Publications - Last Year

2025

Conference paper

P. Christmann and G. Weikum

“Recursive Question Understanding for Complex Question Answering over Heterogeneous Personal Data,” in Findings of the Association for Computational Linguistics (ACL 2025), Vienna, Austria, 2025.

mehr

BibTeX

@inproceedings{Christmann_ACL25,
TITLE = {Recursive Question Understanding for Complex Question Answering over Heterogeneous Personal Data},
AUTHOR = {Christmann, Philipp and Weikum, Gerhard},
LANGUAGE = {eng},
ISBN = {979-8-89176-256-5},
DOI = {10.18653/v1/2025.findings-acl.939},
PUBLISHER = {ACL},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
BOOKTITLE = {Findings of the Association for Computational Linguistics (ACL 2025)},
EDITOR = {Che, Wanxiang and Nabende, Joyce and Shutova, Ekatarina and Pilehvar, Mohammad Taher},
PAGES = {18269--18288},
ADDRESS = {Vienna, Austria},
}

Endnote

%0 Conference Proceedings
%A Christmann, Philipp
%A Weikum, Gerhard
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T Recursive Question Understanding for Complex Question Answering over
  Heterogeneous Personal Data : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0011-437A-9
%R 10.18653/v1/2025.findings-acl.939
%D 2025
%B 63rd Annual Meeting of the Association for Computational Linguistics 
%Z date of event: 2025-07-27 - 2025-08-01
%C Vienna, Austria
%B Findings of the Association for Computational Linguistics
%E Che, Wanxiang; Nabende, Joyce; Shutova, Ekatarina; Pilehvar, Mohammad Taher
%P 18269 - 18288
%I ACL
%@ 979-8-89176-256-5

Paper

P. Christmann and G. Weikum

“The ReQAP System for Question Answering over Personal Information,” 2025. [Online]. Available: https://arxiv.org/abs/2508.06880.

mehr

Abstract

Personal information is abundant on users' devices, from structured data in calendar, shopping records or fitness tools, to unstructured contents in mail and social media posts. This works presents the ReQAP system that supports users with answers for complex questions that involve filters, joins and aggregation over heterogeneous sources. The unique trait of ReQAP is that it recursively decomposes questions and incrementally builds an operator tree for execution. Both the question interpretation and the individual operators make smart use of light-weight language models, with judicious fine-tuning. The demo showcases the rich functionality for advanced user questions, and also offers detailed tracking of how the answers are computed by the operators in the execution tree. Being able to trace answers back to the underlying sources is vital for human comprehensibility and user trust in the system.

BibTeX

@online{Christmann_2508.06880,
TITLE = {The {ReQAP} System for Question Answering over Personal Information},
AUTHOR = {Christmann, Philipp and Weikum, Gerhard},
LANGUAGE = {eng},
URL = {https://arxiv.org/abs/2508.06880},
EPRINT = {2508.06880},
EPRINTTYPE = {arXiv},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
ABSTRACT = {Personal information is abundant on users' devices, from structured data in calendar, shopping records or fitness tools, to unstructured contents in mail and social media posts. This works presents the ReQAP system that supports users with answers for complex questions that involve filters, joins and aggregation over heterogeneous sources. The unique trait of ReQAP is that it recursively decomposes questions and incrementally builds an operator tree for execution. Both the question interpretation and the individual operators make smart use of light-weight language models, with judicious fine-tuning. The demo showcases the rich functionality for advanced user questions, and also offers detailed tracking of how the answers are computed by the operators in the execution tree. Being able to trace answers back to the underlying sources is vital for human comprehensibility and user trust in the system.},
}

