Andrew Parry

profile.jpg

Room 502, Level 5

Sir Alwyn Williams Building

Glasgow, G12 8QQ

I am a 3rd year PhD Candidate at the University of Glasgow working within the IR group. I am supervised by Dr Debasis Ganguly and Dr Sean MacAvaney. My work is currently supported by the Bloomberg PhD Data Science Fellowship. My research interests are primarily in Information Retrieval. Before my PhD I completed my honours degree in Computer Science at Glasgow.

My thesis is focused on inducing control over neural models without post-hoc constraint and how we can incorporate broader features in semantic interactions. A relatively up to date CV can be found here. Reach out on Twitter or by email if you want to chat about research or code. I’m generally interested in collaborations centred around robustness of neural systems under adversarial settings and semi-supervised signals in NLP.

news

Apr 05, 2025 Our reproducibility paper, short paper and demo paper are accepted at SIGIR 2025!
Mar 10, 2025 Our workshop paper “Corpora Performance Prediction” is accepted at QPP++@ECIR2025!
Feb 15, 2025 Our workshop paper “Large Language Model Relevance Assessors Agree With One Another More Than With Human Assessors” is accepted at LLM4Eval@WSDM 2025!
Dec 16, 2024 Our full paper and demo paper are accepted at ECIR 2025!
Dec 08, 2024 I have been awarded the Bloomberg Data Science Fellowship!

selected publications

  1. MechIR: A Mechanistic Interpretability Framework for Information Retrieval
    Andrew Parry, Catherine Chen, Carsten Eickhoff, and Sean MacAvaney
    In Advances in Information Retrieval - 47th European Conference on Information Retrieval, ECIR 2025, Lucca, Italy, April 6-10, 2025, Proceedings, Part V , 2025
  2. Variations in Relevance Judgments and the Shelf Life of Test Collections
    Andrew Parry, Maik Fröbe, Harrisen Scells, Ferdinand Schlatt, Guglielmo Faggioli, and 3 more authors
    CoRR, 2025
  3. Analyzing Adversarial Attacks on Sequence-to-Sequence Relevance Models
    Andrew Parry, Maik Fröbe, Sean MacAvaney, Martin Potthast, and Matthias Hagen
    In Advances in Information Retrieval - 46th European Conference on Information Retrieval, ECIR 2024, Glasgow, UK, March 24-28, 2024, Proceedings, Part II , 2024
  4. Top-Down Partitioning for Efficient List-Wise Ranking
    Andrew ParrySean MacAvaney, and Debasis Ganguly
    CoRR, 2024