MemSyco-Bench：代理記憶中的諂媚行為基準測試

摘要

記憶已成為現代基於LLM的智能體的核心基石，支援其從單輪對話助手進化為長期協作夥伴。然而，記憶並非總是帶來益處：檢索到的記憶往往會引發嚴重的諂媚問題，導致智能體過度迎合使用者，犧牲事實準確性或客觀推理。儘管此類風險日益顯現，現有的記憶基準主要評估記憶是否被正確儲存、檢索或更新，卻忽略了檢索到的記憶如何影響下游推理與決策。為填補此缺口，我們提出MemSyco-Bench，一個用於評估智能體系統中記憶誘發諂媚現象的全面基準。MemSyco-Bench衡量記憶何時應影響決策，以及如何正確使用有效的記憶。具體而言，它涵蓋五項任務，評估智能體是否能拒絕將記憶作為事實證據、尊重其適用範圍、解決記憶與客觀證據之間的衝突、追蹤記憶更新，以及利用有效記憶實現個人化。所有相關資源已整理於社群資源頁面：https://github.com/XMUDeepLIT/MemSyco-Bench。

English

Memory has emerged as a cornerstone of modern LLM-based agents, supporting their evolution from single-turn assistants to long-term collaborators. However, memory is not always beneficial: retrieved memories often induce a critical issue of sycophancy, causing agents to over-align with the user at the cost of factual accuracy or objective reasoning. Despite this emerging risk, existing memory benchmarks primarily evaluate whether memories are correctly stored, retrieved, or updated, while overlooking how retrieved memories influence downstream reasoning and decision-making. To bridge this gap, we propose MemSyco-Bench, a comprehensive benchmark for evaluating memory-induced sycophancy in agent systems. MemSyco-Bench measures when memory should influence a decision and how valid memory should be used. Specifically, it covers five tasks that assess whether agents can reject memory as factual evidence, respect its applicable scope, resolve conflicts between memory and objective evidence, track memory updates, and use valid memory for personalization. All related resources are collected for the community at https://github.com/XMUDeepLIT/MemSyco-Bench.