Endnote

%0 Report
%A Christmann, Philipp
%A Weikum, Gerhard
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T The ReQAP System for Question Answering over Personal Information : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0011-9E07-4
%U https://arxiv.org/abs/2508.06880
%D 2025
%X Personal information is abundant on users' devices, from structured data in calendar, shopping records or fitness tools, to unstructured contents in mail and social media posts. This works presents the ReQAP system that supports users with answers for complex questions that involve filters, joins and aggregation over heterogeneous sources. The unique trait of ReQAP is that it recursively decomposes questions and incrementally builds an operator tree for execution. Both the question interpretation and the individual operators make smart use of light-weight language models, with judicious fine-tuning. The demo showcases the rich functionality for advanced user questions, and also offers detailed tracking of how the answers are computed by the operators in the execution tree. Being able to trace answers back to the underlying sources is vital for human comprehensibility and user trust in the system.
%K Computer Science, Computation and Language, cs.CL,Computer Science, Information Retrieval, cs.IR

Paper

A. Hogan, X. L. Dong, D. Vrandečić, and G. Weikum

“Large Language Models, Knowledge Graphs and Search Engines: A Crossroads for Answering Users’ Questions,” 2025. [Online]. Available: https://arxiv.org/abs/2501.06699.

mehr

Abstract

Much has been discussed about how Large Language Models, Knowledge Graphs and
Search Engines can be combined in a synergistic manner. A dimension largely
absent from current academic discourse is the user perspective. In particular,
there remain many open questions regarding how best to address the diverse
information needs of users, incorporating varying facets and levels of
difficulty. This paper introduces a taxonomy of user information needs, which
guides us to study the pros, cons and possible synergies of Large Language
Models, Knowledge Graphs and Search Engines. From this study, we derive a
roadmap for future research.

BibTeX

@online{Hogan_2501.06699,
TITLE = {Large Language Models, Knowledge Graphs and Search Engines: A Crossroads for Answering Users' Questions},
AUTHOR = {Hogan, Aidan and Dong, Xin Luna and Vrande{\v c}i{\'c}, Denny and Weikum, Gerhard},
LANGUAGE = {eng},
URL = {https://arxiv.org/abs/2501.06699},
EPRINT = {2501.06699},
EPRINTTYPE = {arXiv},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
ABSTRACT = {Much has been discussed about how Large Language Models, Knowledge Graphs and<br>Search Engines can be combined in a synergistic manner. A dimension largely<br>absent from current academic discourse is the user perspective. In particular,<br>there remain many open questions regarding how best to address the diverse<br>information needs of users, incorporating varying facets and levels of<br>difficulty. This paper introduces a taxonomy of user information needs, which<br>guides us to study the pros, cons and possible synergies of Large Language<br>Models, Knowledge Graphs and Search Engines. From this study, we derive a<br>roadmap for future research.<br>},
}

Endnote

%0 Report
%A Hogan, Aidan
%A Dong, Xin Luna
%A Vrande&#269;i&#263;, Denny
%A Weikum, Gerhard
%+ External Organizations
External Organizations
External Organizations
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T Large Language Models, Knowledge Graphs and Search Engines: A Crossroads for Answering Users' Questions : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0010-7837-A
%U https://arxiv.org/abs/2501.06699
%D 2025
%X   Much has been discussed about how Large Language Models, Knowledge Graphs and<br>Search Engines can be combined in a synergistic manner. A dimension largely<br>absent from current academic discourse is the user perspective. In particular,<br>there remain many open questions regarding how best to address the diverse<br>information needs of users, incorporating varying facets and levels of<br>difficulty. This paper introduces a taxonomy of user information needs, which<br>guides us to study the pros, cons and possible synergies of Large Language<br>Models, Knowledge Graphs and Search Engines. From this study, we derive a<br>roadmap for future research.<br>
%K Computer Science, Artificial Intelligence, cs.AI,Computer Science, Information Retrieval, cs.IR,Computer Science, Symbolic Computation, cs.SC

Conference paper

M. Kaiser and G. Weikum

“Preference-based Learning with Retrieval Augmented Generation for Conversational Question Answering,” in WWW Companion ’25, Sydney, Australia, 2025.

