# Don't Go To Extremes: Revealing the Excessive Sensitivity and Calibration Limitations of LLMs in Implicit Hate Speech Detection

arXiv: https://arxiv.org/abs/2402.11406 · HTML: https://mlsanity.com/paper/2402.11406/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 45 references that are in the map.
- First posted on arXiv 18 Feb 2024.

## Builds on (most cited first, 18 of 18)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback](https://mlsanity.com/paper/2305.14975/) (2023)
- [Teaching Models to Express Their Uncertainty in Words](https://mlsanity.com/paper/2205.14334/) (2022)
- [A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT](https://mlsanity.com/paper/2302.11382/) (2023)
- [How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering](https://mlsanity.com/paper/2012.00955/) (2020)
- [Social Bias Frames: Reasoning about Social and Power Implications of Language](https://mlsanity.com/paper/1911.03891/) (2019)
- [From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Models](https://mlsanity.com/paper/2305.08283/) (2023)
- [Latent Hatred: A Benchmark for Understanding Implicit Hate Speech](https://mlsanity.com/paper/2109.05322/) (2021)
- [Mix-n-Match: Ensemble and Compositional Methods for Uncertainty Calibration in Deep Learning](https://mlsanity.com/paper/2003.07329/) (2020)
- [Is ChatGPT better than Human Annotators? Potential and Limitations of ChatGPT in Explaining Implicit Hate Speech](https://mlsanity.com/paper/2302.07736/) (2023)
- [Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning](https://mlsanity.com/paper/2310.03094/) (2023)
- ["HOT" ChatGPT: The promise of ChatGPT in detecting and discriminating hateful, offensive, and toxic comments on social media](https://mlsanity.com/paper/2304.10619/) (2023)
- [Reliability Check: An Analysis of GPT-3's Response to Sensitive Topics and Prompt Wording](https://mlsanity.com/paper/2306.06199/) (2023)

## Built on by (most cited first, 25 of 25)

- [MetaFaith: Faithful Natural Language Uncertainty Expression in LLMs](https://mlsanity.com/paper/2505.24858/) (2025)
- [Better to Ask in English: Evaluation of Large Language Models on English, Low-resource and Cross-Lingual Settings](https://mlsanity.com/paper/2410.13153/) (2024)
- [Is LLM an Overconfident Judge? Unveiling the Capabilities of LLMs in Detecting Offensive Language with Annotation Disagreement](https://mlsanity.com/paper/2502.06207/) (2025)
- [LLM-Based Social Simulations Require a Boundary](https://mlsanity.com/paper/2506.19806/) (2025)
- [Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM](https://mlsanity.com/paper/2508.05775/) (2025)
- [Making FETCH! Happen: Finding Emergent Dog Whistles Through Common Habitats](https://mlsanity.com/paper/2412.12072/) (2024)
- [Semi-Supervised Dialogue Abstractive Summarization via High-Quality Pseudolabel Selection](https://mlsanity.com/paper/2403.04073/) (2024)
- [On the Risk of Evidence Pollution for Malicious Social Text Detection in the Era of LLMs](https://mlsanity.com/paper/2410.12600/) (2024)
- [Enhancing LLM-based Hatred and Toxicity Detection with Meta-Toxic Knowledge Graph](https://mlsanity.com/paper/2412.15268/) (2024)
- [Evaluating Large Language Models for Detecting Antisemitism](https://mlsanity.com/paper/2509.18293/) (2025)
- [AGENT-X: Adaptive Guideline-based Expert Network for Threshold-free AI-generated teXt detection](https://mlsanity.com/paper/2505.15261/) (2025)
- [Can LLMs Use Linguistic Uncertainty Markers to Reliably Reflect Intrinsic Confidence?](https://mlsanity.com/paper/2605.28778/) (2026)
- [Who Speaks Matters: Analysing the Influence of the Speaker's Ethnicity on Hate Classification](https://mlsanity.com/paper/2410.20490/) (2024)
- [CryptoPulse: Short-Term Cryptocurrency Forecasting with Dual-Prediction and Cross-Correlated Market Indicators](https://mlsanity.com/paper/2502.19349/) (2025)
- [Specializing General-purpose LLM Embeddings for Implicit Hate Speech Detection across Datasets](https://mlsanity.com/paper/2508.20750/) (2025)
- [Confident, Calibrated, or Complicit: Safety Alignment and Ideological Bias in LLM Hate Speech Detection](https://mlsanity.com/paper/2509.00673/) (2025)
- [Fine-Tuning Large Language Models with QLoRA for Offensive Language Detection in Roman Urdu-English Code-Mixed Text](https://mlsanity.com/paper/2510.03683/) (2025)
- [Leveraging the Potential of Prompt Engineering for Hate Speech Detection in Low-Resource Languages](https://mlsanity.com/paper/2506.23930/) (2025)
- [Pragmatics beyond humans: meaning, communication, and LLMs](https://mlsanity.com/paper/2508.06167/) (2025)
- [ContiGuard: A Framework for Continual Toxicity Detection Against Evolving Evasive Perturbations](https://mlsanity.com/paper/2603.14843/) (2026)
- [Aligning Implied Statements for Implicit Hate Speech Generalizability with Context-Bounded Semi-hard Negative Mining](https://mlsanity.com/paper/2606.18852/) (2026)
- [Understanding Interpretation Difficulty in Harmful Online Communication: Insights from Cybercrime Communities](https://mlsanity.com/paper/2607.07277/) (2026)
- [Whitewashing Hate, Smearing Harmless Content: Annotator-Style Rebuttal Attacks on LLM-Based Moderation](https://mlsanity.com/paper/2608.22230/) (2026)
- [Sledgehammer or Scalpel? A Fine-grained Adaptive Framework for Implicit Hate Speech](https://mlsanity.com/paper/2608.27462/) (2026)
- [Benevolent Bias in Multi-Turn Human-Agent Dialogue](https://mlsanity.com/paper/2608.29206/) (2026)

## Abstract (arXiv)

The fairness and trustworthiness of Large Language Models (LLMs) are receiving increasing attention. Implicit hate speech, which employs indirect language to convey hateful intentions, occupies a significant portion of practice. However, the extent to which LLMs effectively address this issue remains insufficiently examined. This paper delves into the capability of LLMs to detect implicit hate speech (Classification Task) and express confidence in their responses (Calibration Task). Our evaluation meticulously considers various prompt patterns and mainstream uncertainty estimation methods. Our findings highlight that LLMs exhibit two extremes: (1) LLMs display excessive sensitivity towards groups or topics that may cause fairness issues, resulting in misclassifying benign statements as hate speech. (2) LLMs' confidence scores for each method excessively concentrate on a fixed range, remaining unchanged regardless of the dataset's complexity. Consequently, the calibration performance is heavily reliant on primary classification accuracy. These discoveries unveil new limitations of LLMs, underscoring the need for caution when optimizing models to ensure they do not veer towards extremes. This serves as a reminder to carefully consider sensitivity and confidence in the pursuit of model fairness.