跳到正文
arXiv cs.AI· Chuyi Wang, Xiaohui Xie, Tongze Wang, Fangchen Luo, Yong Cui·· 2 小时前AI 评分51

LIDAR 论文提出通过行为指纹验证编码智能体后端模型身份

Who Is Behind the Harness? Verifying Model Identity in Coding Agents through Behavioral Fingerprinting

AI 导读

arXiv 论文提出 LIDAR,一种针对编码智能体运行的主动黑盒指纹方法,通过后编辑验证、瞬时故障恢复和规格与测试冲突三类探针对比对行为轨迹识别后端模型。

正文 · 原文

View PDF HTML (experimental)

Abstract:LLMs increasingly operate through coding-agent harnesses that inspect repositories, invoke tools, and modify files. Substituting the model behind such an agent can therefore change security-relevant decisions, including whether it verifies changes or recovers safely from failures. Existing LLM fingerprints largely infer identity from direct text or token distributions. In coding agents, these signals are mediated by system instructions, controller logic, tools, and execution feedback, limiting their transfer.
We present LIDAR (LLM Identification from Decisions and Actions at Runtime), an active black-box fingerprinting method for coding-agent execution. Three coding probe pairs expose post-edit verification, transient-failure recovery, and specification--test conflict resolution under controlled changes. LIDAR represents the resulting trajectories with complementary instance-level and distribution-level features and compares them with clean references using a lightweight probabilistic identifier. It requires no access to model weights, logits, or provider internals.
We evaluate 24 models from eight families on OpenCode, OpenHands, and mini-swe-agent, fitting separate references for each harness. Closed-set Top-1 accuracy ranges from 91.39\% to 98.89\% across the three harnesses, exceeding all three baselines in each setting. When the audit also includes unregistered models from the same families, balanced accuracy ranges from 79.58\% to 80.42\%. These results show that agent execution behavior provides model-identity evidence beyond final outputs.
Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
Cite as: arXiv:2609.28559 [cs.CR]
  (or arXiv:2609.28559v2 [cs.CR] for this version)
  https://doi.org/10.48550/arXiv.2609.28559

arXiv-issued DOI via DataCite

Submission history

From: Chuyi Wang [view email]
[v1] Wed, 23 Sep 2026 09:12:46 UTC (418 KB)
[v2] Wed, 18 Nov 2026 13:37:26 UTC (730 KB)

来源:arXiv cs.AI · arxiv.org