Lorenzo Loconte

159 posts

Lorenzo Loconte

Lorenzo Loconte

@loreloc_

PhD Student @ University of Edinburgh

Edinburgh, Scotland Katılım Mart 2017
1.3K Takip Edilen526 Takipçiler
Sabitlenmiş Tweet
Lorenzo Loconte
Lorenzo Loconte@loreloc_·
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🆘
GIF
English
3
32
162
21.4K
Lorenzo Loconte retweetledi
Peyman Milanfar
Peyman Milanfar@docmilanfar·
AI people going on about Johnson-Lindenstrauss lemma like it was discovered yesterday. it’s just another example of how most folks don’t read or know anything more than 5 years old
English
9
12
308
25.8K
Lorenzo Loconte retweetledi
Viacheslav Borovitskiy
Viacheslav Borovitskiy@vabor112·
Happy to share a major milestone: after years of development, we are officially launching Version 1.0 of the GeometricKernels library! To top it off, our accompanying paper has just been published in JMLR (MLOSS)! 🎉 github.com/geometric-kern…
Viacheslav Borovitskiy tweet media
English
7
107
739
34.1K
Lorenzo Loconte retweetledi
Adrián Javaloy
Adrián Javaloy@javaloyML·
I am a bit late to the party, but I am happy to share that our latest work was accepted to #ICLR2026 🥳🥳 📜 How to Square Tensor Networks and Circuits Without Squaring Them arxiv.org/abs/2512.17090
English
1
1
10
330
Lorenzo Loconte retweetledi
Stefano Teso ✈️ AAAI26
Stefano Teso ✈️ AAAI26@looselycorrect·
Good news everyone! This year we will be organizing a workshop on Unifying Concept Representation Learning at ICLR'26! The workshop is about unifying ideas and techniques from #NeSy AI, #XAI and #Causal representation learning. Have a look at the CfP!
UCRL Workshop @ ICLR 2026@UcrlW_iclr2026

📢 Announcing the Workshop on **Unifying Concept Representation Learning** at ICLR’26 ( @iclr_conf ). When? 26 or 27 April 2026 Where? Rio de Janeiro, Brazil Call for papers, schedule, invited speakers & more: ucrl-iclr26.github.io Looking forward to your submissions!

English
2
18
26
3.4K
Lorenzo Loconte retweetledi
NeurIPS Europe Conference
NeurIPS Europe Conference@EurIPSConf·
Congratulations to everyone who got their @NeurIPSConf papers accepted 🎉🎉🎉 At #EurIPS we are looking forward to welcoming presentations of all accepted NeurIPS papers, including a new “Salon des Refusés” track for papers which were rejected due to space constraints!
NeurIPS Europe Conference tweet media
English
4
25
238
51.3K
Lorenzo Loconte retweetledi
NeSy 2026
NeSy 2026@nesyconf·
@luislamb We're glad to announce the NeSy 2025 Test of Time award for "Probabilistic Inference Modulo Theories"! 🏆Rodrigo de Salvo Braz was here to accept the award. This is groundwork for recent NeSy approaches like DeepSeaProbLog and the probabilistic algebraic layer.
NeSy 2026 tweet mediaNeSy 2026 tweet mediaNeSy 2026 tweet media
English
1
8
13
883
Lorenzo Loconte retweetledi
Samy Badreddine
Samy Badreddine@sbadredd·
@deedydas Insightful! We tackled the same problem in Knowledge Graph Completion. Dot-product scoring on low-dim embeddings severely limits what a model can predict. We call this a “rank bottleneck” to align with the existing LM literature. Our paper for context: arxiv.org/abs/2506.22271
English
1
2
4
389
Lorenzo Loconte retweetledi
Cosimo Gregucci@ICML
Cosimo Gregucci@ICML@c_gregucci·
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.
Cosimo Gregucci@ICML tweet media
antonio vergari ⚔️@tetraduzione

