Detoxifying Toxic Communication: A Design Science Approach to Responsible AI

Hossein Arshadi Soufiani; Henry M. Kim; Hjalmar Turesson; Syed Mohammad Arham Noman; Anav Setia

cs.CYAug 31, 2026

Detoxifying Toxic Communication: A Design Science Approach to Responsible AI

Authors: Hossein Arshadi Soufiani, Henry M. Kim, Hjalmar Turesson, Syed Mohammad Arham Noman, Anav Setia

Organizations: Schulich School of Business, York University

Abstract

Toxic language in digital workplaces such as pejoratives, sarcasm, condescension, and subtle incivility can erode trust, morale, and collaboration. Existing moderation tools primarily delete or block harmful messages, disrupting communication and offering no constructive resolution. This study adopts a Design Science Research approach to create a responsible AI artifact that detects and detoxifies toxic communication. The artifact integrates fine-tuned transformer-based classifiers (DistilBERT, DistilRoBERTa) with a generative detoxification model (mT0-XL-Detox-ORPO) that rewrites toxic text into semantically equivalent, non-offensive paraphrases. Technical evaluation demonstrates high accuracy in toxicity detection and strong semantic preservation in rewritten messages, supporting conversation continuity while reinforcing respectful discourse. The paper contributes design principles for responsible AI moderation that prioritize meaning preservation and fairness.

Explore similar work

CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification
Apr 16, 2026Yian Wang, Yuen Chen, Agam Goyal +1Detoxification Toxicity
Detoxification for LLM: From Dataset Itself
Apr 21, 2026Wei Shao, Yihang Wang, Gaoyu Zhu +4Detoxification Toxicity
OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization
Jun 19, 2026Jerry Wang, Hsin-Ling Hsu, Yi-Cheng Lai +2Toxicity Red-Teaming

Apr 16, 2026 · cs.CLJ/K move · Enter open · S save

CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification

Yian Wang, Yuen Chen, Agam Goyal, Hari Sundaram

Department of Computer Science University of Illinois, Urbana-Champaign Champaign, IL 61801

Large language models (LLMs) frequently generate toxic content, posing significant risks for safe deployment. Current mitigation strategies often degrade generation quality or require costly human annotation. We propose CAUSALDETOX, a framework that identifies and intervenes on the specific attention heads causally responsible for toxic generation. Using the Probability of Necessity and Sufficiency (PNS), we isolate a minimal set of heads that are necessary and sufficient for toxicity. We utilize these components via two complementary strategies: (1) Local Inference-Time Intervention, which constructs dynamic, input-specific steering vectors for context-aware detoxification, and (2) PNS-Guided Fine-Tuning, which permanently unlearns toxic representations. We also introduce PARATOX, a novel benchmark of aligned toxic/non-toxic sentence pairs enabling controlled counterfactual evaluation. Experiments on ToxiGen, ImplicitHate, and ParaDetox show that CAUSALDETOX achieves up to 5.34% greater toxicity reduction compared to baselines while preserving linguistic fluency, and offers a 7x speedup in head selection.

Detoxification Toxicity Harmful Content Large Language Model Safety Causal Intervention Causal Reasoning

Open paper Ask about it PDF

Abstract

Explore similar work

CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification

Detoxification for LLM: From Dataset Itself

OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization