KIT International Excellence Fellowship

Univ.-Prof. Adam Jatowt has won the prestigeous International Excellence Fellowship from the Karlsruhe Institute of Technology (KIT) which allows for 2-4 months' long research visit at KIT and will help to streghten the collaboration with KIT researchers.


Research stay of Nicolas at Data Science Group

Nicolas Gutehrle who is a doctoral student at the University of Franche-Comté will stay with us from 4th April, 2022 for three months as a visiting research student. Nicolas is working on topics related to digital humanities and currently investigates automatic timeline summarization methods for primary resources.


Accepted paper at SIGIR 2022

A paper on large-scale question answering dataset has been accepted to appear at the resource track of the SIGIR 2022 conference [1].

[1] Jiexin Wang, Adam Jatowt, Masatoshi Yoshikawa: ArchivalQA: A Large-scale Benchmark Dataset for Open Domain Question Answering over Historical News Collections. Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2022), ACM Press, (to appear in 2022), preprint available here


Keynote talk at the TempWeb 2022 workshop

Adam Jatowt gave a keynote talk entitled: “Temporal Question Answering in News Article Collections” and took part in panel discussion at the 12th Temporal Web Analytics Workshop (TempWeb 2022), which was held in conjunction with The Web Conference 2022 on April 25/26, Lyon, France.


Accepted paper at ACL 2022 Findings

A paper titled “: Fact-Tree Reasoning for N-ary Question Answering over Knowledge Graphs” [1] has been accepted as a findings paper at the 60th Annual Meeting of the Association for Computational Linguistics (ACL 2022) to be held in Dublin, Ireland in May.

[1] Yao Zhang, Peiyao Li, Hongru Liang, Adam Jatowt, Zhenglu Yang: Fact-Tree Reasoning for N-ary Question Answering over Knowledge Graphs. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL 2022), to appear as ACL findings paper


Vannevar Bush Best Paper Award (Best Paper Award at JCDL 2021)

Our paper titled “Estimating Contemporary Relevance of Past News” [1] has received the Vannevar Bush Best Paper Award which was awarded at the 21tst ACM/IEEE-CS Joint Conference on Digital Libraries (JCDL 2021). This award is co-sponsored by ACM SIGIR group (the ACM Special Interest Group on Information Retrieval) together with SIGWEB (the ACM Special Interest Group on Hypertext and the Web).

[1] Mari Sato, Adam Jatowt, Yijun Duan, Ricardo Campos, Masatoshi Yoshikawa: Estimating Contemporary Relevance of Past News, Proceedings of the 21tst ACM/IEEE-CS Joint Conference on Digital Libraries (JCDL 2021), ACM Press, Illinois, USA, pp. 70-79 (2021)


Interview at UIBK

Following the interview with Fabian Oswald the corresponding news report has appeared at the university newsroom (in German).


Accepted paper at CIKM 2021

A recent research work [1] in collaboration with researchers in Alibaba Group on generating multi-role dialogues has been accepted at the 30th Conference on Information and Knowledge Management (CIKM 2021). The proposed approach can have application as a support for judges in leading case debates and for enhancing customer interactions by chatbot systems.

[1] Changzhen Ji, Yating Zhang, Xiaozhong Liu, Adam Jatowt, Changlong Sun, Conghui Zhu and Tiejun Zhao: A Neural Conversation Generation Model via Equivalent Shared Memory Investigation, Proceedings of the 30th International Conference on Information and Knowledge Management (CIKM 2021), ACM Press (to appear in 2021)


Survey on Post-OCR Correction

Our survey paper [1] written with researchers from Univ. of La Rochelle on the different approaches used for correction of OCRed texts has been just published in ACM Computing Surveys journal.

[1] Thi Tuyet Hai Nguyen, Adam Jatowt, Mickaël Coustaty and Antoine Doucet: Survey of Post-OCR Processing Approaches, ACM Computing Surveys, ACM Press (2021) PDF


Edited Book: Computational Approaches to Semantic Change

A new book co-edited with Nina Tahmasebi, Lars Borin, Yang Xu and Simon Hengchen appeared on the topic of computational approaches to semantic change [1]. Semantic change — how the meanings of words change over time — has preoccupied scholars for a long time. Compared to changes in sound and grammar, semantic change is the least understood. Ever since, the study of semantic change has progressed steadily, accumulating a vast store of knowledge for over a century, encompassing many languages and language families. Big-data empirical quantitative investigations are now coming to the forefront, enabled by enormous advances in storage capability and processing power. This volume offers a survey of this exciting new direction in the computational study of semantic change, addressing both more theoretical problems — e.g., discovery of “laws of semantic change” — and practical applications, such as information retrieval in longitudinal text archives.

