כתבה
arXiv cs.LG ·
PACEvolve: הצפנה-מודעת לאינטראקציה פורצלאם
PACEvolve: Enabling Progress-Aware Consistent Evolution
אגנטים המתפתחים בעצמם, סוגדים על ידי LLMs, נועדו להתגבר על זיהום תצורה וקריסת צורה.
תקציר מקורי באנגליתarXiv:2601.10657v3 Announce Type: replace-cross Abstract: Self-evolving agents powered by Large Language Models (LLMs) have emerged as a promising direction across diverse domains, including code optimization and scientific discovery, yet their core failure modes remain underexplored. Through a comprehensive empirical study, we identify that the model's reasoning becomes anchored to the local context of current hypotheses, overemphasizing low-level details while neglecting the broader search landscape. As a result, such agents become prone to context pollution and mode collapse, repeatedly revisiting flawed hypotheses and converging on suboptimal solutions. To address this challenge, we propose Progress-Aware Consistent Evolution (PACEvolve), a systematic framework for governing agent memo
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית