@inproceedings{wang-etal-2026-berry,
title = "Berry-{LR}: Literature Retrieval for Automatic Scientific Survey Generation via Berrypicking Model",
author = "Wang, Yajiao and
Li, Yang and
Wang, Xiaotao and
Feng, Xun and
Liu, Yi and
Zhang, Mengting and
Lin, Xin and
Hu, Maodi and
Zhang, Zhixiong",
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.2678/",
pages = "60908--60930",
ISBN = "979-8-89176-459-0",
abstract = "Automatic scientific survey generation requires a reference set that can support broad topic coverage and citation-worthy reference selection. However, existing systems often approximate this step as one-shot relevance retrieval from a target survey topic, abstract, or generated queries, which may overemphasize papers close to the initial topic wording and miss references needed for broader survey construction. We propose Berry-LR, a literature retrieval framework that models survey reference collection as a berrypicking process. Given a target survey topic, Berry-LR performs move-based search broadening, uses high-confidence retrieved papers as anchors for multi-round retrieval, and records retrieval trajectories as graph evidence for Trajectory-aware Reference Selection. Experiments on SurGE and SurveyBench show that Berry-LR improves reference retrieval, final reference selection, and downstream reference coverage over retrieval and automatic scientific survey generation baselines. Ablations and further analyses suggest that these gains come from move-based search broadening, anchor feedback, and trajectory-aware selection, while many remaining misses are tied to section-local writing contexts."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2026-berry">
<titleInfo>
<title>Berry-LR: Literature Retrieval for Automatic Scientific Survey Generation via Berrypicking Model</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yajiao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaotao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xun</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mengting</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maodi</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhixiong</namePart>
<namePart type="family">Zhang</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>Automatic scientific survey generation requires a reference set that can support broad topic coverage and citation-worthy reference selection. However, existing systems often approximate this step as one-shot relevance retrieval from a target survey topic, abstract, or generated queries, which may overemphasize papers close to the initial topic wording and miss references needed for broader survey construction. We propose Berry-LR, a literature retrieval framework that models survey reference collection as a berrypicking process. Given a target survey topic, Berry-LR performs move-based search broadening, uses high-confidence retrieved papers as anchors for multi-round retrieval, and records retrieval trajectories as graph evidence for Trajectory-aware Reference Selection. Experiments on SurGE and SurveyBench show that Berry-LR improves reference retrieval, final reference selection, and downstream reference coverage over retrieval and automatic scientific survey generation baselines. Ablations and further analyses suggest that these gains come from move-based search broadening, anchor feedback, and trajectory-aware selection, while many remaining misses are tied to section-local writing contexts.</abstract>
<identifier type="citekey">wang-etal-2026-berry</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2678/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>60908</start>
<end>60930</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Berry-LR: Literature Retrieval for Automatic Scientific Survey Generation via Berrypicking Model
%A Wang, Yajiao
%A Li, Yang
%A Wang, Xiaotao
%A Feng, Xun
%A Liu, Yi
%A Zhang, Mengting
%A Lin, Xin
%A Hu, Maodi
%A Zhang, Zhixiong
%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 wang-etal-2026-berry
%X Automatic scientific survey generation requires a reference set that can support broad topic coverage and citation-worthy reference selection. However, existing systems often approximate this step as one-shot relevance retrieval from a target survey topic, abstract, or generated queries, which may overemphasize papers close to the initial topic wording and miss references needed for broader survey construction. We propose Berry-LR, a literature retrieval framework that models survey reference collection as a berrypicking process. Given a target survey topic, Berry-LR performs move-based search broadening, uses high-confidence retrieved papers as anchors for multi-round retrieval, and records retrieval trajectories as graph evidence for Trajectory-aware Reference Selection. Experiments on SurGE and SurveyBench show that Berry-LR improves reference retrieval, final reference selection, and downstream reference coverage over retrieval and automatic scientific survey generation baselines. Ablations and further analyses suggest that these gains come from move-based search broadening, anchor feedback, and trajectory-aware selection, while many remaining misses are tied to section-local writing contexts.
%U https://aclanthology.org/2026.emnlp-main.2678/
%P 60908-60930
Markdown (Informal)
[Berry-LR: Literature Retrieval for Automatic Scientific Survey Generation via Berrypicking Model](https://aclanthology.org/2026.emnlp-main.2678/) (Wang et al., EMNLP 2026)
ACL
- Yajiao Wang, Yang Li, Xiaotao Wang, Xun Feng, Yi Liu, Mengting Zhang, Xin Lin, Maodi Hu, and Zhixiong Zhang. 2026. Berry-LR: Literature Retrieval for Automatic Scientific Survey Generation via Berrypicking Model. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 60908–60930, Budapest, Hungary. Association for Computational Linguistics.