ORFuzz: Fuzzing the "Other Side" of LLM Safety -- Testing Over-Refusal

• URL: http://arxiv.org/abs/2508.11222v1
• Date: Fri, 15 Aug 2025 05:03:26 GMT
• Title: ORFuzz: Fuzzing the "Other Side" of LLM Safety -- Testing Over-Refusal
• Authors: Haonan Zhang, Dongxia Wang, Yi Liu, Kexin Chen, Jiashui Wang, Xinlei Ying, Long Liu, Wenhai Wang,
• Abstract summary: Large Language Models (LLMs) increasingly exhibit over-refusal - erroneously rejecting benign queries due to overly conservative safety measures.<n>This paper introduces the first evolutionary testing framework, ORFuzz, for the systematic detection and analysis of LLM over-refusals.
• Score: 27.26251627767238
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) increasingly exhibit over-refusal - erroneously rejecting benign queries due to overly conservative safety measures - a critical functional flaw that undermines their reliability and usability. Current methods for testing this behavior are demonstrably inadequate, suffering from flawed benchmarks and limited test generation capabilities, as highlighted by our empirical user study. To the best of our knowledge, this paper introduces the first evolutionary testing framework, ORFuzz, for the systematic detection and analysis of LLM over-refusals. ORFuzz uniquely integrates three core components: (1) safety category-aware seed selection for comprehensive test coverage, (2) adaptive mutator optimization using reasoning LLMs to generate effective test cases, and (3) OR-Judge, a human-aligned judge model validated to accurately reflect user perception of toxicity and refusal. Our extensive evaluations demonstrate that ORFuzz generates diverse, validated over-refusal instances at a rate (6.98% average) more than double that of leading baselines, effectively uncovering vulnerabilities. Furthermore, ORFuzz's outputs form the basis of ORFuzzSet, a new benchmark of 1,855 highly transferable test cases that achieves a superior 63.56% average over-refusal rate across 10 diverse LLMs, significantly outperforming existing datasets. ORFuzz and ORFuzzSet provide a robust automated testing framework and a valuable community resource, paving the way for developing more reliable and trustworthy LLM-based software systems.

Related papers

• FormalJudge: A Neuro-Symbolic Paradigm for Agentic Oversight [21.731032636844237]
  This paper proposes a neuro-symbolic framework that employs a bidirectional Formal-of-Thought architecture.<n>We validate across three benchmarks spanning behavioral safety, multi-domain constraint adherence, and agentic upward deception detection.
  arXiv  Detail & Related papers  (2026-02-11T18:48:11Z)
• LLMEval-3: A Large-Scale Longitudinal Study on Robust and Fair Evaluation of Large Language Models [51.55869466207234]
  Existing evaluation of Large Language Models (LLMs) on static benchmarks is vulnerable to data contamination and leaderboard overfitting.<n>We introduce LLMEval-3, a framework for dynamic evaluation of LLMs.<n>LLEval-3 is built on a proprietary bank of 220k graduate-level questions, from which it dynamically samples unseen test sets for each evaluation run.
  arXiv  Detail & Related papers  (2025-08-07T14:46:30Z)
• Beyond Benchmarks: Dynamic, Automatic And Systematic Red-Teaming Agents For Trustworthy Medical Language Models [87.66870367661342]
  Large language models (LLMs) are used in AI applications in healthcare.<n>Red-teaming framework that continuously stress-test LLMs can reveal significant weaknesses in four safety-critical domains.<n>A suite of adversarial agents is applied to autonomously mutate test cases, identify/evolve unsafe-triggering strategies, and evaluate responses.<n>Our framework delivers an evolvable, scalable, and reliable safeguard for the next generation of medical AI.
  arXiv  Detail & Related papers  (2025-07-30T08:44:22Z)
• LLMs Cannot Reliably Judge (Yet?): A Comprehensive Assessment on the Robustness of LLM-as-a-Judge [44.6358611761225]
  Large Language Models (LLMs) have demonstrated remarkable intelligence across various tasks.<n>These systems are susceptible to adversarial attacks that can manipulate evaluation outcomes.<n>Existing evaluation methods adopted by LLM-based judges are often piecemeal and lack a unified framework for comprehensive assessment.
  arXiv  Detail & Related papers  (2025-06-11T06:48:57Z)
• A Framework for Creating Non-Regressive Test Cases via Branch Consistency Analysis Driven by Descriptions [9.141981611891715]
  DISTINCT is a Description-guided, branch-consistency analysis framework.<n>It transforms Large Language Model (LLM)-based generators into fault-aware test generators.<n>It achieves an average improvement of 14.64% in Compilation Success Rate (CSR) and 6.66% in Passing Rate (PR)
  arXiv  Detail & Related papers  (2025-06-09T07:05:48Z)
• Everything You Wanted to Know About LLM-based Vulnerability Detection But Were Afraid to Ask [30.819697001992154]
  Large Language Models are a promising tool for automated vulnerability detection.<n>Despite widespread adoption, a critical question remains: Are LLMs truly effective at detecting real-world vulnerabilities?<n>This paper challenges three widely held community beliefs: that LLMs are (i) unreliable, (ii) insensitive to code patches, and (iii) performance-plateaued across model scales.
  arXiv  Detail & Related papers  (2025-04-18T05:32:47Z)
• ABFS: Natural Robustness Testing for LLM-based NLP Software [8.833542944724465]
  Large Language Models (LLMs) in Natural Language Processing (NLP) software has rapidly gained traction across various domains.<n>These applications frequently exhibit robustness deficiencies, where slight perturbations in input may lead to erroneous outputs.<n>Current robustness testing methods face two main limitations: (1) low testing effectiveness, and (2) insufficient naturalness of test cases.
  arXiv  Detail & Related papers  (2025-03-03T09:02:06Z)
• The Dual-use Dilemma in LLMs: Do Empowering Ethical Capacities Make a Degraded Utility? [54.18519360412294]
  Large Language Models (LLMs) must balance between rejecting harmful requests for safety and accommodating legitimate ones for utility.<n>This paper presents a Direct Preference Optimization (DPO) based alignment framework that achieves better overall performance.<n>We analyze experimental results obtained from testing DeepSeek-R1 on our benchmark and reveal the critical ethical concerns raised by this highly acclaimed model.
  arXiv  Detail & Related papers  (2025-01-20T06:35:01Z)
• SeCodePLT: A Unified Platform for Evaluating the Security of Code GenAI [58.29510889419971]
  Existing benchmarks for evaluating the security risks and capabilities of code-generating large language models (LLMs) face several key limitations.<n>We introduce a general and scalable benchmark construction framework that begins with manually validated, high-quality seed examples and expands them via targeted mutations.<n>Applying this framework to Python, C/C++, and Java, we build SeCodePLT, a dataset of more than 5.9k samples spanning 44 CWE-based risk categories and three security capabilities.
  arXiv  Detail & Related papers  (2024-10-14T21:17:22Z)
• Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
  We reformulate open-ended generation tasks into token-level prediction tasks. We instruct an LLM to self-evaluate its answers. We benchmark a range of scoring methods based on self-evaluation.
  arXiv  Detail & Related papers  (2023-12-14T19:09:22Z)
• LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond [135.8013388183257]
  We propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits. Most LLMs struggle on SummEdits, with performance close to random chance. The best-performing model, GPT-4, is still 8% below estimated human performance.
  arXiv  Detail & Related papers  (2023-05-23T21:50:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.