準確但不謙遜：評估知識衝突下 LLM 智能體的認知謙遜
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict
這項研究評估 LLM 智能體在知識衝突下的認知謙遜，發現任務準確率較高不一定代表更能辨認、處理及表達不確定性。研究以 Identify、Solve、Escalate 三個行為維度評估四個智能體，並發現智能體常在執行初期察覺衝突，卻未能在後續維持或解決衝突。模型層面的幹預可提升認知謙遜，但往往以任務準確率下降為代價。
Published on Oct 8
Authors:
,
,
,
,
,
Abstract
When retrieved evidence contradicts an agent's prior beliefs, does it revise its answer, acknowledge uncertainty, or persist with an incorrect conclusion? Existing evaluations of agentic systems focus primarily on task success, offering limited insight into how agents handle such conflicts. We propose to evaluate agents on epistemic humility (EH): the agent's willingness to recognize, act on, and communicate uncertainty during task execution. We operationalize EH through three trajectory-level behavioral dimensions: Identify, Solve, and Escalate (ISE). Through knowledge conflict, situations where the backbone language model's parametric knowledge contradicts the evidence it encounters, or where two contextual sources disagree, we evaluate two conflict settings: (1) controlled conflict and (2) naturally occurring conflict during multi-step agentic execution, each paired with matched no-conflict controls. Evaluating four agents, we find that higher task accuracy does not necessarily correspond to greater epistemic humility: some high-accuracy configurations recognize conflicts during execution but do not communicate unresolved uncertainty in their incorrect final answers. Trajectory-level analysis further reveals that agents frequently detect conflicts in early steps of execution but fail to maintain or resolve them in later steps. Finally, we show that model-level interventions can improve EH, but often at the cost of task accuracy, suggesting that epistemic humility emerges from the interaction among the backbone model, agent harness, and evaluation environment.
View arXiv page View PDF GitHub Add to collection
Get this paper in your agent:
hf papers read 2610.12360
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.12360 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.12360 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.12360 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源：HuggingFace Daily Papers（社區熱門論文） · huggingface.co