SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.02997 · AAAI · 2025

CALM: Curiosity-Driven Auditing for Large Language Models

Xingjun Ma, Yi Liu, Cong Wang, Chao Shen, Xiang Zheng, Longxiang Wang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Auditing Large Language Models (LLMs) is a crucial and challenging task. In this study, we focus on auditing blackbox LLMs without access to their parameters, only to the provided service. We treat this type of auditing as a black-box optimization problem where the goal is to automatically uncover input-output pairs of the target LLMs that exhibit illegal, immoral, or unsafe behaviors. For instance, we may seek a nontoxic input that the target LLM responds to with a toxic output or an input that induces the hallucinative response from the target LLM containing politically sensitive individuals. This black-box optimization is challenging due to the scarcity of feasible points, the discrete nature of the prompt space, and the large search space. To address these challenges, we propose Curiosity-Driven Auditing for Large Language Models (CALM), which uses intrinsically motivated reinforcement learning to finetune an LLM as the auditor agent to uncover potential harmful and biased input-output pairs of the target LLM. CALM successfully identifies derogatory completions involving celebrities and uncovers inputs that elicit specific names under the black-box setting. This work offers a promising direction for auditing black-box LLMs. Content warning: Please note that this paper includes examples that may be offensive.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2501.02997")
get_code_for_paper("2501.02997")
have("2501.02997")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2501.02997.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2501.02997.svg)](https://syntology.ai/paper/2501.02997)

Connect an agent — have() is free.