SoftHateBench: Evaluating Moderation Models Against Reasoning-Driven, Policy-Compliant Hostility

• URL: http://arxiv.org/abs/2601.20256v1
• Date: Wed, 28 Jan 2026 05:04:18 GMT
• Title: SoftHateBench: Evaluating Moderation Models Against Reasoning-Driven, Policy-Compliant Hostility
• Abstract summary: We introduce textbftextscSoftHateBench, a generative benchmark that produces soft-hate variants while preserving the hostile standpoint.<n>The benchmark spans textbf7 sociocultural domains and textbf28 target groups, comprising textbf4,745 soft-hate instances.<n> Evaluations across encoder-based detectors, general-purpose LLMs, and safety models show a consistent drop from hard to soft tiers.
• Score: 6.293833791034378
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Online hate on social media ranges from overt slurs and threats (\emph{hard hate speech}) to \emph{soft hate speech}: discourse that appears reasonable on the surface but uses framing and value-based arguments to steer audiences toward blaming or excluding a target group. We hypothesize that current moderation systems, largely optimized for surface toxicity cues, are not robust to this reasoning-driven hostility, yet existing benchmarks do not measure this gap systematically. We introduce \textbf{\textsc{SoftHateBench}}, a generative benchmark that produces soft-hate variants while preserving the underlying hostile standpoint. To generate soft hate, we integrate the \emph{Argumentum Model of Topics} (AMT) and \emph{Relevance Theory} (RT) in a unified framework: AMT provides the backbone argument structure for rewriting an explicit hateful standpoint into a seemingly neutral discussion while preserving the stance, and RT guides generation to keep the AMT chain logically coherent. The benchmark spans \textbf{7} sociocultural domains and \textbf{28} target groups, comprising \textbf{4,745} soft-hate instances. Evaluations across encoder-based detectors, general-purpose LLMs, and safety models show a consistent drop from hard to soft tiers: systems that detect explicit hostility often fail when the same stance is conveyed through subtle, reasoning-based language. \textcolor{red}{\textbf{Disclaimer.} Contains offensive examples used solely for research.}

Related papers

• Benchmarking Factual Robustness of LLMs via Multi-conversation Persuasion [0.0]
  Red-teaming frameworks evaluate models in multi-turn dialogues where the target model retains conversation history.<n>We introduce the textbfSAST-IR (Stateful Attacker, Stateless Target - Iterative Refinement) framework.<n>We simulate a worst-case adversarial setting using textbfmulti-turn (stateless) iterations.<n>Simple, diverse attack strategies achieved a staggering textbf96% success rate, exposing severe brittleness in memory-less defense.
  arXiv  Detail & Related papers  (2026-09-15T07:46:01Z)
• Beyond the Prompt: Jailbreaking Function-Calling LLMs via Simulated Moderation Traces [32.850551109913845]
  Jailbreak attacks remain a critical threat to the safe deployment of large language models (LLMs)<n>We show that this prompt-centric paradigm overlooks a structural vulnerability in stateful, function-calling environments.<n>We exploit this architectural flaw through SMT, a black-box attack framework based on Simulated Moderation Traces.
  arXiv  Detail & Related papers  (2026-07-01T06:08:07Z)
• Towards Human-Like Interactive Speech Recognition With Agentic Correction and Semantic Evaluation [53.844308305341166]
  We propose textbfAgentic ASR, a closed-loop framework that combines a single-pass ASR front-end with semantic correction, intent routing, and reasoning-based editing.<n> Experiments on multilingual, named-entity-intensive, and code-switching benchmarks show that iterative interaction consistently reduces semantic errors.
  arXiv  Detail & Related papers  (2026-05-28T06:23:31Z)
• When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning [60.714416943748866]
  In single-stream autoregressive interfaces, the same tokens both update the model state and constitute an irreversible public commitment.<n>We introduce Side-by-Side (SxS) Interleaved Reasoning, which makes disclosure timing a controllable decision within standard autoregressive generation.
  arXiv  Detail & Related papers  (2026-05-05T02:59:58Z)
• TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning [60.68349524623048]
  Decompositional jailbreaks pose a critical threat to large language models.<n>We present TwinGate, a stateful dual-encoder defense framework.<n>We construct a comprehensive dataset of over 3.62 million instructions spanning 8,600 distinct malicious intents.