[1] Nina Tahmasebi, Lars Borin, Adam Jatowt, Yang Xu and Simon Hengchen: Computational Approaches to Semantic Change. Language Variation, Language Science Press (2021)


Organization of Special Session: Multimedia Datasets for Repeatable Experimentation

Multimedia Datasets for Repeatable Experimentation (MDRE 2022) will be held in conjunction with the 27th International Conference on Multimedia Modeling (MMM 2022). MDRE 2022 presents an opportunity for researchers and practitioners to submit their datasets to this track, making their work permanently available and citable via the http://mmdatasets.org forum, as well as to increase the public awareness of their considerable efforts. The special session is organized together with Cathal Gurrin (Dublin City University), Bjorn Por Jonsson (IT University of Copenhagen), Duc-Tien Dang-Nguyen (University of Bergen), Liting Zhou (Dublin City University) and Graham Healy (Dublin City University).


Accepted paper at ACL-IJCNLP 2021

Our paper on multi-timeline summarization has been accepted at the prime NLP conference (ACL-IJCNLP 2021). The paper proposes the new task of generating multiple timeline summaries and demonstrates effective solution for this using two-stage affinity propagation algorithm. This work is the result of collaboration with Kyoto University and University of La Rochelle.

[1] Yu Yi, Adam Jatowt, Antoine Doucet, Kazunari Sugiyama, Masatoshi Yoshikawa: Multi-TimeLine Summarization (MTLS): Improving Timeline Summarization by Generating Multiple Summaries. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, pp. 377-387 (ACL-IJCNLP 2021) (2021) PDF


Organization of the 2nd International Workshop on Computational Approaches to Historical Language Change 2019 (LChange 2021)

The 2nd International Workshop on Computational Approaches to Historical Language Change 2019 ( LChange 2021) was organized together with Nina Tahmasebi (University of Gothenburg), Yang Xu (University of Toronto), Simon Hengchen (University of Gothenburg) Syrielle Montariol (INRIA Paris) and Haim Dubossarsky (University of Cambridge) and was held in conjunction with ACL 2021. The proceedings from the workshop are available at ACL Anthology.


Organization of the 6th International Workshop on Computational History (HistoInformatics 2021)

The 6th International Workshop on Computational History ( HistoInformatics 2021) is organized together with Yasunobu Sumikawa (Takushoku University), Ryohei Ikejiri (The University of Tokyo), Antoine Doucet (University of La Rochelle), Eva Pfanzelter (University of Innsbruck), Mohammed Hasanuzzaman (Munster Technological University), Gaël Dias (University of Caen Normandy, GREYC CNRS) and Ian Milligan (University of Waterloo). The workshop will take place in September in conjunction with JCDL 2021 conference.


Organization of the 4th workshop on Narrative Extraction from Text (Text2Story 2021)

The 4th workshop on Narrative Extraction from Text (Text2Story 2021) was held in conjunction with ECIR 2021 conference and was organized together with Ricardo Campos (INESC TEC; 2Ci2-Smart Cities Research Center - Polytechnic Institute of Tomar), Alípio M. Jorge (INESC TEC; University of Porto), Sumit Bhatia (IBM Research AI) and Mark Finlayson (Florida International University). The workshop proceedings are available here.


Two papers accepted at SIGIR 2021

Two papers have been accepted at the main Information Retrieval conference (The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2021)). One paper is about predicting the date of events described in text [1] (collaboration with Kyoto University) using transformers and multi-variate time series analysis and another on multi-modal text summarization (collaboration with IIT Patna).

[1] Jiexin Wang, Adam Jatowt, Masatoshi Yoshikawa: Event Occurrence Date Estimation based on Multivariate Time Series Analysis over Temporal Document Collections. Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2021), ACM Press, pp. 398-407 (2021) PDF

[2] Anubhav Jangra, Sriparna Saha, Adam Jatowt and Mohammed Hasanuzzaman: Multi-Modal Supplementary-Complementary Summarization using Multi-Objective Optimization. Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2021), ACM Press, pp. 818-828 (2021) PDF