🚨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

English
0
5
20
4.1K
Lorenzo Loconte retweetledi
Emanuele Marconato
Emanuele Marconato@ema_marconato·
🧵Why are linear properties so ubiquitous in LLM representations? We explore this question through the lens of 𝗶𝗱𝗲𝗻𝘁𝗶𝗳𝗶𝗮𝗯𝗶𝗹𝗶𝘁𝘆: “All or None: Identifiable Linear Properties of Next-token Predictors in Language Modeling” Published at #AISTATS2025🌴 1/9
Emanuele Marconato tweet media
English
8
56
318
29.4K
Lorenzo Loconte retweetledi
Emile van Krieken
Emile van Krieken@EmilevanKrieken·
We propose Neurosymbolic Diffusion Models! We find diffusion is especially compelling for neurosymbolic approaches, combining powerful multimodal understanding with symbolic reasoning 🚀 Read more 👇
GIF
English
32
106
589
54.3K
Flavio Adamo
Flavio Adamo@flavioAd·
I asked Codex to convert a legacy project from Python 2.7 to 3.11 and from Django 1.x to 5.0 It literally took 12 minutes If you know, that’s usually weeks of pain This is actually insane
Flavio Adamo tweet media
English
134
220
3.5K
833.5K
Lorenzo Loconte retweetledi
Lennert De Smet
Lennert De Smet@LennertDS_·
Just under 10 days left to submit your latest endeavours in ⚡#tractable⚡ probabilistic models❗ Join us at TPM @auai.org #UAI2025 and show how to build #neurosymbolic / #probabilistic AI that is both fast and trustworthy!
antonio vergari ⚔️@tetraduzione

the #TPM ⚡Tractable Probabilistic Modeling ⚡Workshop is back at @UncertaintyInAI #UAI2025! Submit your works on: - fast and #reliable inference - #circuits and #tensor #networks - normalizing #flows - scaling #NeSy #AI 🕓 deadline: 23/05/25 👉 …able-probabilistic-modeling.github.io/tpm2025/

English
0
4
6
635
Lorenzo Loconte retweetledi
Jaron Maene
Jaron Maene@jjcmoon·
We developed a library to make logical reasoning embarrassingly parallel on the GPU. For those at ICLR 🇸🇬: you can get the juicy details tomorrow (poster #414 at 15:00). Hope to see you there!
Jaron Maene tweet media
English
1
7
15
655
Lorenzo Loconte retweetledi
Eleonora Giunchiglia
Eleonora Giunchiglia@e_giunchiglia·
🚨New at #ICLR: we introduce the first ever 𝐥𝐚𝐲𝐞𝐫 that makes 𝐚𝐧𝐲 neural network 𝐜𝐨𝐦𝐩𝐥𝐢𝐚𝐧𝐭 𝐛𝐲 𝐝𝐞𝐬𝐢𝐠𝐧 with constraints expressed as 𝐝𝐢𝐬𝐣𝐮𝐧𝐜𝐭𝐢𝐨𝐧𝐬 𝐨𝐟 𝐥𝐢𝐧𝐞𝐚𝐫 𝐢𝐧𝐞𝐪𝐮𝐚𝐥𝐢𝐭𝐢𝐞𝐬—even if they define 𝐧𝐨𝐧-𝐜𝐨𝐧𝐯𝐞𝐱 𝐬𝐩𝐚𝐜𝐞𝐬!
Eleonora Giunchiglia tweet media
English
1
7
35
3.3K
Russell Tsuchida
Russell Tsuchida@RussellTsuchida·
Squared families can be viewed as simplified versions of previous models, such as squared probabilistic circuits (@loreloc_ ), or squared neural families. We fix the "hidden feature" for extra tractable theory, while still allowing for rich representations.
English
1
0
1
167
Lorenzo Loconte retweetledi
Pascal Hitzler
Pascal Hitzler@pascalhitzler·
Pascal Hitzler tweet media
ZXX
0
6
37
1.1K
Lorenzo Loconte retweetledi
Bhavya Kailkhura
Bhavya Kailkhura@bkailkhu·
Our paper "Low-rank finetuning for LLMs is inherently unfair" won a 𝐛𝐞𝐬𝐭 𝐩𝐚𝐩𝐞𝐫 𝐚𝐰𝐚𝐫𝐝 at the @RealAAAI colorai workshop! #AAAI2025 Congratulations to amazing co-authors @nandofioretto @WatIsDas @CuongTr95450563 and M. Romanelli 🥳🥳🥳
Bhavya Kailkhura tweet media
Bhavya Kailkhura@bkailkhu

Fine-tuning your LLM with LoRA for critical areas like ⚖️ criminal justice, 🏥 healthcare, or 💼 hiring? ⚠️ Think again! ⚠️ 🚨 We found that LoRA can amplify #AI #harms: ❗️False sense of #safety #alignment 🚫Increased #unfairness and #bias, hitting minority groups the hardest

English
4
3
39
4K

Keşfet

@NeurIPSConf @luislamb @deedydas @ELLISforEurope @icmlconf @flavioAd @auai @looselycorrect