  arXiv  Detail & Related papers  (2026-04-30T13:44:01Z)
• SWAY: A Counterfactual Computational Linguistic Approach to Measuring and Mitigating Sycophancy [1.4732811715354452]
  SWAY is an unsupervised computational linguistic measure of sycophancy.<n>We develop a counterfactual prompting mechanism to identify how much a model's agreement shifts under positive or negative linguistic pressure.<n>We introduce a counterfactual mitigation strategy teaching models to consider what the answer would be if opposite assumptions were suggested.
  arXiv  Detail & Related papers  (2026-04-02T18:00:14Z)
• More Than Sum of Its Parts: Deciphering Intent Shifts in Multimodal Hate Speech Detection [49.07221734365168]
  Hate speech on social media is critical for securing cyberspace, yet relies heavily on automated detection systems.<n>We characterize semantic intent shifts where modalities interact to construct implicit hate from benign cues or neutralize toxicity through semantic inversion.<n>We propose the Asymmetric Reasoning via Courtroom Agent DEbate framework to effectively decipher these complex cues.
  arXiv  Detail & Related papers  (2026-03-22T15:46:23Z)
• LaSER: Internalizing Explicit Reasoning into Latent Space for Dense Retrieval [74.72139580745511]
  LaSER is a novel self-distillation framework that internalizes explicit reasoning into the latent space of retrievers.<n>Our method successfully combines the reasoning depth of explicit CoT pipelines with the inference efficiency of standard dense retrievers.
  arXiv  Detail & Related papers  (2026-03-02T04:11:18Z)
• CORE: Context-Robust Remasking for Diffusion Language Models [51.59514489363897]
  We propose Context-Robust Remasking (CORE), a training-free framework for inference-time revision.<n>Rather than trusting static token probabilities, CORE identifies context-brittle tokens by probing their sensitivity to targeted masked-context perturbations.<n>On LLaDA-8B-Base, CORE delivers consistent improvements across reasoning and code benchmarks, outperforming compute-matched baselines and improving MBPP by up to 9.2 percentage points.
  arXiv  Detail & Related papers  (2026-02-04T00:12:30Z)
• Defenses Against Prompt Attacks Learn Surface Heuristics [40.392588465939106]
  Large language models (LLMs) are increasingly deployed in security-sensitive applications.<n>LLMs may override intended logic when adversarial instructions appear in user queries or retrieved content.<n>Recent defenses rely on supervised fine-tuning with benign and malicious labels.
  arXiv  Detail & Related papers  (2026-01-12T04:12:48Z)
• Prototypicality Bias Reveals Blindspots in Multimodal Evaluation Metrics [25.374192139098284]
  We study prototypicality bias as a systematic failure mode in multimodal evaluation.<n>We introduce a controlled contrastive benchmark ProtoBias, spanning Animals, Objects, and Demography images.<n>Our results show that widely used metrics, including CLIPScore, PickScore, and VQA-based scores, frequently misrank these pairs.<n>We propose ProtoScore, a robust 7B- parameter metric that substantially reduces failure rates and suppresses misranking.
  arXiv  Detail & Related papers  (2026-01-08T13:49:14Z)
• FVA-RAG: Falsification-Verification Alignment for Mitigating Sycophantic Hallucinations [0.0]
  Falsification-Verification Alignment RAG (FVA-RAG) is a framework that shifts the retrieval paradigm from Inductive Verification (seeking support) to Deductive Falsification (seeking disproof)<n>We introduce a dual-verification mechanism that explicitly weighs the draft answer against this "Anti-Context"<n>Preliminary experiments on a dataset of common misconceptions demonstrate that FVA-RAG significantly improves robustness against sycophantic hallucinations compared to standard RAG baselines.
  arXiv  Detail & Related papers  (2025-12-07T21:28:42Z)
• In-Context Representation Hijacking [15.706479613839967]
  Doublespeak is an in-context representation hijacking attack against large language models.<n>We show that the substitution leads to the internal representation of the benign token converging toward that of the harmful one.<n>We show that this semantic emerges layer by layer, with benign meanings in early layers converging into harmful semantics in later ones.
