Debasis Ganguly
@debforit
Followers
761
Following
1K
Statuses
580
Lecturer/Asst. Professor at the School of Computing, University of Glasgow (@UofGlasgow/@GlasgowCS/@IDAglasgow/@ir_glasgow)
Glasgow, Scotland
Joined February 2010
RT @iadh: We have 2 open-ended (tenure-track) faculty openings in Machine Learning at @GlasgowCS. Applications are welcome in the general a…
0
12
0
Excited to hear Prasenjit Mitra talk about efficient and safe LLMs.
0
0
1
@VirginAtlantic Unacceptable to not get any information about missing baggage even after 4 days! Very bad customer support!
2
0
0
My baggage went missing at the destination of a 1-stop flight from Glasgow to Mumbai via London (@British_Airways Glasgow to Heathrow and @VirginAtlantic from @HeathrowAirport to Mumbai). My baggage hasn't been traced yet! A complete lack of sync btwn the two airlines!
5
0
0
Listening to a talk on the use of LLMs for legal documents by Yiqun Liu of Tsinghua University @fire_forum
0
0
5
RT @fire_forum: 🎉 Exciting News for Students! 🎉 We are thrilled to announce a limited number of student travel grants for FIRE 2024 🔗 App…
0
3
0
A very well attended Open Day at the @GlasgowCS where, in addition to the lab tour, we, for the 1st time, tried an engagement session with talks and posters from PhD and ungergrad students. Special thanks to Jeremy for his lively talk on CS as an abstraction.
0
1
11
Very rewarding as a researcher to see that our #sigir2024 perspective paper on the use of IR for RAG is gaining some significant interest from the software industry.
I am excited to hold a talk about In-Context Learning, with inspiration from the “In-Context Learning” or: How I learned to stop worrying and love “Applied Information Retrieval" paper from the @UofGlasgow! This is a promising way to integrate large language models to improve the quality of completing specific tasks without hallucinations. Also, getting started does not take much, particularly when using @vespaengine! Today, at @mlconference in Brooklyn, New York:
0
1
19
Paper "In-Context Learning with Localised Examples as an Effective Non-Parametric Retrieval Model" accepted in #emnlp2024. We show: all that's needed for effective search is a small number of examples of relevant documents for similar information needs. @MrParryParry @nilanjansb
2
0
22
Glad to organise 3 tasks at FIRE-2024 @fire_forum. Please participate or spread the word. ILSUM: Detect misinformed news summaries. IRSE: Are code comments useful? Is LLM-generated code trustworthy? SqCLIR: CLIR on spoken queries.
0
2
12
Working on Query Performance Prediction (#QPP) and/or its applications for IR and NLP? Consider submitting to this special issue in @acmtois (IF: 5.6), where as as fellow guest editors I'm joined by @suchanadatta, @guglielm0f, @frrncl, & @iadh.
0
19
34
Glad that this work has gained some attention, which is the whole point of a "perspective". Thanks for the mention @jobergum
There is an excellent SIGIR perspective paper on how ICL can be seen upon as a retrieval problem. More on this soon 🔜
1
2
6
RT @MrParryParry: 🚨 Happy to say that I'll be presenting our work (w/@macavaney & @debforit) "Top-Down Partitioning for Efficient List-Wise…
0
5
0
Very happy for my PhD student @MrParryParry for this excellent analysis on the vulnerabilities of neural rankers. #ACL2024NLP
Happy to share that our work "Exploiting Positional Bias for Query-Agnostic Generative Content in Search" has been accepted into ACL findings 🤠, here's a pre-print with updates coming soon:
0
0
16
@nicolayr_ The #SIGIR2024 paper has the same title... I'm too lazy and old-school for a composing a breakdown tweet... but maybe I should pick up that habit 😆
0
0
1
Here's the arxiv link of the paper: And here's the youtube talk:
0
0
3