mehr

BibTeX

@inproceedings{Kaiser_WWW25,
TITLE = {Preference-based Learning with Retrieval Augmented Generation for Conversational Question Answering},
AUTHOR = {Kaiser, Magdalena and Weikum, Gerhard},
LANGUAGE = {eng},
ISBN = {979-8-4007-1331-6},
DOI = {10.1145/3701716.3715544},
PUBLISHER = {ACM},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
BOOKTITLE = {WWW Companion '25},
EDITOR = {Long, Guodong and Blumestein, Michael and Chang, Yi},
PAGES = {1053--1057},
ADDRESS = {Sydney, Australia},
}

Endnote

%0 Conference Proceedings
%A Kaiser, Magdalena
%A Weikum, Gerhard
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T Preference-based Learning with Retrieval Augmented Generation for
  Conversational Question Answering : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0010-FBD2-6
%R 10.1145/3701716.3715544
%D 2025
%B ACM Web Conference
%Z date of event: 2025-04-28 - 2025-05-02
%C Sydney, Australia
%B WWW Companion '25
%E Long, Guodong; Blumestein, Michael; Chang, Yi
%P 1053 - 1057
%I ACM
%@ 979-8-4007-1331-6

Thesis

D5IMPR-CS

T.-P. Nguyen

“Large-scale Acquisition of Refined Commonsense Knowledge,” Universität des Saarlandes, Saarbrücken, 2025.

mehr

BibTeX

@phdthesis{ThesisPhDNguyen2025,
TITLE = {Large-scale Acquisition of Refined Commonsense Knowledge},
AUTHOR = {Nguyen, Tuan-Phong},
LANGUAGE = {eng},
URL = {urn:nbn:de:bsz:291--ds-452237},
DOI = {10.22028/D291-45223},
SCHOOL = {Universit{\"a}t des Saarlandes},
ADDRESS = {Saarbr{\"u}cken},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
}

Endnote

%0 Thesis
%A Nguyen, Tuan-Phong
%Y Weikum, Gerhard
%A referee: Razniewski, Simon
%A referee: Suchanek, Fabian
%A referee: Gummadi, Krishna
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
International Max Planck Research School, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Group K. Gummadi, Max Planck Institute for Software Systems, Max Planck Society
%T Large-scale Acquisition of Refined Commonsense Knowledge : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0011-FD98-5
%R 10.22028/D291-45223
%U urn:nbn:de:bsz:291--ds-452237
%F OTHER: hdl:20.500.11880/40027
%I Universit&#228;t des Saarlandes
%C Saarbr&#252;cken
%D 2025
%P XII, 148 p.
%V phd
%9 phd
%U http://dx.doi.org/10.22028/D291-45223

Conference paper

S. Singhania, S. Razniewski, and G. Weikum

“Recall Them All: Long List Generation from Long Novels,” in Proceedings of the First Workshop NLP and Language Models for Digital Humanities associated with RANLP 2025 (LM4DH 2025), Varna, Bulgaria, 2025.

mehr

BibTeX

@inproceedings{SinghaniaLM4DH25,
TITLE = {Recall Them All: {L}ong List Generation from Long Novels},
AUTHOR = {Singhania, Sneha and Razniewski, Simon and Weikum, Gerhard},
LANGUAGE = {eng},
ISBN = {978-954-452-106-6},
DOI = {10.26615/978-954-452-106-6-013},
PUBLISHER = {ACL Bulgaria},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
BOOKTITLE = {Proceedings of the First Workshop NLP and Language Models for Digital Humanities associated with RANLP 2025 (LM4DH 2025)},
EDITOR = {Arachchige, Isuri Nanomi and Frontini, Francesca and Mitkov, Ruslan and Rayson, Paul},
PAGES = {133--142},
ADDRESS = {Varna, Bulgaria},
}

Endnote

%0 Conference Proceedings
%A Singhania, Sneha
%A Razniewski, Simon
%A Weikum, Gerhard
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T Recall Them All: Long List Generation from Long Novels : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0012-315D-D
%R 10.26615/978-954-452-106-6-013
%D 2025
%B First Workshop NLP and Language Models for Digital Humanities
%Z date of event: 2025-09-11 - 2025-09-11
%C Varna, Bulgaria
%B Proceedings of the First Workshop NLP and Language Models for Digital Humanities associated with RANLP 2025
%E Arachchige, Isuri Nanomi; Frontini, Francesca; Mitkov, Ruslan; Rayson, Paul
%P 133 - 142
%I ACL Bulgaria
%@ 978-954-452-106-6
%U https://acl-bg.org/proceedings/2025/LM4DH%202025/pdf/2025.lm4dh-1.13.pdf

