@inproceedings{huang-etal-2026-role,
title = "Role-Sensitive Neurons: A Neuron-Level Gain Control Mechanism for Confidence Steering",
author = "Huang, Peiwen and
Hsu, Chih-Hao and
Huang, Tzu-Hung and
Lin, Shou-De",
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.294/",
pages = "5924--5944",
ISBN = "979-8-89176-395-1",
abstract = "Role-playing prompts effectively steer Large Language Models (LLMs), yet the neural mechanism driving this behavioral shift remains unclear. In this work, we identify Role-Sensitive Neurons (RSNs){---}a sparse sub-network ({\ensuremath{\approx}} 0.5{\%} of all neurons) governing the transition from hesitation to action. Using a novel evaluation framework with explicit abstention (MMLU-E), we reveal a Confidence-Performance Decoupling: roles primarily modulate the model{'}s probabilistic ``willingness to act'' rather than its underlying knowledge representation. We demonstrate that RSNs function as a mechanistic gain control system: causal intervention on this subspace allows precise regulation of abstention behavior. Furthermore, cross-model transfer experiments confirm that these circuits are indigenous to pre-training, with Instruction Tuning (SFT) acting merely as a ``signal sharpener'' to refine latent gain dynamics. Finally, we identify a critical safety boundary: in knowledge-deficient models, amplifying RSNs induces ``unwarranted certainty,'' highlighting decisiveness as a tunable gain parameter distinct from epistemic truth."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2026-role">
<titleInfo>
<title>Role-Sensitive Neurons: A Neuron-Level Gain Control Mechanism for Confidence Steering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Peiwen</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chih-Hao</namePart>
<namePart type="family">Hsu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tzu-Hung</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shou-De</namePart>
<namePart type="family">Lin</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>Role-playing prompts effectively steer Large Language Models (LLMs), yet the neural mechanism driving this behavioral shift remains unclear. In this work, we identify Role-Sensitive Neurons (RSNs)—a sparse sub-network (\ensuremath\approx 0.5% of all neurons) governing the transition from hesitation to action. Using a novel evaluation framework with explicit abstention (MMLU-E), we reveal a Confidence-Performance Decoupling: roles primarily modulate the model’s probabilistic “willingness to act” rather than its underlying knowledge representation. We demonstrate that RSNs function as a mechanistic gain control system: causal intervention on this subspace allows precise regulation of abstention behavior. Furthermore, cross-model transfer experiments confirm that these circuits are indigenous to pre-training, with Instruction Tuning (SFT) acting merely as a “signal sharpener” to refine latent gain dynamics. Finally, we identify a critical safety boundary: in knowledge-deficient models, amplifying RSNs induces “unwarranted certainty,” highlighting decisiveness as a tunable gain parameter distinct from epistemic truth.</abstract>
<identifier type="citekey">huang-etal-2026-role</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.294/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>5924</start>
<end>5944</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Role-Sensitive Neurons: A Neuron-Level Gain Control Mechanism for Confidence Steering
%A Huang, Peiwen
%A Hsu, Chih-Hao
%A Huang, Tzu-Hung
%A Lin, Shou-De
%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 huang-etal-2026-role
%X Role-playing prompts effectively steer Large Language Models (LLMs), yet the neural mechanism driving this behavioral shift remains unclear. In this work, we identify Role-Sensitive Neurons (RSNs)—a sparse sub-network (\ensuremath\approx 0.5% of all neurons) governing the transition from hesitation to action. Using a novel evaluation framework with explicit abstention (MMLU-E), we reveal a Confidence-Performance Decoupling: roles primarily modulate the model’s probabilistic “willingness to act” rather than its underlying knowledge representation. We demonstrate that RSNs function as a mechanistic gain control system: causal intervention on this subspace allows precise regulation of abstention behavior. Furthermore, cross-model transfer experiments confirm that these circuits are indigenous to pre-training, with Instruction Tuning (SFT) acting merely as a “signal sharpener” to refine latent gain dynamics. Finally, we identify a critical safety boundary: in knowledge-deficient models, amplifying RSNs induces “unwarranted certainty,” highlighting decisiveness as a tunable gain parameter distinct from epistemic truth.
%U https://aclanthology.org/2026.findings-acl.294/
%P 5924-5944
Markdown (Informal)
[Role-Sensitive Neurons: A Neuron-Level Gain Control Mechanism for Confidence Steering](https://aclanthology.org/2026.findings-acl.294/) (Huang et al., Findings 2026)
ACL