννλ λ°©λ²
A Layered Analysis of Disagreement And Answer Quality in Multi-Agent LLM Debate
λ©ν°μμ΄μ νΈ ν λ‘ μ΄ μ€μ μ견 λΆμΌμΉμ λ΅λ³ κ°μ μ μ΄λ£¨λμ§ μ€μ¦ λΆμ
An LLM-based multi-agent framework for automated reliability analysis of spatially variable soils
ν μ 곡ν λλ©μΈμ μλνλ μ λ’°μ± λΆμμ μννλ λ©ν°μμ΄μ νΈ μμ€ν
The worst spam emails: iLands AI agent hustle
AI μμ΄μ νΈλ₯Ό μ μ©ν λκ·λͺ¨ μ€νΈ λ©μΌ λ°°ν¬ κ΄λ ¨ λΉν κΈ°μ¬
LRU is harder to beat than the KV-cache papers suggest
LLM 컨ν μ€νΈ μΊμ μ΅μ νμμ LRU λ©λͺ¨λ¦¬ κ΄λ¦¬ μκ³ λ¦¬μ¦μ ν¨μ¨μ± μ¬κ²μ¦