←
Cosimo Gregucci47 posts
Cosimo Gregucci avatar

Cosimo Gregucci

@c_gregucci

#MSCA PhD student in #ML #AI at @Uni_Stuttgart | #knowledge #graphs #neuro #symbolic #reasoning

◷ Joined October 2021▧ 9 Photos and videos
Cosimo Gregucci avatar
Cosimo Gregucci@c_gregucci·17 Jul 2025

Thanks to everyone who stopped by our work! If you missed it and want to know more, just drop me a message! #ICML2025

Post attachment

Spotlight poster coming soon at #ICML2025 @icmlconf! 📌East Exhibition Hall A-B E-1806 🗓️Wed 16 Jul 4:30 p.m. PDT — 7 p.m. PDT 📜arxiv.org/pdf/2410.12537 Let’s chat! I’m always up for conversations about knowledge graphs, reasoning, neuro-symbolic AI, and benchmarking.

Post attachment
Cosimo Gregucci avatar
Cosimo Gregucci@c_gregucci·10 Jul 2025

Spotlight poster coming soon at #ICML2025 @icmlconf! 📌East Exhibition Hall A-B E-1806 🗓️Wed 16 Jul 4:30 p.m. PDT — 7 p.m. PDT 📜arxiv.org/pdf/2410.12537 Let’s chat! I’m always up for conversations about knowledge graphs, reasoning, neuro-symbolic AI, and benchmarking.

Post attachment

🚨Is complex query answering really complex?🚨 unfortunately not! the current benchmarks boil down to link prediction 98% of the time... how to fix this??? 👇👇👇 📜arxiv.org/abs/2410.12537 with @c_gregucci @BoXiongs @loreloc_ @PMinervini @ststaab

Post attachment
Pasquale Minervini avatar
⟳ Cosimo Gregucci retweeted
Pasquale Minervini@PMinervini·14 Mar 2025

hi, we found problematic benchmarks in complex query answering as well! arxiv.org/abs/2410.12537 @tetraduzione @Mniepert @c_gregucci @chrsmrrs @michael_galkin

Pasquale Minervini avatar
⟳ Cosimo Gregucci retweeted
Pasquale Minervini@PMinervini·7 Mar 2025

We will be hiring one research fellow to work on multi-modal foundation models! The job ad will be online in the next few days 🙂

Our Generative AI Hub (genai.ac.uk/) will soon be hiring a Research Fellow to work on multi-modal AI models at @InfAtEd! If you're interested, please watch this space! 🚀🚀🚀

Lorenzo Loconte avatar
⟳ Cosimo Gregucci retweeted
Lorenzo Loconte@loreloc_·27 Feb 2025

We are going to present our poster "Sum of Squares Circuits" at AAAI in Philadelphia today Hall E 12:30pm-14:00pm poster #840 We trace expressiveness connections of different types of additive and subtractive deep mixture models and tensor networks 📜 arxiv.org/abs/2408.11778

We learn more expressive mixture models that can subtract probability density by squaring them 🚨We show squaring can reduce expressiveness To tackle this we build sum of squares circuits🆘 🚀We explain why complex parameters help, and show an expressiveness hierarchy around🆘

Diego Calanzone avatar
⟳ Cosimo Gregucci retweeted
Diego Calanzone@diegocalanzone·22 Jan 2025

Finally, LOgically COnsistent (LoCo) LLaMas can outperform solver-based baselines and SFT! I thank @tetraduzione and @looselycorrect for the guidance in realizing this project, get in touch or come to chat in Singapore! arxiv.org/abs/2409.13724

antonio vergari ⚔️ avatar
⟳ Cosimo Gregucci retweeted
antonio vergari ⚔️@tetraduzione·17 Oct 2024

🚨Is complex query answering really complex?🚨 unfortunately not! the current benchmarks boil down to link prediction 98% of the time... how to fix this??? 👇👇👇 📜arxiv.org/abs/2410.12537 with @c_gregucci @BoXiongs @loreloc_ @PMinervini @ststaab

Post attachment

yes! the way missing edges are placed is the key ingredient to complexity in CQA. 🚨🚨🚨 Unfortunately, all current benchmarks are baby benchmarks requiring to predict a single missing edge for 99% queries... Bottom line: we are just evaluating link prediction...

Post attachmentPost attachment
Santiago avatar
⟳ Cosimo Gregucci retweeted
Santiago@svpino·11 Oct 2024

Large Language Models don't reason. Thank you, Apple.

Post attachment
Lorenzo Loconte avatar
⟳ Cosimo Gregucci retweeted
Lorenzo Loconte@loreloc_·7 Oct 2024

We learn more expressive mixture models that can subtract probability density by squaring them 🚨We show squaring can reduce expressiveness To tackle this we build sum of squares circuits🆘 🚀We explain why complex parameters help, and show an expressiveness hierarchy around🆘