@inproceedings{wang-etal-2025-mixture,
title = "Mixture of {L}o{RA} Experts for Continual Information Extraction with {LLM}s",
author = "Wang, Zitao and
Wang, Xinyi and
Hu, Wei",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.718/",
doi = "10.18653/v1/2025.findings-emnlp.718",
pages = "13324--13339",
ISBN = "979-8-89176-335-7",
abstract = "We study continual information extraction (IE), which aims to extract emerging information across diverse IE tasks incessantly while avoiding forgetting. Existing approaches are either task-specialized for a single IE task or suffer from catastrophic forgetting and insufficient knowledge transfer in continual IE. This paper proposes a new continual IE model using token-level mixture of LoRA experts with LLMs. We leverage a LoRA router to route each token to the most relevant LoRA experts, facilitating effective knowledge transfer among IE tasks. We guide task experts' selection by task keys to retain the IE task-specific knowledge and mitigate catastrophic forgetting. We design a gate reflection method based on knowledge distillation to address forgetting in the LoRA router and task keys. The experimental results show that our model achieves state-of-the-art performance, effectively mitigating catastrophic forgetting and enhancing knowledge transfer in continual IE."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-mixture">
<titleInfo>
<title>Mixture of LoRA Experts for Continual Information Extraction with LLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zitao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinyi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>We study continual information extraction (IE), which aims to extract emerging information across diverse IE tasks incessantly while avoiding forgetting. Existing approaches are either task-specialized for a single IE task or suffer from catastrophic forgetting and insufficient knowledge transfer in continual IE. This paper proposes a new continual IE model using token-level mixture of LoRA experts with LLMs. We leverage a LoRA router to route each token to the most relevant LoRA experts, facilitating effective knowledge transfer among IE tasks. We guide task experts’ selection by task keys to retain the IE task-specific knowledge and mitigate catastrophic forgetting. We design a gate reflection method based on knowledge distillation to address forgetting in the LoRA router and task keys. The experimental results show that our model achieves state-of-the-art performance, effectively mitigating catastrophic forgetting and enhancing knowledge transfer in continual IE.</abstract>
<identifier type="citekey">wang-etal-2025-mixture</identifier>
<identifier type="doi">10.18653/v1/2025.findings-emnlp.718</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.718/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>13324</start>
<end>13339</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Mixture of LoRA Experts for Continual Information Extraction with LLMs
%A Wang, Zitao
%A Wang, Xinyi
%A Hu, Wei
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F wang-etal-2025-mixture
%X We study continual information extraction (IE), which aims to extract emerging information across diverse IE tasks incessantly while avoiding forgetting. Existing approaches are either task-specialized for a single IE task or suffer from catastrophic forgetting and insufficient knowledge transfer in continual IE. This paper proposes a new continual IE model using token-level mixture of LoRA experts with LLMs. We leverage a LoRA router to route each token to the most relevant LoRA experts, facilitating effective knowledge transfer among IE tasks. We guide task experts’ selection by task keys to retain the IE task-specific knowledge and mitigate catastrophic forgetting. We design a gate reflection method based on knowledge distillation to address forgetting in the LoRA router and task keys. The experimental results show that our model achieves state-of-the-art performance, effectively mitigating catastrophic forgetting and enhancing knowledge transfer in continual IE.
%R 10.18653/v1/2025.findings-emnlp.718
%U https://aclanthology.org/2025.findings-emnlp.718/
%U https://doi.org/10.18653/v1/2025.findings-emnlp.718
%P 13324-13339
Markdown (Informal)
[Mixture of LoRA Experts for Continual Information Extraction with LLMs](https://aclanthology.org/2025.findings-emnlp.718/) (Wang et al., Findings 2025)
ACL