Dennis Aumiller

Phone: +49 (0) 6221 / 54 - 14353
Fax: +49 (0) 6221 / 54 - 14351
Office: INF 205, room 1/312 (first floor)
Email: aumiller(at)informatik.uni-heidelberg(dot)de
Office hours: By appointment

---

News

2023-07-25: After concluding my time as a research assistant in the Data Science Group, I will be joining Cohere's Data and Evaluation team as a Member of Technical Staff in September!

2023-05-16: Our paper "Evaluating Factual Consistency of Texts with Semantic Role Labeling" has been accepted at the The 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023)! I will be presenting our insights in person at the conference, held jointly with ACL 2023 in Toronto, Canada.

2023-01-18: A new long paper, titled "On the State of German (Abstractive) Text Summarization" has been accepted at the 20th Conference on Database Systems for Business, Technology and Web (BTW'23).

2022-11-01: Our submission to the shared task on Lexical Simplification at the Text Simplification, Accessibility and Readability Workshop (TSAR-2022) obtained the highest scores across all systems! The paper describing our approach will be presented at the TSAR Workshop, co-located with EMNLP'22. A pre-print is available on arXiv.

2022-10-06: Our work Eur-Lex-Sum: A Multi- and Cross-lingual Dataset for Long-form Summarization in the Legal Domain has been accepted to the main track of The 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP'22)! A pre-print and the public version of our dataset are available now.

2022-04-04: More great news! The previously announced pre-print Klexikon: A German Dataset for Joint Summarization and Simplification has been accepted with oral presentation at the 13th Conference on Language Resources and Evaluation (LREC'22)! Similarly, I am looking forward to attending a physical conference again. Feel free to reach out and let me know if you are around Marseilles in late June.

2022-03-31: Our demonstration paper Online DATEing: A Web Interface for Temporal Annotations has been accepted at the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR'22)! You can check out a system demonstration online. I will be attending the conference in person, so feel free to reach out if you want to have a coffee in Madrid this summer!

2022-03-01: Our short paper Time for some German? Pre-Training a Transformer-based Temporal Tagger for German has been accepted at the Fifth International Workshop on Narrative Extraction from Texts (Text2Story@ECIR'22)! You can also check out our framework on GitHub.

2022-01-18: A pre-print of a newly released data resource Klexikon: A German Dataset for Joint Summarization and Simplification has been published on arXiv! See here for the paper, or check out the code on GitHub.

2021-07-01: I will be joining Amazon Research Berlin as an Applied Scientist Intern later this year.

---

About

My main interests are focused on Natural Language Processing on large document collections. Specifically, I investigate suitable models for (multi-)document summarization, particularly on collections that work with domain-specific texts or non-English languages. A secondary focus lies on the incorporation of user-specified aspects during the generation process of summaries, which can enable a wider audience to benefit from the (currently rather limited) applicability of summarization models.
We also previously experimented with other forms of text exploration, including the visualization of information that allow for improved content density, or the re-structuring of document sections in a more (temporally) consistent manner. I strive to ensure our research stays reproducible; as a commitment, most of our methods and artifacts can be found on my Github profile, or on the Huggingface Hub.

I previously studied Applied Computer Science with a minor in Computational Linguistics, also at Heidelberg University and finished my Master of Science in May 2019. Afterwards, I spent four wonderful years as part of the Data Science Group group at the Institute of Computer Science, under the supervision of Prof. Dr. Michael Gertz.

---

Reviewing Activities

• LIRAI@ACM Hypertext 2023 (PC member)
  
• Text2Story@ECIR 2023 (PC member, "Recognized Reviewer Award")
  
• EDBT 2023 (external)
• ECIR 2022, ECIR 2023 (external)
• TSAR@EMNLP 2022 (shared task reviewer)
  
• ARR September 21 Cycle (emergency reviewer)
• SIGMOD 2021 (external)
• BTW 2021, BTW 2023 (external)
  
• SustaiNLP@EMNLP 2020, SustaiNLP@EMNLP 2021 (PC member)
  
• SDP@EMNLP 2020 (shared task reviewer)
• AI4Narratives@IJCAI-PRICAI 2020 (PC member)
  
• AAAI 2020 (external)
• EMNLP 2019, EMNLP 2020 (external)
  

---

Teaching & Supervision

Courses:

• Lecture Assistant for "Data Science for Text Analytics" (Winter 2022)
  
• Lecture Assistant for graduate course "Text Analytics" (Winter 2020)
  
• Lecture Assistant for "Databases 1" (Summer 2019, Summer 2020, Summer 2021, Summer 2022)
• Head Teaching Assistant for graduate course "Complex Network Analysis" (Winter 2018)
• Teaching Assistant for "Databases 1" (Summer 2016, Summer 2017)
• Head Teaching Assistant for graduate course "Computer Graphics" (Winter 2016, with Prof. Dr. Filip Sadlo)

Student Practicals:

• continual supervision of (under-)graduate semester research projects, since Summer 2019

Co-supervised Seminars:

• "Modern Information Retrieval" (Summer 2023)
  
• "Domain-Specific Question Answering" (Summer 2022)
  
• "Knowledge Graphs and NLP" (Summer 2021)
• "Trends and Topics in Text Analytics" (Summer 2020)
  

Supervised Master Theses (co-supervision with Michael Gertz):

• Jiahui Li: "Styled Text Summarization via Domain-Specific Paraphrasing" (Summer 2023)
  
• Fabio Becker: "A Generative Model for Dynamic Networks with Community Structures" (Winter 2020)

Supervised Undergraduate Theses (co-supervision with Michael Gertz):

• Jing Fan: "Assessing Factual Accuracy of Generated Text" (Winter 2022)
  
• Mateusz Chrzastek: "Extractive Keyphrases from Noun Chunk Similarity" (Winter 2021)
  
• Jan-Gabriel Mylius : "Visual Analysis of Paragraph Similarity" (Winter 2020)
  
• Stefan Hickl: "Automatisierte Generierung von Inhaltsverzeichnissen aus PDF-Dokumenten" (Summer 2020)

Faculty Responsibilities:

• Member of the "Prüfungsausschuss Informatik" (since October 2022)
  

---

Research Interests

•     (Multi-)Document Summarization
•     Text Simplification
•     Temporal Tagging
•     Keyphrase Extraction
•     (Temporal) Information Retrieval
•     Broader (German) Natural Language Processing / Machine Learning