@inproceedings{liu-etal-2026-accelerating,
title = "Accelerating Neural Architecture Search with Code Semantics",
author = "Liu, Zhen and
Zhou, Wanqi and
Bai, Shuanghao and
Song, Wei and
Zhang, He and
Liu, Yuhan and
Fu, Jingwen",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.557/",
pages = "12778--12795",
ISBN = "979-8-89176-459-0",
abstract = "Neural architecture search (NAS) is a core technique for automating AI model design, yet its efficiency is limited by the high cost of reliable architecture evaluation. To address this bottleneck, we introduce code-semantic transition modeling for NAS acceleration: before spending evaluator budget, we ask whether a proposal{'}s semantic change from its parent suggests a worthwhile architecture evolution. We instantiate this idea with ECO-SCE, an Evolutionary Code Semantic Change Estimator placed between the architecture generator and evaluator. Given a parent-to-proposal code transition, ECO-SCE filters likely low-gain proposals and reserves evaluator calls for proposals that are more likely to provide useful evolutionary progress. To train ECO-SCE, we construct NB101-SemPair from NAS-Bench-101 by sampling beneficial, harmful, and marginal transitions with diverse semantic and structural changes. ECO-SCE learns risk-aware transition utility from global semantic shifts, sparse semantic overlaps, and explicit structural edits, allowing it to skip likely low-gain proposals while sending promising or uncertain transitions to the evaluator. Across seven downstream benchmarks including CLRS, ECO-SCE reduces average GPU time by 70.88{\%} and improves task performance in 6 out of 7 cases when inserted zero-shot into OpenEvolve."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-accelerating">
<titleInfo>
<title>Accelerating Neural Architecture Search with Code Semantics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhen</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wanqi</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuanghao</namePart>
<namePart type="family">Bai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">He</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuhan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingwen</namePart>
<namePart type="family">Fu</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Neural architecture search (NAS) is a core technique for automating AI model design, yet its efficiency is limited by the high cost of reliable architecture evaluation. To address this bottleneck, we introduce code-semantic transition modeling for NAS acceleration: before spending evaluator budget, we ask whether a proposal’s semantic change from its parent suggests a worthwhile architecture evolution. We instantiate this idea with ECO-SCE, an Evolutionary Code Semantic Change Estimator placed between the architecture generator and evaluator. Given a parent-to-proposal code transition, ECO-SCE filters likely low-gain proposals and reserves evaluator calls for proposals that are more likely to provide useful evolutionary progress. To train ECO-SCE, we construct NB101-SemPair from NAS-Bench-101 by sampling beneficial, harmful, and marginal transitions with diverse semantic and structural changes. ECO-SCE learns risk-aware transition utility from global semantic shifts, sparse semantic overlaps, and explicit structural edits, allowing it to skip likely low-gain proposals while sending promising or uncertain transitions to the evaluator. Across seven downstream benchmarks including CLRS, ECO-SCE reduces average GPU time by 70.88% and improves task performance in 6 out of 7 cases when inserted zero-shot into OpenEvolve.</abstract>
<identifier type="citekey">liu-etal-2026-accelerating</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.557/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>12778</start>
<end>12795</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Accelerating Neural Architecture Search with Code Semantics
%A Liu, Zhen
%A Zhou, Wanqi
%A Bai, Shuanghao
%A Song, Wei
%A Zhang, He
%A Liu, Yuhan
%A Fu, Jingwen
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F liu-etal-2026-accelerating
%X Neural architecture search (NAS) is a core technique for automating AI model design, yet its efficiency is limited by the high cost of reliable architecture evaluation. To address this bottleneck, we introduce code-semantic transition modeling for NAS acceleration: before spending evaluator budget, we ask whether a proposal’s semantic change from its parent suggests a worthwhile architecture evolution. We instantiate this idea with ECO-SCE, an Evolutionary Code Semantic Change Estimator placed between the architecture generator and evaluator. Given a parent-to-proposal code transition, ECO-SCE filters likely low-gain proposals and reserves evaluator calls for proposals that are more likely to provide useful evolutionary progress. To train ECO-SCE, we construct NB101-SemPair from NAS-Bench-101 by sampling beneficial, harmful, and marginal transitions with diverse semantic and structural changes. ECO-SCE learns risk-aware transition utility from global semantic shifts, sparse semantic overlaps, and explicit structural edits, allowing it to skip likely low-gain proposals while sending promising or uncertain transitions to the evaluator. Across seven downstream benchmarks including CLRS, ECO-SCE reduces average GPU time by 70.88% and improves task performance in 6 out of 7 cases when inserted zero-shot into OpenEvolve.
%U https://aclanthology.org/2026.emnlp-main.557/
%P 12778-12795
Markdown (Informal)
[Accelerating Neural Architecture Search with Code Semantics](https://aclanthology.org/2026.emnlp-main.557/) (Liu et al., EMNLP 2026)
ACL
- Zhen Liu, Wanqi Zhou, Shuanghao Bai, Wei Song, He Zhang, Yuhan Liu, and Jingwen Fu. 2026. Accelerating Neural Architecture Search with Code Semantics. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 12778–12795, Budapest, Hungary. Association for Computational Linguistics.