From Intrinsic Toxicity to Reception-Based Toxicity: A Contextual Framework for Prediction and Evaluation

Most toxicity detection models treat toxicity as an intrinsic property of text, overlooking the role of context in shaping its impact. In this position paper, drawing on insights from psychology, neuroscience, and computational social science, we reconceptualise toxicity as a socially emergent signal of stress. We formalise this perspective in the Contextual Stress Framework (CSF), which defines toxicity as a stress-inducing norm violation within a given context and introduces an additional dimension for toxicity detection. As one possible realisation of CSF, we introduce PONOS (Proportion Of Negative Observed Sentiments), a metric that quantifies toxicity through collective social reception rather than lexical features. We validate this approach on a novel dataset, demonstrating improved contextual sensitivity and adaptability when used alongside existing models.

• Publication year

  2025

• Venue

  Unknown venue

• Publication date

  2025-03-20

• Fields of study

  Computer Science, Environmental Science

• Identifiers
  arXiv 2503.16072
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• The TIP of the Iceberg: Revealing a Hidden Class of Task-In-Prompt Adversarial Attacks on LLMs

  2025cited by this paper
• Bottom-up influences on social norms: How observers' responses to transgressions drive norm maintenance versus change.

  2024cited by this paper
• The role of emotions in collective responses to in-group norm violations: the case of university’s sensitivity to the natural environment norm

  2024cited by this paper
• A survey on multi-lingual offensive language detection

  2024cited by this paper
• The Llama 3 Herd of Models

  2024cited by this paper
• Users' Behavioral and Emotional Response to Toxicity in Twitter Conversations

  2024cited by this paper
• No offence, Bert - I insult only humans! Multilingual sentence-level attack on toxicity detection networks

  2023cited by this paper
• Is Your Toxicity My Toxicity? Exploring the Impact of Rater Identity on Toxicity Annotation

  2022cited by this paper
• Do People Get Used to Insulting Language?

  2022cited by this paper
• Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions

  2022cited by this paper
• Handling Bias in Toxic Speech Detection: A Survey

  2022cited by this paper
• The Authors Matter: Understanding and Mitigating Implicit Bias in Deep Text Classification

  2021cited by this paper
• Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detection

  2021cited by this paper
• Context Sensitivity Estimation in Toxicity Detection

  2021cited by this paper
• Abuse is Contextual, What about NLP? The Role of Context in Abusive Language Annotation and Detection

  2021cited by this paper
• Toxic, Hateful, Offensive or Abusive? What Are We Really Classifying? An Empirical Analysis of Hate Speech Datasets

  2020cited by this paper
• The Language of Derogation and Hate: Functions, Consequences, and Reappropriation

  2020cited by this paper
• Directions in abusive language training data, a systematic review: Garbage in, garbage out

  2020cited by this paper
• Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective

  2020cited by this paper
• In Data We Trust: A Critical Analysis of Hate Speech Detection Datasets

  2020cited by this paper
• TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification

  2020cited by this paper
• Toxicity Detection: Does Context Really Matter?

  2020cited by this paper
• Hate speech

  2020cited by this paper
• Review on Psychological Stress Detection Using Biosignals

  2019cited by this paper
• The Risk of Racial Bias in Hate Speech Detection

  2019cited by this paper
• Challenges and frontiers in abusive content detection

  2019cited by this paper
• Jigsaw

  2019cited by this paper
• Legal Framework, Dataset and Annotation Schema for Socially Unacceptable Online Discourse Practices in Slovene

  2017cited by this paper
• On Calibration of Modern Neural Networks

  2017cited by this paper
• What is Hate Speech? Part 2: Family Resemblances

  2017cited by this paper
• Forbidden Words Taboo And The Censoring Of Language

  2016cited by this paper
• Regularity In Semantic Change

  2016cited by this paper
• Comment

  2015cited by this paper
• A healthy heart is not a metronome: an integrative review of the heart's anatomy and heart rate variability

  2014cited by this paper
• The Emerging Field of Human Social Genomics

  2013cited by this paper
• Objective measures, sensors and computational techniques for stress recognition and classification: A survey

  2012cited by this paper
• The Content and Context of Hate Speech: Rethinking Regulation and Responses

  2012cited by this paper
• A meta-analysis of heart rate variability and neuroimaging studies: implications for heart rate variability as a marker of stress and health.

  2012cited by this paper
• The Content and Context of Hate Speech: Is There a Case for Banning Hate Speech?

  2012cited by this paper
• Extreme speech and democracy

  2012cited by this paper
• Moral Foundations Theory: The Pragmatic Validity of Moral Pluralism

  2012cited by this paper
• Scikit-learn: Machine Learning in Python

  2011influential reference
• The free-energy principle: a unified brain theory?

  2010cited by this paper
• Pains and Pleasures of Social Life

  2009cited by this paper
• Explaining Social Behavior: More Nuts and Bolts for the Social Sciences

  2007cited by this paper
• The grammar of society: the nature and dynamics of social norms

  2005cited by this paper
• The Emotional Force of Swearwords and Taboo Words in the Speech of Multilinguals

  2004cited by this paper
• Does Rejection Hurt? An fMRI Study of Social Exclusion

  2003cited by this paper
• Style and Sociolinguistic Variation: “Style” as distinctiveness: the culture and ideology of linguistic differentiation

  2002cited by this paper
• Protective and damaging effects of stress mediators.

  1998cited by this paper
• Why Zebras Don't Get Ulcers

  1994cited by this paper
• Emotion and Adaptation

  1993cited by this paper
• A Focus Theory of Normative Conduct: A Theoretical Refinement and Reevaluation of the Role of Norms in Human Behavior

  1991cited by this paper
• Politeness: some universals in language usage

  1987cited by this paper
• The language of ideology

  1984cited by this paper
• On Using Language

  1956cited by this paper

• No citing papers are available for this paper.