#agent
5 articles
CliffCompaction：長任務程式碼 Agent 的高效能 context 自動壓縮技術
CliffCompaction is an autocompaction technique for long-horizon coding agents that reduces token costs by up to 50% and prevents context drift by strictly truncating rather than rewriting original context.
SWE-Serve：首個針對「生產級推論服務」的 AI Agent 軟體工程基準測試
SWE-Serve is a new benchmark featuring 53 real-world SGLang tasks, designed to evaluate AI agents' capability to implement complex features and achieve production correctness across the inference serving stack.
破解 MCP 生態系：新型 A2M 攻擊框架揭示 AI Agent 的語意供應鏈安全威脅
This study introduces A2M, a two-stage black-box hijacking framework, exposing critical semantic supply-chain vulnerabilities in Model Context Protocol (MCP) agents and highlighting the need for stronger security.
別撐大脈絡，改建構程式框架：Growing Harness 透過「失敗導向學習」自動生成高效能 Agent
Growing Harness is a novel paradigm that turns recurring agent control decisions into reusable, executable code via failure-guided synthesis, slashing inference costs by up to 98.6% while keeping small models highly capable.
Meta 發表 Muse Image 與 Muse Video：首款導入 Agentic 機制與推理期算力擴展的媒體生成模型
Meta launches Muse Image and previews Muse Video, pioneering agentic capabilities like tool use, self-refinement, and test-time compute scaling for media generation.