Glasgow IR Group

Glasgow Information Retrieval Group @GlasgowCS

[Shen Dong @ShenDongML]

🎉 Excited to share that our paper "APR: Adaptive Personalised Reranking For Conversational Search" (with @iadh and @debforit) has been accepted to @SIGIRConf 2026!

This will be my first SIGIR. See you in Melbourne!
#IR #ConversationalSearch #NLP #SIGIR2026

[Craig Macdonald @craig_macdonald]

#sigir2026 accepted reproducibility track paper: "A Replicability Study of Joint Product Quantisation for Effective Space-Efficient Dense Retrieval", joint work with @ntonellotto and Zhili Shen

[Craig Macdonald @craig_macdonald]

Delighted that our paper “PLAID-PRF — Pseudo-Relevance Feedback with Centroid-like Tokens in PLAID” has been accepted to #sigir2026, w/ Xiao Wang and @macavaney

[Tetsuya Sakai (酒井哲也) @tetsuyasakai]

Slide deck

https://t.co/2gzwp5Ao3Y

[Tetsuya Sakai (酒井哲也) @tetsuyasakai]

I have just updated Slide #7
as @iadh kindly told me that the definition of F can be found in a paper from 1971.

[Tetsuya Sakai (酒井哲也) @tetsuyasakai]

Thank you ❤️

[Glasgow IR Group @ir_glasgow]

@KaustubhDhole @macavaney The advent mega-thread continues here 👇:
https://t.co/fXESAMEsls

[Glasgow IR Group @ir_glasgow]

@macavaney 🎄PyTerrier Advent 16/25: Speaking of Learned Sparse Retrieval, PyTerrier has bindings to two backend search engines that provide blazing-fast retrieval over sparse vectors: PISA and BMP.

You can see that we really work to keep the look-and-feel uniform between implementations.

[Glasgow IR Group @ir_glasgow]

🎄We want to try something new and fun this year – an “Advent Calendar” of PyTerrier pipelines

We’ll kick it off with *the* baseline: BM25 on MSMARCO. One line to download a pre-built index, one line to make a BM25 retriever, one line to search.

[Glasgow IR Group @ir_glasgow]

@KaustubhDhole @macavaney 🎄PyTerrier Advent 9/25: Yesterday—dense retrieval with E5 via PyTerrier_DR. Today—RAG! PyTerrier_RAG readers generate answers from retrieved docs. Example: a FiD reader over E5 results, w/ & w/o monoT5 reranking 👇 Check the notebook for answer quality.
https://t.co/0JHApuIA9B

[Siwei Liu @TedSiwei]

Happy 575th birthday to @UofGlasgow!

App made with @HowOneAI. This AI-generated app can turn an image into Line art in seconds.

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel @mengzaiqiao @JinyuanF @rodrygo_santos @iadh @craig_macdonald 🎄 PyTerrier Advent 25/25: To wrap up the our advent series, we'd like thank the contributors shown below, and to the many others who support the PyTerrier ecosystem! #WorldChangersTogether

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel @mengzaiqiao @JinyuanF @rodrygo_santos @iadh @craig_macdonald 🎄 PyTerrier Advent 25/25: To wrap up the our advent series, we'd like thank the contributors shown below, and to the many others who support the PyTerrier ecosystem! #WorldChangersTogether

[Glasgow IR Group @ir_glasgow]

🎄 PyTerrier Advent 10/25: Dense retrieval often improves with pseudo-relevance feedback (Rocchio-style).

In PyTerrier_DR it’s easy: retrieve docs, a transformer to mix doc vectors w/ the query vector, and re-retrieve.
https://t.co/NxQJYUcTIf

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel @mengzaiqiao @JinyuanF @rodrygo_santos @iadh @craig_macdonald 🎄 PyTerrier Advent 24/25: Removing low-quality docs can boost search quality and cut indexing costs. Our SIGIR’24 paper QT5 trains a T5 model to filter passages at indexing time—easy to integrate, and works with dense, PISA, or SPLADE indexers too.

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel @mengzaiqiao @JinyuanF 🎄 PyTerrier Advent 23/25: Results too homogeneous? Try a diversification reranker. Shown: implicit MMR on BM25 or dense retrieval—but explicit methods like xQuAD (@rodrygo_santos) are easy too.

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel 🎄 PyTerrier Advent 22/25: A more complex pipeline—knowledge-graph–enhanced RAG from our EMNLP 2024 paper TRACE. We build a KG over retrieved docs, then use a transformer to reason over triples for better QA. Uses cache (see yesterday's advent) on LLM KG extraction

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel 🎄PyTerrier Advent 21/25: Bounded recall blues got you down? You can use Adaptive Retrieval techniques, like GAR, LADR, and LAFF, to efficiently surface missing relevant documents.

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel 🎄PyTerrier Advent 20/25: Every PyTerrier transformer as a function mapping from one dataframe type to another. This makes them easily cachable, courtesy of pyterrier_caching. We have cache object for retrievers, rerankers, or even indexing-time transformers (e.g. Doc2Query)

[Glasgow IR Group @ir_glasgow]

@macavaney @ntonellotto @joelmmackenzie @torstensuel 🎄 PyTerrier Advent 19/25: PyTerrier-RAG brings agentic RAG to your workflows with support for SOTA methods like Search-R1 and R1-Searcher, to combine retrievers and reasoning. You could even swap BM25 out for dense or LSR retriever.