@inproceedings{teplica-etal-2025-sciurus,
title = "{SCIUR}us: Shared Circuits for Interpretable Uncertainty Representations in Language Models",
author = "Teplica, Carter and
Liu, Yixin and
Cohan, Arman and
Rudner, Tim G. J.",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.618/",
doi = "10.18653/v1/2025.naacl-long.618",
pages = "12451--12469",
ISBN = "979-8-89176-189-6",
abstract = "We investigate the mechanistic sources of uncertainty in large language models (LLMs), an area with important implications for language model reliability and trustworthiness. To do so, we conduct a series of experiments designed to identify whether the factuality of generated responses and a model{'}s uncertainty originate in separate or shared circuits in the model architecture. We approach this question by adapting the well-established mechanistic interpretability techniques of causal tracing and zero-ablation to study the effect of different circuits on LLM generations. Our experiments on eight different models and five datasets, representing tasks predominantly requiring factual recall, provide strong evidence that a model{'}s uncertainty is produced in the same parts of the network that are responsible for the factuality of generated responses."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="teplica-etal-2025-sciurus">
<titleInfo>
<title>SCIURus: Shared Circuits for Interpretable Uncertainty Representations in Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Carter</namePart>
<namePart type="family">Teplica</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yixin</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arman</namePart>
<namePart type="family">Cohan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tim</namePart>
<namePart type="given">G</namePart>
<namePart type="given">J</namePart>
<namePart type="family">Rudner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>We investigate the mechanistic sources of uncertainty in large language models (LLMs), an area with important implications for language model reliability and trustworthiness. To do so, we conduct a series of experiments designed to identify whether the factuality of generated responses and a model’s uncertainty originate in separate or shared circuits in the model architecture. We approach this question by adapting the well-established mechanistic interpretability techniques of causal tracing and zero-ablation to study the effect of different circuits on LLM generations. Our experiments on eight different models and five datasets, representing tasks predominantly requiring factual recall, provide strong evidence that a model’s uncertainty is produced in the same parts of the network that are responsible for the factuality of generated responses.</abstract>
<identifier type="citekey">teplica-etal-2025-sciurus</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.618</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.618/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>12451</start>
<end>12469</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SCIURus: Shared Circuits for Interpretable Uncertainty Representations in Language Models
%A Teplica, Carter
%A Liu, Yixin
%A Cohan, Arman
%A Rudner, Tim G. J.
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F teplica-etal-2025-sciurus
%X We investigate the mechanistic sources of uncertainty in large language models (LLMs), an area with important implications for language model reliability and trustworthiness. To do so, we conduct a series of experiments designed to identify whether the factuality of generated responses and a model’s uncertainty originate in separate or shared circuits in the model architecture. We approach this question by adapting the well-established mechanistic interpretability techniques of causal tracing and zero-ablation to study the effect of different circuits on LLM generations. Our experiments on eight different models and five datasets, representing tasks predominantly requiring factual recall, provide strong evidence that a model’s uncertainty is produced in the same parts of the network that are responsible for the factuality of generated responses.
%R 10.18653/v1/2025.naacl-long.618
%U https://aclanthology.org/2025.naacl-long.618/
%U https://doi.org/10.18653/v1/2025.naacl-long.618
%P 12451-12469
Markdown (Informal)
[SCIURus: Shared Circuits for Interpretable Uncertainty Representations in Language Models](https://aclanthology.org/2025.naacl-long.618/) (Teplica et al., NAACL 2025)
ACL