LeanGuard 論文提出經 Lean 驗證的智能體工具調用護欄
LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents
LeanGuard 是一套經驗證的執行期系統,會在工具調用送往環境前,按 Lean4 政策檢查語言模型智能體的每次調用,並可透過證明引導搜尋修復遭拒的調用。
Published on Nov 2
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
LeanGuard is a verified runtime that blocks policy-violating tool calls by language-model agents and can repair them via proof-guided search, reducing unauthorized executions and prompt-injection success without harming benign task performance.
Tool-using language-model agents often execute actions that their domain policies forbid. We present LeanGuard, a runtime that checks every proposed tool call against a policy written in Lean4 before the call reaches the environment. Policies combine state predicates with scoped past-time temporal operators over the execution history, evaluation is fail-closed, and the decision procedure is proved equivalent to a LeanLTL semantics. A denied call can be corrected in two ways. With feedback, the agent receives the violated rule and a hint; with proof-guided repair, Lean metaprogramming searches for a nearby call that preserves the user's intent and proves that the policy admits it, without the model. On Tau-Bench with five Qwen models (0.8B-27B), unguarded agents execute at least 33 calls that the Lean policy forbids, while LeanGuard executes none; with feedback, task success does not change significantly. On DTap, LeanGuard reduces direct prompt-injection success from 57.9% to 5.9%, and in telecom, where benign tasks need no writes, benign success is unchanged. Feedback reduces repeated blocked calls and raises task success for four of five models over a bare denial. On replayed denials, repair proves an admissible edit for 88 of 252 denials caused by argument values, in a median of 2,ms.
View arXiv page View PDF GitHub Add to collection
Get this paper in your agent:
hf papers read 2611.02332
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/2611.02332 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/2611.02332 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/2611.02332 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