iiinova — where your digital self keeps evolving

Pasquale Minervini @neuralnoise.com · 02/08/2026

stochastic parrots are getting pretty lucky

Pasquale Minervini @neuralnoise.com · 04/07/2026

New blog post on our {ICML, ACL} 2026 papers, plus some new interesting results on open-ended learning and multi-modal retrieval! neuralnoise.com/2026/icml-ac...

7170

Pasquale Minervini @neuralnoise.com · 27/03/2026

If you are interested in privacy-preserving clinical NLP, we are recruiting a postdoc at the University of Edinburgh! The work is on LLMs/VLMs, AI privacy, and real-world health data in secure research environments. Apply by April 6th, 2026! More details: elxw.fa.em3.oraclecloud.com/hcmUI/Candid...

2102

Pasquale Minervini @neuralnoise.com · 17/10/2025

Check out Yu Zhao's (@yuzhaouoe.bsky.social) latest work, “Learning GUI Grounding with Spatial Reasoning from Visual Feedback” (www.arxiv.org/abs/2509.21552), done during his internship at MSR (@msftresearch.bsky.social)! New SOTA 🏆 results on ScreenSpot-v2 (+5.7%) and ScreenSpot-Pro (+110.8%)!

131

Pasquale Minervini @neuralnoise.com · 23/08/2025

I really needed a Deep Research MCP server to use with Claude Code and other tools — here it is: github.com/pminervini/d...

060

Pasquale Minervini @neuralnoise.com · 31/07/2025

Supermassive congrats to Giwon Hong (@giwonhong.bsky.social) for the amazing feat! 🙂

141

Pasquale Minervini @neuralnoise.com · 26/07/2025

The amazing folks at EdinburghNLP will be presenting a few papers at ACL 2025 (@aclmeeting.bsky.social); if you're in Vienna, touch base with them!

0120

Pasquale Minervini @neuralnoise.com · 05/07/2025

"in 2025 we will have flying cars" 😂😂😂

839991

Pasquale Minervini @neuralnoise.com · 25/05/2025

claude-code is pretty good at updating personal websites! it has browser use, so it can e.g. scrape your latest papers from arxiv and dblp and use that to update your website's publication list

250

Pasquale Minervini @neuralnoise.com · 30/04/2025

Extremely proud of the squad 💕

030

Pasquale Minervini @neuralnoise.com · 25/04/2025

Massive congrats @manueldileo.bsky.social!!! 🚀🚀🚀🚀🚀🚀🚀

070

Pasquale Minervini @neuralnoise.com · 19/04/2025

My amazing collaborators will present several works at ICLR and NAACL later this month -- please catch up with them if you're attending! I tried to summarise our recent work in a blog post: neuralnoise.com/2025/march-r...

0175

Pasquale Minervini @neuralnoise.com · 10/04/2025

us & some of the topics we're working on! neuralnoise.com/publications/

020

Pasquale Minervini @neuralnoise.com · 01/04/2025

Still ~8 days to apply for a postdoc position in multimodal foundation models at the University of Edinburgh! (@edinburgh-uni.bsky.social) -- Fully funded position until 2029 by the Generative AI Hub (@genaihub.bsky.social) to work with outstanding research teams! neuralnoise.com/2025/multimo...

084

Pasquale Minervini @neuralnoise.com · 17/03/2025

madly in love with this article from @thesun.co.uk covering a paper from @rohit-saxena.bsky.social and @aryopg.bsky.social Paper: arxiv.org/abs/2502.05092 The Sun: www.thesun.co.uk/tech/3384555...

081

Pasquale Minervini @neuralnoise.com · 13/03/2025

Please share it within your circles! edin.ac/3DDQK1o

0139

Pasquale Minervini @neuralnoise.com · 19/02/2025

if you're into RL you likely know this one!

110

Pasquale Minervini @neuralnoise.com · 10/02/2025

LLMs can tackle math olympiad problems, but… can they read a clock or calendar? 🕰️📆 Our experiments reveal surprising failures in temporal reasoning—multimodal models struggle with analogue clock reading and date inference! Paper Link: arxiv.org/abs/2502.05092 Dataset: huggingface.co/datasets/roh...

