No One Model Catches Every Harm: Benchmarking Content Moderation Across Safety Scenarios

• URL: http://arxiv.org/abs/2608.21775v1
• Date: Sat, 22 Aug 2026 04:44:24 GMT
• Title: No One Model Catches Every Harm: Benchmarking Content Moderation Across Safety Scenarios
• Abstract summary: Large Language Models (LLMs) are increasingly deployed in real-world applications, yet they remain vulnerable to generating harmful content.<n>From adversarial jailbreaks that bypass safety filters to implicit hate that evades detection, the range of risks these models pose continues to grow.<n>We present the most comprehensive evaluation of LLM safety capabilities to date, systematically testing textbf53 models across textbf11 datasets.
• Score: 0.12277343096128711
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Large Language Models (LLMs) are increasingly deployed in real-world applications, yet they remain vulnerable to generating harmful content. From adversarial jailbreaks that bypass safety filters to implicit hate that evades detection, the range of risks these models pose continues to grow. While both specialized content moderators and general-purpose LLMs are being used as safety layers, the question of which model is best suited for which type of harmful content remains unanswered. We present the most comprehensive evaluation of LLM safety capabilities to date, systematically testing \textbf{53} models across \textbf{11} datasets that we organize into four distinct categories. Our evaluation under both prompt-only and prompt-response settings uncovers critical blind spots: large frontier models that lead on one category fall significantly behind smaller, specialized alternatives on others, and real-world conversational safety remains largely unsolved across all model families. These findings challenge the assumption that scale alone ensures safety, and provide the community with a structured framework for informed model selection.

Related papers

• SafeAtlas-VL: Beyond Binary Multimodal Safety with Large-Scale Data and Guard Models [73.04773649437375]
  Multimodal safety moderation requires distinguishing risks arising from visual content, user intent, and assistant behavior.<n>We introduce SafeAtlas-VL, a dataset of 1.5M training instances that places image-, request-, and response-level judgments on a five-level ordered scale.<n>We also construct SafeAtlas-Bench, a set of 5,000 instances for evaluating five-level predictions and continuous risk scores.
  arXiv  Detail & Related papers  (2026-08-29T07:08:01Z)
• Enhancing Safety of Large Language Models via Embedding Space Separation [24.821172423424866]
  Large language models (LLMs) have achieved impressive capabilities, yet ensuring their safety against harmful prompts remains a critical challenge.<n>Recent work has revealed that the latent representations (embeddings) of harmful and safe queries in LLMs typically exhibit linear separability.<n>Motivated by this observation, we propose a representation-level fine-tuning approach, named Embedding Space Separation (ES2)
  arXiv  Detail & Related papers  (2026-03-01T07:22:38Z)
• OutSafe-Bench: A Benchmark for Multimodal Offensive Content Detection in Large Language Models [54.80460603255789]
  We introduce OutSafe-Bench, the first most comprehensive content safety evaluation test suite designed for the multimodal era.<n>OutSafe-Bench includes a large-scale dataset that spans four modalities, featuring over 18,000 bilingual (Chinese and English) text prompts, 4,500 images, 450 audio clips and 450 videos, all systematically annotated across nine critical content risk categories.<n>In addition to the dataset, we introduce a Multidimensional Cross Risk Score (MCRS), a novel metric designed to model and assess overlapping and correlated content risks across different categories.
  arXiv  Detail & Related papers  (2025-11-13T13:18:27Z)
• DUAL-Bench: Measuring Over-Refusal and Robustness in Vision-Language Models [59.45605332033458]
  Safety mechanisms can backfire, causing over-refusal, where models decline benign requests out of excessive caution.<n>No existing benchmark has systematically addressed over-refusal in the visual modality.<n>This setting introduces unique challenges, such as dual-use cases where an instruction is harmless, but the accompanying image contains harmful content.
  arXiv  Detail & Related papers  (2025-10-12T23:21:34Z)
• The Problem with Safety Classification is not just the Models [3.2634122554914002]
  We show how multilingual disparities exist in 5 safety classification models by considering datasets covering 18 languages.<n>We identify potential issues with the evaluation datasets, arguing that the shortcomings of current safety classifiers are not only because of the models themselves.
  arXiv  Detail & Related papers  (2025-07-29T13:09:40Z)
• HoliSafe: Holistic Safety Benchmarking and Modeling for Vision-Language Model [58.12612140992874]
  We introduce a holistic safety dataset and benchmark, textbfHoliSafe, that spans all five safe/unsafe image-text combinations.<n>We also propose a novel modular framework for enhancing VLM safety with a visual guard module (VGM) designed to assess the harmfulness of input images.<n> Experiments show that Safe-VLM with VGM, trained on our HoliSafe, achieves state-of-the-art safety performance across multiple VLM benchmarks.
  arXiv  Detail & Related papers  (2025-06-05T07:26:34Z)
• Understanding and Mitigating Overrefusal in LLMs from an Unveiling Perspective of Safety Decision Boundary [28.247658612894668]
  RASS is an automated framework for prompt generation and selection that strategically targets overrefusal prompts near the safety boundary.<n> RASS efficiently identifies and curates boundary-aligned prompts, enabling more effective and targeted mitigation of overrefusal.
  arXiv  Detail & Related papers  (2025-05-23T19:30:49Z)
• ShieldGemma: Generative AI Content Moderation Based on Gemma [49.91147965876678]
  ShieldGemma is a suite of safety content moderation models built upon Gemma2. Models provide robust, state-of-the-art predictions of safety risks across key harm types.
  arXiv  Detail & Related papers  (2024-07-31T17:48:14Z)
• Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training [67.30423823744506]
  We introduce a novel approach, Decoupled Refusal Training (DeRTa), designed to empower LLMs to refuse compliance to harmful prompts at any response position.<n>DeRTa incorporates two novel components: (1) Maximum Likelihood Estimation with Harmful Response Prefix, which trains models to recognize and avoid unsafe content by appending a segment of harmful response to the beginning of a safe response, and (2) Reinforced Transition Optimization (RTO), which equips models with the ability to transition from potential harm to safety refusal consistently throughout the harmful response sequence.
  arXiv  Detail & Related papers  (2024-07-12T09:36:33Z)
• AEGIS: Online Adaptive AI Content Safety Moderation with Ensemble of LLM Experts [0.0]
  As Large Language Models (LLMs) and generative AI become more widespread, the content safety risks associated with their use also increase. We find a notable deficiency in high-quality content safety datasets and benchmarks that comprehensively cover a wide range of critical safety areas. To address this, we define a broad content safety risk taxonomy, comprising 13 critical risk and 9 sparse risk categories.
  arXiv  Detail & Related papers  (2024-04-09T03:54:28Z)
• Developing Safe and Responsible Large Language Model : Can We Balance Bias Reduction and Language Understanding in Large Language Models? [2.089112028396727]
  This study explores whether Large Language Models can produce safe, unbiased outputs without sacrificing knowledge or comprehension.<n>We introduce the Safe and Responsible Large Language Model (textbfSR$_textLLM$)<n>Experiments on our specialized dataset and out-of-distribution test sets reveal that textbfSR$_textLLM$ effectively reduces biases while preserving knowledge integrity.
  arXiv  Detail & Related papers  (2024-04-01T18:10:05Z)

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.