@inproceedings{hu-etal-2026-lada,
title = "{LADA}: Enabling Adaptation of Black-Box {LLM}s to Dynamic Domain Changes at Test Time",
author = "Hu, Yihao and
Qiao, Congyu and
Yang, Junming and
Song, Mofei and
Geng, Xin and
Xu, Ning",
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.1190/",
pages = "27274--27291",
ISBN = "979-8-89176-459-0",
abstract = "Adaptation of black-box large language models (LLMs) at test time seeks to adjust LLMs to target domains in order to handle distribution shifts during testing, without access to model parameters. Most existing approaches rely on adapters trained with substantial target-domain data, which is often scarce in practical scenarios, and these adapters are tightly coupled to their training distribution, leading to degraded performance in dynamic scenarios with changing domains. To solve this, we propose LADA, a framework leveraging meta-training for continuous and rapid adaptation of black-box LLMs to unseen and changing domains. Specifically, LADA meta-trains an adapter on diverse tasks that cover multiple datasets and error combinations to learn transferable adaptation skills. At test time, the adapter needs only a few target-domain examples for lightweight adaptation and guides the LLM stepwise toward domain-appropriate reasoning via adaptive selection of reliable steps, without any updates to the LLM parameters. Experiments on datasets spanning multiple domains validate the effectiveness of the proposed approach."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hu-etal-2026-lada">
<titleInfo>
<title>LADA: Enabling Adaptation of Black-Box LLMs to Dynamic Domain Changes at Test Time</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yihao</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Congyu</namePart>
<namePart type="family">Qiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junming</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mofei</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Geng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ning</namePart>
<namePart type="family">Xu</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>Adaptation of black-box large language models (LLMs) at test time seeks to adjust LLMs to target domains in order to handle distribution shifts during testing, without access to model parameters. Most existing approaches rely on adapters trained with substantial target-domain data, which is often scarce in practical scenarios, and these adapters are tightly coupled to their training distribution, leading to degraded performance in dynamic scenarios with changing domains. To solve this, we propose LADA, a framework leveraging meta-training for continuous and rapid adaptation of black-box LLMs to unseen and changing domains. Specifically, LADA meta-trains an adapter on diverse tasks that cover multiple datasets and error combinations to learn transferable adaptation skills. At test time, the adapter needs only a few target-domain examples for lightweight adaptation and guides the LLM stepwise toward domain-appropriate reasoning via adaptive selection of reliable steps, without any updates to the LLM parameters. Experiments on datasets spanning multiple domains validate the effectiveness of the proposed approach.</abstract>
<identifier type="citekey">hu-etal-2026-lada</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1190/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>27274</start>
<end>27291</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LADA: Enabling Adaptation of Black-Box LLMs to Dynamic Domain Changes at Test Time
%A Hu, Yihao
%A Qiao, Congyu
%A Yang, Junming
%A Song, Mofei
%A Geng, Xin
%A Xu, Ning
%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 hu-etal-2026-lada
%X Adaptation of black-box large language models (LLMs) at test time seeks to adjust LLMs to target domains in order to handle distribution shifts during testing, without access to model parameters. Most existing approaches rely on adapters trained with substantial target-domain data, which is often scarce in practical scenarios, and these adapters are tightly coupled to their training distribution, leading to degraded performance in dynamic scenarios with changing domains. To solve this, we propose LADA, a framework leveraging meta-training for continuous and rapid adaptation of black-box LLMs to unseen and changing domains. Specifically, LADA meta-trains an adapter on diverse tasks that cover multiple datasets and error combinations to learn transferable adaptation skills. At test time, the adapter needs only a few target-domain examples for lightweight adaptation and guides the LLM stepwise toward domain-appropriate reasoning via adaptive selection of reliable steps, without any updates to the LLM parameters. Experiments on datasets spanning multiple domains validate the effectiveness of the proposed approach.
%U https://aclanthology.org/2026.emnlp-main.1190/
%P 27274-27291
Markdown (Informal)
[LADA: Enabling Adaptation of Black-Box LLMs to Dynamic Domain Changes at Test Time](https://aclanthology.org/2026.emnlp-main.1190/) (Hu et al., EMNLP 2026)
ACL