0132

Pasquale Minervini @neuralnoise.com · 28/01/2025

Our work on fixing and improving MMLU ("Are We Done with MMLU?", arxiv.org/abs/2406.04127; NAACL 2025) is featured on the DeepSeek home page! deepseek.com -- it's always amazing to see real-world applications and the impact of academic research 🚀🚀🚀

0236

Pasquale Minervini @neuralnoise.com · 14/01/2025

Friendly reminder that we are advertising for a permanent faculty position in Embodied NLP at the School of Informatics, University of Edinburgh (informatics.ed.ac.uk) -- apply by Jan 31st! edin.ac/4fqgawg

050

Pasquale Minervini @neuralnoise.com · 21/12/2024

Is the “wall” in the room with us right now?

290

Pasquale Minervini @neuralnoise.com · 20/12/2024

I tried googling some subspans but it seems original content

000

Pasquale Minervini @neuralnoise.com · 19/12/2024

The School of Informatics at the University of Edinburgh is hiring a lecturer or reader in embodied natural language processing. Apply by 31 Jan 2025 at edin.ac/4fqgawg

21915

Pasquale Minervini @neuralnoise.com · 13/12/2024

- Manning: can you review a book? - me: sure! what's the compensation? - Manning:

2100

Pasquale Minervini @neuralnoise.com · 12/12/2024

Hey I know this paper! 😍 www.nature.com/articles/s41...

040

Pasquale Minervini @neuralnoise.com · 06/12/2024

The correct answer for this one is C 🙂 you can read more about errors in MMLU here: arxiv.org/abs/2406.04127

020

Pasquale Minervini @neuralnoise.com · 06/12/2024

Garbage in, garbage out -- nice gem for the Italian-speaking folks on this platform 😅 TLDR, in arxiv.org/abs/2406.04127 we found that MMLU contains TONS of errors, and looks like all these seamlessly propagated to this new "Global MMLU" dataset

0112

Pasquale Minervini @neuralnoise.com · 29/11/2024

still not enough agents on this platform 😅

010

Pasquale Minervini @neuralnoise.com · 25/11/2024

Thank you ARR! So we can avoid working through the weekend! 😊

060

Pasquale Minervini @neuralnoise.com · 20/11/2024

here's the guy! 5.4 tok/sec with no batching -- you can very likely increase the throughput by batching requests!

210

Pasquale Minervini @neuralnoise.com · 20/11/2024

Still wrapping my head around the fact that a consumer laptop can now run 90B VLMs without batting an eye (best spent grant funding so far)

4311

Pasquale Minervini @neuralnoise.com · 19/11/2024

My fav reference on the manifold assumption -- where datapoints (in this case, activations) are distributed on locally-linear lower-dimensional manifolds that can be approximated via e.g. generative models or neighbourhood graph structures -- is this amazing book: www.molgen.mpg.de/3659531/MITP...

2191

Pasquale Minervini @neuralnoise.com · 16/11/2024

"LLMs can't reason, look at how their accuracy drops if you change the numbers in the problem!!!" The accuracy drop (%):

161

Pasquale Minervini @neuralnoise.com · 12/11/2024

A Simple and Effective L2 Norm-Based Strategy for KV Cache Compression arxiv.org/abs/2406.11430! -- Check Alessio's talk today at the EMNLP'24 oral session on Low Resource Methods for NLP! #EMNLP2024

091

Pasquale Minervini @neuralnoise.com · 12/11/2024

Folks, is this expected? I'm using Cursor with GPT4o in the backend

130

Pasquale Minervini @neuralnoise.com · 09/11/2024

My amazing collaborators will be presenting three papers next week at EMNLP 2024! I wrote a blog post about our EMNLP papers and some of the other projects we're brewing 🚀🙂 neuralnoise.com/2024/nov-res...

0115