  arXiv  Detail & Related papers  (2025-12-03T13:19:34Z)
• Hallucination Detection via Internal States and Structured Reasoning Consistency in Large Language Models [7.18947815679122]
  Internal State Probing and Chain-of-Thought Verification are used to detect hallucinations in large language models.<n>We develop a unified framework that bridges the gap between the two methods.<n>Our framework consistently and significantly outperforms strong baselines.
  arXiv  Detail & Related papers  (2025-10-13T15:31:21Z)
• Implicit Reasoning in Large Language Models: A Comprehensive Survey [67.53966514728383]
  Large Language Models (LLMs) have demonstrated strong generalization across a wide range of tasks.<n>Recent studies have shifted attention from explicit chain-of-thought prompting toward implicit reasoning.<n>This survey introduces a taxonomy centered on execution paradigms, shifting the focus from representational forms to computational strategies.
  arXiv  Detail & Related papers  (2025-09-02T14:16:02Z)
• The Silent Saboteur: Imperceptible Adversarial Attacks against Black-Box Retrieval-Augmented Generation Systems [101.68501850486179]
  We explore adversarial attacks against retrieval-augmented generation (RAG) systems to identify their vulnerabilities.<n>This task aims to find imperceptible perturbations that retrieve a target document, originally excluded from the initial top-$k$ candidate set.<n>We propose ReGENT, a reinforcement learning-based framework that tracks interactions between the attacker and the target RAG.
  arXiv  Detail & Related papers  (2025-05-24T08:19:25Z)
• Retrieval is Not Enough: Enhancing RAG Reasoning through Test-Time Critique and Optimization [58.390885294401066]
  Retrieval-augmented generation (RAG) has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs)<n>RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions.<n>We propose AlignRAG, a novel iterative framework grounded in Critique-Driven Alignment (CDA)<n>We introduce AlignRAG-auto, an autonomous variant that dynamically terminates refinement, removing the need to pre-specify the number of critique iterations.
  arXiv  Detail & Related papers  (2025-04-21T04:56:47Z)
• LLM-based Rewriting of Inappropriate Argumentation using Reinforcement Learning from Machine Feedback [16.57980268646285]
  This paper studies how inappropriate language in arguments can be computationally mitigated. We propose a reinforcement learning-based rewriting approach that balances content preservation and appropriateness. We evaluate different weighting schemes for the reward function in both absolute and relative human assessment studies.
  arXiv  Detail & Related papers  (2024-06-05T15:18:08Z)
• Defending Large Language Models against Jailbreak Attacks via Semantic Smoothing [107.97160023681184]
  Aligned large language models (LLMs) are vulnerable to jailbreaking attacks. We propose SEMANTICSMOOTH, a smoothing-based defense that aggregates predictions of semantically transformed copies of a given input prompt.
  arXiv  Detail & Related papers  (2024-02-25T20:36:03Z)
• Constructing Highly Inductive Contexts for Dialogue Safety through Controllable Reverse Generation [65.48908724440047]
  We propose a method called emphreverse generation to construct adversarial contexts conditioned on a given response. We test three popular pretrained dialogue models (Blender, DialoGPT, and Plato2) and find that BAD+ can largely expose their safety problems.
  arXiv  Detail & Related papers  (2022-12-04T12:23:41Z)
• Discriminator-Free Generative Adversarial Attack [87.71852388383242]
  Agenerative-based adversarial attacks can get rid of this limitation. ASymmetric Saliency-based Auto-Encoder (SSAE) generates the perturbations. The adversarial examples generated by SSAE not only make thewidely-used models collapse, but also achieves good visual quality.
  arXiv  Detail & Related papers  (2021-07-20T01:55:21Z)
• Towards Variable-Length Textual Adversarial Attacks [68.27995111870712]
  It is non-trivial to conduct textual adversarial attacks on natural language processing tasks due to the discreteness of data. In this paper, we propose variable-length textual adversarial attacks(VL-Attack) Our method can achieve $33.18$ BLEU score on IWSLT14 German-English translation, achieving an improvement of $1.47$ over the baseline model.
  arXiv  Detail & Related papers  (2021-04-16T14:37:27Z)

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.