@inproceedings{li-etal-2026-toxitrace,
title = "{T}oxi{T}race: Gradient-Aligned Training for Explainable {C}hinese Toxicity Detection",
author = "Li, Boyang and
Shou, Hongzhe and
Liang, Yuanyuan and
Zhang, JingBin and
Zhou, Fang",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.354/",
pages = "7121--7138",
ISBN = "979-8-89176-395-1",
abstract = "Existing Chinese toxic content detection methods mainly target sentence-level classification but often fail to provide readable and contiguous toxic evidence spans. We propose ToxiTrace, an explainability-oriented method for BERT-style encoders with three components: (1) CuSA, which refines encoder-derived saliency cues into fine-grained toxic spans with lightweight LLM guidance; (2) GCLoss, a gradient-constrained objective that concentrates token-level saliency on toxic evidence while suppressing irrelevant activations; and (3) ARCL, which constructs sample-specific contrastive reasoning pairs to sharpen the semantic boundary between toxic and non-toxic content. Experiments show that ToxiTrace improves classification accuracy and toxic span extraction while preserving efficient encoder-based inference and producing more coherent, human-readable explanations. The core training code is available at https://github.com/ZhouF-ECNU/ToxiTrace."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-toxitrace">
<titleInfo>
<title>ToxiTrace: Gradient-Aligned Training for Explainable Chinese Toxicity Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Boyang</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hongzhe</namePart>
<namePart type="family">Shou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuanyuan</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">JingBin</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fang</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Existing Chinese toxic content detection methods mainly target sentence-level classification but often fail to provide readable and contiguous toxic evidence spans. We propose ToxiTrace, an explainability-oriented method for BERT-style encoders with three components: (1) CuSA, which refines encoder-derived saliency cues into fine-grained toxic spans with lightweight LLM guidance; (2) GCLoss, a gradient-constrained objective that concentrates token-level saliency on toxic evidence while suppressing irrelevant activations; and (3) ARCL, which constructs sample-specific contrastive reasoning pairs to sharpen the semantic boundary between toxic and non-toxic content. Experiments show that ToxiTrace improves classification accuracy and toxic span extraction while preserving efficient encoder-based inference and producing more coherent, human-readable explanations. The core training code is available at https://github.com/ZhouF-ECNU/ToxiTrace.</abstract>
<identifier type="citekey">li-etal-2026-toxitrace</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.354/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>7121</start>
<end>7138</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ToxiTrace: Gradient-Aligned Training for Explainable Chinese Toxicity Detection
%A Li, Boyang
%A Shou, Hongzhe
%A Liang, Yuanyuan
%A Zhang, JingBin
%A Zhou, Fang
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F li-etal-2026-toxitrace
%X Existing Chinese toxic content detection methods mainly target sentence-level classification but often fail to provide readable and contiguous toxic evidence spans. We propose ToxiTrace, an explainability-oriented method for BERT-style encoders with three components: (1) CuSA, which refines encoder-derived saliency cues into fine-grained toxic spans with lightweight LLM guidance; (2) GCLoss, a gradient-constrained objective that concentrates token-level saliency on toxic evidence while suppressing irrelevant activations; and (3) ARCL, which constructs sample-specific contrastive reasoning pairs to sharpen the semantic boundary between toxic and non-toxic content. Experiments show that ToxiTrace improves classification accuracy and toxic span extraction while preserving efficient encoder-based inference and producing more coherent, human-readable explanations. The core training code is available at https://github.com/ZhouF-ECNU/ToxiTrace.
%U https://aclanthology.org/2026.findings-acl.354/
%P 7121-7138
Markdown (Informal)
[ToxiTrace: Gradient-Aligned Training for Explainable Chinese Toxicity Detection](https://aclanthology.org/2026.findings-acl.354/) (Li et al., Findings 2026)
ACL