FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning

• URL: http://arxiv.org/abs/2505.08054v2
• Date: Tue, 15 Jul 2025 10:03:15 GMT
• Title: FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning
• Authors: Zhehao Zhang, Weijie Xu, Fanyou Wu, Chandan K. Reddy,
• Abstract summary: FalseReject is a comprehensive resource containing 16k seemingly toxic queries accompanied by structured responses across 44 safety-related categories.<n>We propose a graph-informed adversarial multi-agent interaction framework to generate diverse and complex prompts.<n>We show that supervised finetuning with FalseReject substantially reduces unnecessary refusals without compromising overall safety or general language capabilities.
• Score: 12.467239356591238
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Safety alignment approaches in large language models (LLMs) often lead to the over-refusal of benign queries, significantly diminishing their utility in sensitive scenarios. To address this challenge, we introduce FalseReject, a comprehensive resource containing 16k seemingly toxic queries accompanied by structured responses across 44 safety-related categories. We propose a graph-informed adversarial multi-agent interaction framework to generate diverse and complex prompts, while structuring responses with explicit reasoning to aid models in accurately distinguishing safe from unsafe contexts. FalseReject includes training datasets tailored for both standard instruction-tuned models and reasoning-oriented models, as well as a human-annotated benchmark test set. Our extensive benchmarking on 29 state-of-the-art (SOTA) LLMs reveals persistent over-refusal challenges. Empirical results demonstrate that supervised finetuning with FalseReject substantially reduces unnecessary refusals without compromising overall safety or general language capabilities.

Related papers

• ROSE: Toward Reality-Oriented Safety Evaluation of Large Language Models [60.28667314609623]
  Large Language Models (LLMs) are increasingly deployed as black-box components in real-world applications.<n>We propose Reality-Oriented Safety Evaluation (ROSE), a novel framework that uses multi-objective reinforcement learning to fine-tune an adversarial LLM.
  arXiv  Detail & Related papers  (2025-06-17T10:55:17Z)
• Advancing Neural Network Verification through Hierarchical Safety Abstract Interpretation [52.626086874715284]
  We introduce a novel problem formulation called Abstract DNN-Verification, which verifies a hierarchical structure of unsafe outputs.<n>By leveraging abstract interpretation and reasoning about output reachable sets, our approach enables assessing multiple safety levels during the formal verification process.<n>Our contributions include a theoretical exploration of the relationship between our novel abstract safety formulation and existing approaches.
  arXiv  Detail & Related papers  (2025-05-08T13:29:46Z)
• Retrieval-Augmented Generation with Conflicting Evidence [57.66282463340297]
  Large language model (LLM) agents are increasingly employing retrieval-augmented generation (RAG) to improve the factuality of their responses.<n>In practice, these systems often need to handle ambiguous user queries and potentially conflicting information from multiple sources.<n>We propose RAMDocs (Retrieval with Ambiguity and Misinformation in Documents), a new dataset that simulates complex and realistic scenarios for conflicting evidence for a user query.
  arXiv  Detail & Related papers  (2025-04-17T16:46:11Z)
• Exploring LLM Reasoning Through Controlled Prompt Variations [0.9217021281095907]
  We evaluate how well state-of-the-art models maintain logical consistency and correctness when confronted with four categories of prompt perturbations.<n>Our experiments, conducted on thirteen open-source and closed-source LLMs, reveal that introducing irrelevant context within the model's context window significantly degrades performance.<n>Certain perturbations inadvertently trigger chain-of-thought-like reasoning behaviors, even without explicit prompting.
  arXiv  Detail & Related papers  (2025-04-02T20:18:50Z)
• Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-tuning [23.71517734919702]
  Vision-language models (VLMs) have made remarkable strides in generative modeling with multimodal inputs.<n>Current alignment strategies rely on supervised safety fine-tuning with curated datasets.<n>We show that supervised fine-tuning inadvertently reinforces spurious correlations between superficial textual patterns and safety responses.
  arXiv  Detail & Related papers  (2025-03-14T19:52:08Z)
• Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning [21.423429565221383]
  Large language models (LLMs) are vital for a wide range of applications yet remain susceptible to jailbreak threats.<n>We propose a novel defense strategy, Safety Chain-of-Thought (SCoT), which harnesses the enhanced textitreasoning capabilities of LLMs for proactive assessment of harmful inputs.
  arXiv  Detail & Related papers  (2025-01-31T14:45:23Z)
• Enhancing AI Safety Through the Fusion of Low Rank Adapters [7.384556630042846]
  Low-Rank Adapter Fusion mitigates harmful responses when faced with malicious prompts.<n>We show a 42% reduction in the harmfulness rate by leveraging LoRA fusion between a task adapter and a safety adapter.<n>We also observe exaggerated safety behaviour, where the model rejects safe prompts that closely resemble unsafe ones.
  arXiv  Detail & Related papers  (2024-12-30T13:12:27Z)
• Safe to Serve: Aligning Instruction-Tuned Models for Safety and Helpfulness [0.0]
  Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning and text generation.<n>LLMs can inadvertently generate unsafe or biased responses when prompted with problematic inputs.<n>This research addresses the critical challenge of developing language models that generate both helpful and harmless content.
  arXiv  Detail & Related papers  (2024-11-26T06:52:22Z)
• FaithEval: Can Your Language Model Stay Faithful to Context, Even If "The Moon is Made of Marshmallows" [74.7488607599921]
  FaithEval is a benchmark to evaluate the faithfulness of large language models (LLMs) in contextual scenarios.<n>FaithEval comprises 4.9K high-quality problems in total, validated through a rigorous four-stage context construction and validation framework.<n>Our study reveals that even state-of-the-art models often struggle to remain faithful to the given context, and that larger models do not necessarily exhibit improved faithfulness.
  arXiv  Detail & Related papers  (2024-09-30T06:27:53Z)
• SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal [64.9938658716425]
  SORRY-Bench is a proposed benchmark for evaluating large language models' (LLMs) ability to recognize and reject unsafe user requests.<n>First, existing methods often use coarse-grained taxonomy of unsafe topics, and are over-representing some fine-grained topics.<n>Second, linguistic characteristics and formatting of prompts are often overlooked, like different languages, dialects, and more -- which are only implicitly considered in many evaluations.
  arXiv  Detail & Related papers  (2024-06-20T17:56:07Z)
• On Prompt-Driven Safeguarding for Large Language Models [172.13943777203377]
  We find that in the representation space, the input queries are typically moved by safety prompts in a "higher-refusal" direction. Inspired by these findings, we propose a method for safety prompt optimization, namely DRO. Treating a safety prompt as continuous, trainable embeddings, DRO learns to move the queries' representations along or opposite the refusal direction, depending on their harmfulness.
  arXiv  Detail & Related papers  (2024-01-31T17:28:24Z)

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.