Sushil Awale

Research Associate, Visual Analytics Research Group, TIB.

prof_pic.jpg

I am a Research Associate in the Visual Analytics Research Group at TIB, Hannover and a PhD candidate at Leibniz Universität Hannover. I am also part of the Multimodal Modelling and Machine Learning (M3L) Team at Universität Marburg. I’m supervised by Prof. Dr. Ralph Ewerth.

My research focuses on Multimodal Information Retrieval for scientific, technical, and cultural domains. I build advanced IR systems leveraging vision-language models to support search in patents and improve cultural heritage access. My broader research interests include domain adaptation for better multimodal representations, Knowledge Graphs, and Natural Language Processing.

Previously, I worked as a Student Assistant in the Language Technology Group, Universität Hamburg building an intelligent scholarly system powered by Natural Language Processing (NLP).

I completed my Masters in Machine Learning from Universität Hamburg, Germany in 2023 and my Bachelors in Computer Science in 2020 from Tribhuwan University, Nepal.

news

Jun 1, 2026 Our paper Examining Gender and Racial Bias in Vision-language Models within Colonial Contexts (Awale et. al., WebSci 2026) wins Best Poster Paper Award at ACM Web Science Conference 2026.
May 1, 2026 Our paper Examining Gender and Racial Bias in Vision-language Models within Colonial Contexts (Awale et. al., WebSci 2026) accepted as poster paper at ACM Web Science Conference 2026.
Sep 9, 2025 Presented a hands-on full-day tutorial with Eric Müller-Budack titled Fusing Vision and Language: A Tutorial on Vision-Language Models for Multimodal Content Analysis at KONVENS 2025 - Conference on Natural Language Processing in Hildesheim, Germany. Slides and materials can be found here.
May 30, 2025 Our paper Exploring Patents Visually: An Interactive Search System for Multimodal Patent Image Search and Interpretation (Awale et. al., PatentSemTech@SIGIR 2025) accepted in PatentSemTech Workshop at SIGIR 2025.
Dec 16, 2024 Our paper Patent Figure Classification using Large Vision-language Models (Awale et. al., ECIR 2025) accepted as full paper at ECIR 2025.

selected publications

  1. WebSci
    Examining Gender and Racial Bias in Vision-language Models within Colonial Contexts
    Sushil Awale, Omkar Gavali, Ratan Sebastian, and 4 more authors
    In ACM Web Science Conference Companion, WebSci 2026, Braunschweig, Germany, May 26-29, 2026 2026
  2. PatentSemTech
    Exploring Patents Visually: An Interactive Search System for Multimodal Patent Image Search and Interpretation
    Sushil Awale, Eric Müller-Budack, Rahim Delaviz, and 1 more author
    In Patent Text Mining and Semantic Technologies co-located with the ACM SIGIR Conference on Research and Development in Information Retrieval, PatentSemTech@SIGIR 2025, Padua, Italy, July 17, 2025 2025
  3. ECIR
    Patent Figure Classification Using Large Vision-Language Models
    Sushil Awale, Eric Müller-Budack, and Ralph Ewerth
    In European Conference on Information Retrieval, ECIR 2025, Lucca, Italy, April 6-11, 2025 2025
  4. BIR
    DBLP-QuAD: A Question Answering Dataset over the DBLP Scholarly Knowledge Graph
    Debayan Banerjee, Sushil Awale, Ricardo Usbeck, and 1 more author
    In Workshop on Bibliometric-enhanced Information Retrieval co-located with the European Conference on Information Retrieval, BIR@ECIR 2023, Dublin, Ireland, April 2, 2023 2023