Conference paper

S. Singhania, S. Razniewski, and G. Weikum

“L3X: Long Object List Extraction from Long Documents,” in CIKM ’25, 34th ACM International Conference on Information and Knowledge Management, Seoul, Republic of Korea, 2025.

mehr

BibTeX

@inproceedings{SinghaniaCIKM25,
TITLE = {{L3X}: {L}ong Object List Extraction from Long Documents},
AUTHOR = {Singhania, Sneha and Razniewski, Simon and Weikum, Gerhard},
LANGUAGE = {eng},
ISBN = {979-8-4007-2040-6},
DOI = {10.1145/3746252.3761460},
PUBLISHER = {ACM},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
BOOKTITLE = {CIKM '25, 34th ACM International Conference on Information and Knowledge Management},
EDITOR = {Cha, Meeyoung and Park, Chanyoung and Park, Noseong and Yang, Carl and Roy, Senjuti Basu and Li, Jessie and Kamps, Jaap and Shin, Kijung and Hooi, Bryan and He, Lifang},
PAGES = {6693--6697},
ADDRESS = {Seoul, Republic of Korea},
}

Endnote

%0 Conference Proceedings
%A Singhania, Sneha
%A Razniewski, Simon
%A Weikum, Gerhard
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
External Organizations
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T L3X: Long Object List Extraction from Long Documents : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0012-4948-A
%R 10.1145/3746252.3761460
%D 2025
%B 34th ACM International Conference on Information and Knowledge Management
%Z date of event: 2025-11-10 - 2025-11-14
%C Seoul, Republic of Korea
%B CIKM '25
%E Cha, Meeyoung; Park, Chanyoung; Park, Noseong; Yang, Carl; Roy, Senjuti Basu; Li, Jessie; Kamps, Jaap; Shin, Kijung ; Hooi, Bryan; He, Lifang
%P 6693 - 6697
%I ACM
%@ 979-8-4007-2040-6

Conference paper

G. H. Torbati, A. Tigunova, G. Weikum, and A. Yates

“CUP: A Framework for Resource-Efficient Review-Based Recommenders,” in Advances in Information Retrieval (ECIR 2025), Lucca, Italy, 2025.

mehr

BibTeX

@inproceedings{Torbati_ECIR25,
TITLE = {{CUP}: {A} Framework for Resource-Efficient Review-Based Recommenders},
AUTHOR = {Torbati, Ghazaleh Haratinezhad and Tigunova, Anna and Weikum, Gerhard and Yates, Andrew},
LANGUAGE = {eng},
ISBN = {978-3-031-88710-9},
DOI = {10.1007/978-3-031-88711-6_23},
PUBLISHER = {Springer},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
BOOKTITLE = {Advances in Information Retrieval (ECIR 2025)},
EDITOR = {Hauff, Claudia and Macdonald, Craig and Jannach, Dietmar and Kazai, Gabriella and Nardini, Franco Maria and Pinelli, Fabio and Silvestri, Fabrizio and Tonellotto, Nicola},
PAGES = {360--375},
SERIES = {Lecture Notes in Computer Science},
VOLUME = {15573},
ADDRESS = {Lucca, Italy},
}

