Giovanni Trappolini

Research Fellow in AI @Sapienza_Engineering

[Giovanni Trappolini @GioTrappolini]

It was great working with you Peter!

[Peter W. Kruger @pwk]

👩‍🔬AutoBench goes scientific!🎉Started 6 months ago almost as a game, then turned into business, now it's got a fancy arXiv paper to prove it's not just fun and money. Introducing the first scientific paper that validates our Collective-LLM-as-a-Judge method!🤖📜
1/12

[Giovanni Trappolini @GioTrappolini]

@athleticKoder Great post! We address many of these points in our latest paper https://t.co/ESJGqy90ux where we introduce UDCG, a new metric for evaluating retrieval for RAG, you might want to check it out!

[Giovanni Trappolini @GioTrappolini]

Huge thanks to my amazing co-authors for making this work possible! 🙏 @FlorinCuconasu @SimoneFilice @Yoellem @fabreetseo
(n/n)

[Giovanni Trappolini @GioTrappolini]

LLMs aren't humans, so why evaluate RAG with human-centric metrics?
Traditional IR metrics fail to predict RAG performance because:
❌They assume sequential reading
❌They ignore that some docs actively distract LLMs
Meet UDCG: a new metric w/ 36% ⬆️ corr. to actual performance✅

[Giovanni Trappolini @GioTrappolini]

Retrieval evaluation needs to evolve for the LLM era. UDCG is a step toward metrics that align with machine consumers, not human browsing patterns.
📄 Full paper: https://t.co/ESJGqy90ux
💻 Code & data: https://t.co/p2gW6dw7Ni
(8/n)

[Giovanni Trappolini @GioTrappolini]

🚀Huge congratulations to Florin! 🏆 If reading the research paper isn't your thing, no worries—you can listen to the generated podcast instead, and it's actually pretty good. Quoting from it, "It's going to sound crazy!
https://t.co/OScA07oDrn

[M2L school @M2lSchool]

Well done to our 'Best Poster Award' winners! One of our winners is @FlorinCuconasu with "The Power of Noise: Redefining Retrieval for RAG Systems". Congrats!! #M2LSummerSchool

[Giovanni Trappolini @GioTrappolini]

👀👀👀

[Deedy @deedydas]

Research published by Google Deepmind reveals OpenAI Strawberry's approach.

Searches at inference through potential responses to reason better.

“test-time compute can be used to outperform a 14× larger model.”

[ECIR2025 @ecir2025]

#ECIR2025 call for Tutorials: https://t.co/KjeQZkmEzE. Tutorial proposals submission: Nov. 08, 2024. Notification: Dec. 16, 2024.

[Giovanni Trappolini @GioTrappolini]

@Fabio_Petroni @fabreetseo @_FedeSiciliano_ @ir_rag_sigir @SIGIRConf

[Giovanni Trappolini @GioTrappolini]

Almost ready for the second keynote here at the IR-RAG workshop!
Finish up your lunch and come to the presidential ballroom for Dr. Zhang presentation! @yuhaozhangx

[Giovanni Trappolini @GioTrappolini]

"Seldom is a glance at the statistics enough to understand the meaning of the figures" 🤔
Fromm had clearly RAG in mind when uttering these words! ✍️
Read our paper to find out more about BaseVsInstruct for RAG! 🚩

[Giovanni Trappolini @GioTrappolini]

@RSTLessGroup @giulio_deras 🚀

[Giovanni Trappolini @GioTrappolini]

🚨 If you, or someone you know, are looking for a PhD you should definitely check this out! 🤌

[Giovanni Trappolini @GioTrappolini]

Huge shout-out to the amazing @FlorinCuconasu and the team @_FedeSiciliano_ @SimoneFilice @caesar_one_ @Yoellem @ntonellotto @fabreetseo

[Giovanni Trappolini @GioTrappolini]

📢 Check out the refined version of “The Power of Noise: Redefining Retrieval for RAG Systems”! accepted @SIGIR2024.

A huge thank you to the reviewers and the research community for your invaluable feedback—it has significantly improved the paper. (1/n)

[Giovanni Trappolini @GioTrappolini]

p.s. We have now added a statement clarifying that we use the instruct-version of the LLMs.