@inproceedings{shi-etal-2026-beyond-shadow,
title = "Beyond My Own Shadow: Collaborative Latent Prompting for Personalized Question Answering",
author = "Shi, Yinghan and
Fu, Kai and
Liu, Xinyu and
Chu, Quanyou and
Du, Ming and
Wang, Hongya and
Xiao, Yanghua and
Xu, Bo",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1684/",
pages = "35949--35965",
ISBN = "979-8-89176-462-0",
abstract = "While Large Language Models (LLMs) excel in general question answering, they typically generate uniform responses, ignoring users' diverse intents. Current Personalized Question Answering (PQA) methods face dual limitations: individual-based prompting struggles with data sparsity, while collaborative explicit prompting suffers from inaccurate user matching and attention dispersion caused by text concatenation. To tackle these challenges, we propose Collaborative Latent Prompting (CLP), a PQA framework that fuses collaborative knowledge within a dynamic latent space rather than relying on explicit textual concatenation. First, our Multi-Grained Collaborative Retrieval (MGCR) module ensures precise retrieval of relevant reference users through a three-stage refinement. Next, the Dynamic Q-Former (DQ-Former) utilizes question-modulated dynamic latent queries to extract compact personalized features from noisy texts. Finally, these features are projected into dense soft prompts to implicitly steer a frozen LLM. Experiments across three PQA domains and three backbone LLMs show consistent improvements over representative PQA baselines."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shi-etal-2026-beyond-shadow">
<titleInfo>
<title>Beyond My Own Shadow: Collaborative Latent Prompting for Personalized Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yinghan</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">Fu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinyu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Quanyou</namePart>
<namePart type="family">Chu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming</namePart>
<namePart type="family">Du</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hongya</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanghua</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>While Large Language Models (LLMs) excel in general question answering, they typically generate uniform responses, ignoring users’ diverse intents. Current Personalized Question Answering (PQA) methods face dual limitations: individual-based prompting struggles with data sparsity, while collaborative explicit prompting suffers from inaccurate user matching and attention dispersion caused by text concatenation. To tackle these challenges, we propose Collaborative Latent Prompting (CLP), a PQA framework that fuses collaborative knowledge within a dynamic latent space rather than relying on explicit textual concatenation. First, our Multi-Grained Collaborative Retrieval (MGCR) module ensures precise retrieval of relevant reference users through a three-stage refinement. Next, the Dynamic Q-Former (DQ-Former) utilizes question-modulated dynamic latent queries to extract compact personalized features from noisy texts. Finally, these features are projected into dense soft prompts to implicitly steer a frozen LLM. Experiments across three PQA domains and three backbone LLMs show consistent improvements over representative PQA baselines.</abstract>
<identifier type="citekey">shi-etal-2026-beyond-shadow</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1684/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>35949</start>
<end>35965</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Beyond My Own Shadow: Collaborative Latent Prompting for Personalized Question Answering
%A Shi, Yinghan
%A Fu, Kai
%A Liu, Xinyu
%A Chu, Quanyou
%A Du, Ming
%A Wang, Hongya
%A Xiao, Yanghua
%A Xu, Bo
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F shi-etal-2026-beyond-shadow
%X While Large Language Models (LLMs) excel in general question answering, they typically generate uniform responses, ignoring users’ diverse intents. Current Personalized Question Answering (PQA) methods face dual limitations: individual-based prompting struggles with data sparsity, while collaborative explicit prompting suffers from inaccurate user matching and attention dispersion caused by text concatenation. To tackle these challenges, we propose Collaborative Latent Prompting (CLP), a PQA framework that fuses collaborative knowledge within a dynamic latent space rather than relying on explicit textual concatenation. First, our Multi-Grained Collaborative Retrieval (MGCR) module ensures precise retrieval of relevant reference users through a three-stage refinement. Next, the Dynamic Q-Former (DQ-Former) utilizes question-modulated dynamic latent queries to extract compact personalized features from noisy texts. Finally, these features are projected into dense soft prompts to implicitly steer a frozen LLM. Experiments across three PQA domains and three backbone LLMs show consistent improvements over representative PQA baselines.
%U https://aclanthology.org/2026.findings-emnlp.1684/
%P 35949-35965
Markdown (Informal)
[Beyond My Own Shadow: Collaborative Latent Prompting for Personalized Question Answering](https://aclanthology.org/2026.findings-emnlp.1684/) (Shi et al., Findings 2026)
ACL
- Yinghan Shi, Kai Fu, Xinyu Liu, Quanyou Chu, Ming Du, Hongya Wang, Yanghua Xiao, and Bo Xu. 2026. Beyond My Own Shadow: Collaborative Latent Prompting for Personalized Question Answering. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 35949–35965, Budapest, Hungary. Association for Computational Linguistics.