Craig Macdonald
@craig_macdonald
Followers
3K
Following
7K
Media
310
Statuses
3K
Professor of Information Retrieval
Glasgow
Joined July 2009
.@mvlacho1 presenting the last of her PhD work at #RecSys2025 : Fashion-AlterEval: A Dataset for Improved Evaluation of Conversational Recommendation Systems with Alternative Relevant Items Cc/ @ir_glasgow @GlasgowCS ๐ https://t.co/KibFfanDpB
0
2
13
๐ At #RecSys2025 spot #24, Jackson Dam is demoing RecViz, a graph-based visual analytics tool for exploring recommendation datasets & results. ๐ Great for qualitative analysis & insight generation. ๐ A fantastic undergraduate achievement at @GlasgowCS with @iadh & @ZixuanYI_
0
4
14
Alberto Mancino from Politecnico Bari explaining our joint work on Balancing Accuracy and Novelty with Sub-Item popularity at #recsys2025. First author Chiara Mallamaci was a @ir_glasgow @GlasgowCS visitor earlier this year. W/ @asash Walter Annelli, Tommaso Di Noia
1
2
14
Thrilled to join @asash on the Recsperts podcast! Thanks @MarcelKurovski for having us. We had a blast discussing our #RecSys research & transformer-based sequential recommendation. Tune in on your favorite podcast platform
recsperts.com
In episode 29 of Recsperts, I welcome Craig Macdonald, Professor of Information Retrieval at the University of Glasgow, and Aleksandr โSashaโ Petrov, PhD researcher and former applied scientist at ...
2
5
20
RA job in @ir_glasgow on an EU Horizon project, with @iadh @richardm_ and myself, on the use of LLMs in developing AI agents for the risk and ESG (Environmental, Social and Governance) assessment of sustainability-focused green investment projects. ๐
1
12
25
๐ข Come by Poster #130 right now at #ACL2025NLP ! @mengzaiqiao is here! KiRAG: Knowledge-Driven Iterative Retriever for Enhancing Retrieval-Augmented Generation ๐๐ w/ @JinyuanF and @craig_macdonald โ donโt miss the deep dive into knowledge-grounded retrieval! @ir_glasgow
0
3
21
4. ๐ Neural Passage Quality Estimation for Static Pruning https://t.co/7G9i6FTWRD
dl.acm.org
0
1
2
3. ๐ Static Pruning for Multi-Representation Dense Retrieval https://t.co/xZvilNMphH
dl.acm.org
1
0
1
Our works that Nicola cited in his talk (also @iadh, @macavaney and others): 1. ๐ E๏ฌcient Query Processing for Scalable Web Search (Foundations and Trends in IR) https://t.co/go3xMXLY3t
1
0
2
With some of us already en route home, the rest of #TeamUofG bids farewell to #SIGIR2025 & the beautiful city of Padova. We had a great week full of ideas, activities & connections. Thanks to the organisers for a fantastic conference! Until next time ๐ฎ๐น #GoodbyePadova #ByeSIGIR
1
11
39
Watching .@ntonellotto presenting at ReNeuIR workshop at #sigir2025, discussing key ideas like static pruning in neural era
2
3
21
Go to stall 091 if you want to check out the poster of this @ir_glasgow work by @DanielTian97 called "Am I on the Right Track?" #SIGIR2025 (done w/t @JinyuanF, @debforit, @mengzaiqiao and @craig_macdonald)
0
6
18
.@DanielTian97 is presenting our work on using QPP in agentic RAG w/ @JinyuanF @debforit @mengzaiqiao ๐ https://t.co/OkVsq7hQ0a
0
2
19
Currently listening to @iadh give his keynote speech on evaluating financial asset recommendation, in the FinIR workshop. Interesting stuff! @ir_glasgow
#SIGIR2025
0
8
30
Starting the day at #SIGIR2025 FinIR Workshop with a lively keynote presentation by @iadh: Beyond Profit: Evaluating Financial Recommenders with Real-World Transactions. @ir_glasgow @IDAglasgow @GlasgowCS
1
11
30
Huge congratulations to @macavaney on receiving the prestigious ACM SIGIR Early Career Researcher Award in the research category! This well-deserved recognition highlights the excellence & impact of his work in the IR community ๐๐#sigir2025 Cc @GlasgowCS @UofGlasgow @ACMSIGIR
1
16
71
How does retrieval impact on Agentic RAG? Let's see what predicted intermediate retrieval quality can tell us! Our IR-RAG@SIGIR'25 paper is on Arxiv: https://t.co/FPBEFQbuu5.
#SIGIR2025 cc/ @JinyuanF @debforit
@mengzaiqiao @craig_macdonald
arxiv.org
Agentic Retrieval-Augmented Generation (RAG) is a new paradigm where the reasoning model decides when to invoke a retriever (as a "tool") when answering a question. This paradigm, exemplified by...
1
10
27