@inproceedings{rayhanul-masud-parvez-2026-tame,
title = "{TAME}: Token Attribution and Masking for Emergent misalignment",
author = "Rayhanul Masud, Md and
Parvez, Md Rizwan",
editor = {Aziz, Wilker and
Berant, Jonathan and
Eikema, Bryan and
Ilia, Evgenia and
de Marneffe, Marie-Catherine and
Plank, Barbara and
Shelmanov, Artem and
Swayamdipta, Swabha and
Tiedemann, J{\"o}rg and
Vazhentsev, Artem and
V{\'a}zquez, Ra{\'u}l and
Zerva, Chrysoula},
booktitle = "The Proceedings of the Third Workshop on Uncertainty-Aware {NLP}",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.uncertainlp-main.29/",
pages = "475--480",
ISBN = "979-8-89176-486-6",
abstract = "Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weights, activations, and training documents, but it remains unclear which training tokens carry the relevant fine-tuning signal. We introduce TAME (Token Attribution and Masking for Emergent misalignment), a three-stage framework: token attribution scores how strongly the fine-tuning update raises each response token{'}s likelihood, using forward passes through a released LoRA adapter; signal characterization finds patterns among high-attribution tokens; and causal validation tests them by attribution-guided loss masking. On released EM organisms and a 6,849-example medical-advice split, attribution is concentrated (top 5{\%} of tokens hold 32{\%} of the mass) and, in Llama, depleted for medical vocabulary but enriched for a register of unwarranted certainty, even after controlling for token rarity. Masking high-attribution tokens during fresh fine-tuning cuts EM 23{\texttimes} in Llama and 36{\texttimes} in Qwen, with the perplexity cost concentrated on the targeted register rather than on medical content; an equal random mask leaves EM unchanged. In Llama, the attribution pattern suggests that EM- relevant signal lies more in how confidently flawed content is expressed than in its domain vocabulary; the causal masking effect itself holds across both families."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rayhanul-masud-parvez-2026-tame">
<titleInfo>
<title>TAME: Token Attribution and Masking for Emergent misalignment</title>
</titleInfo>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="family">Rayhanul Masud</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="given">Rizwan</namePart>
<namePart type="family">Parvez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>The Proceedings of the Third Workshop on Uncertainty-Aware NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wilker</namePart>
<namePart type="family">Aziz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathan</namePart>
<namePart type="family">Berant</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bryan</namePart>
<namePart type="family">Eikema</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Evgenia</namePart>
<namePart type="family">Ilia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Plank</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Shelmanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Swabha</namePart>
<namePart type="family">Swayamdipta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Tiedemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Artem</namePart>
<namePart type="family">Vazhentsev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raúl</namePart>
<namePart type="family">Vázquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chrysoula</namePart>
<namePart type="family">Zerva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-486-6</identifier>
</relatedItem>
<abstract>Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weights, activations, and training documents, but it remains unclear which training tokens carry the relevant fine-tuning signal. We introduce TAME (Token Attribution and Masking for Emergent misalignment), a three-stage framework: token attribution scores how strongly the fine-tuning update raises each response token’s likelihood, using forward passes through a released LoRA adapter; signal characterization finds patterns among high-attribution tokens; and causal validation tests them by attribution-guided loss masking. On released EM organisms and a 6,849-example medical-advice split, attribution is concentrated (top 5% of tokens hold 32% of the mass) and, in Llama, depleted for medical vocabulary but enriched for a register of unwarranted certainty, even after controlling for token rarity. Masking high-attribution tokens during fresh fine-tuning cuts EM 23× in Llama and 36× in Qwen, with the perplexity cost concentrated on the targeted register rather than on medical content; an equal random mask leaves EM unchanged. In Llama, the attribution pattern suggests that EM- relevant signal lies more in how confidently flawed content is expressed than in its domain vocabulary; the causal masking effect itself holds across both families.</abstract>
<identifier type="citekey">rayhanul-masud-parvez-2026-tame</identifier>
<location>
<url>https://aclanthology.org/2026.uncertainlp-main.29/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>475</start>
<end>480</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TAME: Token Attribution and Masking for Emergent misalignment
%A Rayhanul Masud, Md
%A Parvez, Md Rizwan
%Y Aziz, Wilker
%Y Berant, Jonathan
%Y Eikema, Bryan
%Y Ilia, Evgenia
%Y de Marneffe, Marie-Catherine
%Y Plank, Barbara
%Y Shelmanov, Artem
%Y Swayamdipta, Swabha
%Y Tiedemann, Jörg
%Y Vazhentsev, Artem
%Y Vázquez, Raúl
%Y Zerva, Chrysoula
%S The Proceedings of the Third Workshop on Uncertainty-Aware NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-486-6
%F rayhanul-masud-parvez-2026-tame
%X Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weights, activations, and training documents, but it remains unclear which training tokens carry the relevant fine-tuning signal. We introduce TAME (Token Attribution and Masking for Emergent misalignment), a three-stage framework: token attribution scores how strongly the fine-tuning update raises each response token’s likelihood, using forward passes through a released LoRA adapter; signal characterization finds patterns among high-attribution tokens; and causal validation tests them by attribution-guided loss masking. On released EM organisms and a 6,849-example medical-advice split, attribution is concentrated (top 5% of tokens hold 32% of the mass) and, in Llama, depleted for medical vocabulary but enriched for a register of unwarranted certainty, even after controlling for token rarity. Masking high-attribution tokens during fresh fine-tuning cuts EM 23× in Llama and 36× in Qwen, with the perplexity cost concentrated on the targeted register rather than on medical content; an equal random mask leaves EM unchanged. In Llama, the attribution pattern suggests that EM- relevant signal lies more in how confidently flawed content is expressed than in its domain vocabulary; the causal masking effect itself holds across both families.
%U https://aclanthology.org/2026.uncertainlp-main.29/
%P 475-480
Markdown (Informal)
[TAME: Token Attribution and Masking for Emergent misalignment](https://aclanthology.org/2026.uncertainlp-main.29/) (Rayhanul Masud & Parvez, UncertaiNLP 2026)
ACL