Giovanni Trappolini

49 posts

Giovanni Trappolini

@GioTrappolini

Research Fellow in AI @Sapienza_Engineering

Katılım Aralık 2016

280 Takip Edilen146 Takipçiler

Giovanni Trappolini@GioTrappolini·29 Eki

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

English

128

Giovanni Trappolini@GioTrappolini·28 Eki

@athleticKoder Great post! We address many of these points in our latest paper arxiv.org/abs/2510.21440 where we introduce UDCG, a new metric for evaluating retrieval for RAG, you might want to check it out!

English

athletic coder@athleticKoder·28 Eki

You're in a ML Engineer interview at Perplexity , and they ask: "Your RAG system retrieves at 80% accuracy but only answers correctly 50% of the time. What's wrong?" Here's is how you answer:

English

917

116.9K

Giovanni Trappolini@GioTrappolini·27 Eki

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

English

Giovanni Trappolini@GioTrappolini·27 Eki

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: arxiv.org/abs/2510.21440 💻 Code & data: github.com/GiovanniTRA/UD… (8/n)

English

Giovanni Trappolini@GioTrappolini·27 Eki

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✅

English

168

Giovanni Trappolini@GioTrappolini·16 Eyl

🚀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! shorturl.at/NlVGc

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

English

1.1K

Giovanni Trappolini@GioTrappolini·10 Eyl

👀👀👀

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.”

ART

146

Giovanni Trappolini retweetledi

ECIR2025@ecir2025·5 Eyl

#ECIR2025 call for Tutorials: ecir2025.eu/call-for-tutor…. Tutorial proposals submission: Nov. 08, 2024. Notification: Dec. 16, 2024.

Català

359

Giovanni Trappolini@GioTrappolini·18 Tem

@Fabio_Petroni @fabreetseo @_FedeSiciliano_ @ir_rag_sigir @SIGIRConf

QAM

112

Giovanni Trappolini@GioTrappolini·18 Tem

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

GIF

English

252

Giovanni Trappolini@GioTrappolini·5 Tem

"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! 🚩

Fabrizio Silvestri@fabreetseo

📷 Excited to share our latest research on Retrieval Augmented Generation (RAG) systems! We found that base LLMs outperform instructed ones in RAG tasks by 20%! Dive into the details here: arxiv.org/abs/2406.14972 #AI #NLP #RAG #Research

English

495

Giovanni Trappolini@GioTrappolini·2 Tem

@RSTLessGroup @giulio_deras 🚀

QME

RSTLess group@RSTLessGroup·1 Tem

Meet @giulio_deras, a first-year #PhD student in #AI at #Sapienza sponsored by #LeonardoLabs. He holds a B.Sc. in #Math and a M.Sc. in #DataScience. His research interests are focused on Open Domain Question and Answer systems #Q&A. #LLM #QA

English

306

Giovanni Trappolini@GioTrappolini·24 May

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

Fabrizio Silvestri@fabreetseo

🚀 Exciting PhD Opportunities at @SapienzaRoma ! 🚀 Looking for a fully-funded 3-year #PhD in #DataScience? @SapienzaRoma has openings in: * #DataScience * #MachineLearning * #GraphLearning * #NLP * #eXplainableAI - #XAI Apply by June 20th! Details: lnkd.in/dPyQeSWU

English

218

Giovanni Trappolini@GioTrappolini·2 May

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

English

228

Giovanni Trappolini@GioTrappolini·2 May

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

English

249

Giovanni Trappolini@GioTrappolini·2 May

📢 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)

English

1.3K

Keşfet

@athleticKoder @FlorinCuconasu @SimoneFilice @Yoellem @fabreetseo @Fabio_Petroni @_FedeSiciliano_ @ir_rag_sigir