Endnote

%0 Conference Proceedings
%A Torbati, Ghazaleh Haratinezhad
%A Tigunova, Anna
%A Weikum, Gerhard
%A Yates, Andrew
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
External Organizations
Databases and Information Systems, MPI for Informatics, Max Planck Society
External Organizations
%T CUP: A Framework for Resource-Efficient Review-Based Recommenders : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0011-0BBF-B
%R 10.1007/978-3-031-88711-6_23
%D 2025
%B 47th European Conference on Information Retrieval
%Z date of event: 2025-04-06 - 2025-04-10
%C Lucca, Italy
%B Advances in Information Retrieval
%E Hauff, Claudia; Macdonald, Craig; Jannach, Dietmar; Kazai, Gabriella; Nardini, Franco Maria; Pinelli, Fabio; Silvestri, Fabrizio; Tonellotto, Nicola
%P 360 - 375
%I Springer
%@ 978-3-031-88710-9
%B Lecture Notes in Computer Science
%N 15573
%U https://rdcu.be/eh6FF

Conference paper

H. D. Tran, G. Weikum, and A. Yates

“Efficient and Effective Conversational Search with Tail Entity Selection,” in Advances in Information Retrieval (ECIR 2025), Lucca, Italy, 2025.

mehr

BibTeX

@inproceedings{Tran_ECIR25,
TITLE = {Efficient and Effective Conversational Search with Tail Entity Selection},
AUTHOR = {Tran, Hai Dang and Weikum, Gerhard and Yates, Andrew},
LANGUAGE = {eng},
ISBN = {978-3-031-88713-0},
DOI = {978-3-031-88714-7_26},
PUBLISHER = {Springer},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
BOOKTITLE = {Advances in Information Retrieval (ECIR 2025)},
EDITOR = {Hauff, Claudia and Macdonald, Craig and Jannach, Dietmar and Kazai, Gabriella and Nardini, Franco Maria and Pinelli, Fabio and Silvestri, Fabrizio and Tonellotto, Nicola},
PAGES = {275--283},
SERIES = {Lecture Notes in Computer Science},
VOLUME = {15574},
ADDRESS = {Lucca, Italy},
}

Endnote

%0 Conference Proceedings
%A Tran, Hai Dang
%A Weikum, Gerhard
%A Yates, Andrew
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T Efficient and Effective Conversational Search with Tail Entity Selection : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0011-0BC4-4
%R 978-3-031-88714-7_26
%D 2025
%B 47th European Conference on Information Retrieval
%Z date of event: 2025-04-06 - 2025-04-10
%C Lucca, Italy
%B Advances in Information Retrieval
%E Hauff, Claudia; Macdonald, Craig; Jannach, Dietmar; Kazai, Gabriella; Nardini, Franco Maria; Pinelli, Fabio; Silvestri, Fabrizio; Tonellotto, Nicola
%P 275 - 283
%I Springer
%@ 978-3-031-88713-0
%B Lecture Notes in Computer Science
%N 15574
%U https://rdcu.be/eh6Lu

Thesis

D5IMPR-CS

H. D. Tran

“Incorporating Knowledge about Entities in Conversational Search,” Universität des Saarlandes, Saarbrücken, 2025.

mehr

BibTeX

@phdthesis{ThesisPhDTran2025,
TITLE = {Incorporating Knowledge about Entities in Conversational Search},
AUTHOR = {Tran, Hai Dang},
LANGUAGE = {eng},
URL = {urn:nbn:de:bsz:291--ds-460133},
DOI = {10.22028/D291-46013},
SCHOOL = {Universit{\"a}t des Saarlandes},
ADDRESS = {Saarbr{\"u}cken},
YEAR = {2025},
MARGINALMARK = {$\bullet$},
DATE = {2025},
}

Endnote

%0 Thesis
%A Tran, Hai Dang
%Y Weikum, Gerhard
%A referee: Yates, Andrew
%A referee: Berberich, Klaus
%+ Databases and Information Systems, MPI for Informatics, Max Planck Society
International Max Planck Research School, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
Databases and Information Systems, MPI for Informatics, Max Planck Society
%T Incorporating Knowledge about Entities in Conversational Search : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0011-FD68-C
%R 10.22028/D291-46013
%U urn:nbn:de:bsz:291--ds-460133
%F OTHER: hdl:20.500.11880/40471
%I Universit&#228;t des Saarlandes
%C Saarbr&#252;cken
%D 2025
%P XI, 105 p.
%V phd
%9 phd
%U http://dx.doi.org/10.22028/D291-46013