MilaNLP Lab
@milanlp.bsky.social
📤 447
📥 178
📝 175
The Milan Natural Language Processing Group
#NLProc
#AI
milanlproc.github.io
Another exhausting day in the lab… conducting very rigorous panettone analysis. Pandoro was evaluated too, because we believe in fair experimental design.
about 18 hours ago
0
19
7
#TBT
#NLProc
'@donyarn.bsky.social &
@dirkhovy.bsky.social
's 2024 paper, 'Conversations as a Source for Teaching Scientific Concepts' turns video dialogues into effective teaching tools.'
loading . . .
https://arxiv.org/pdf/2404.10475
about 18 hours ago
0
3
2
We're happy to have
@veraneplenbroek.bsky.social
at our lab this week! She presented her
#EMNLP2025
work "Reading Between the Prompts: How Stereotypes Shape LLM's Implicit Personalization" and shared more of her exciting ongoing work.
#NLProc
2 days ago
0
11
3
#MemoryModay
#NLProc
'Hey Siri. Ok Google. Alexa: A topic modeling of user reviews for smart speakers,' by Nguyen &
@dirkhovy.bsky.social
decodes speaker reviews for user preferences using topic models. Domain knowledge needed for market analysis.
loading . . .
Hey Siri. Ok Google. Alexa: A topic modeling of user reviews for smart speakers
Hanh Nguyen, Dirk Hovy. Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). 2019.
https://aclanthology.org/D19-5510
4 days ago
0
3
2
What an inspiring week at
#EMNLP2025
in Suzhou🇨🇳! Huge thanks to the organizers and everyone who stopped by our poster/talk!
4 days ago
1
17
5
For our weekly lab seminar, it was a pleasure to have
@andersgiovanni.com
presenting his research "How AI Affects Us: Controlled Experiments in Human-AI Interaction".
#NLProc
7 days ago
0
9
2
#TBT
#NLProc
' Attanasio et al. study asks 'Is It Worth the (Environmental) Cost?' analyzing continuous training for language models. Balances benefits, environmental impacts, for responsible use.
#Sustainability
'
loading . . .
https://arxiv.org/pdf/2210.07365
8 days ago
0
3
3
For our weekly reading group,
@joachimbaumann.bsky.social
presented the upcoming PNAS article "The potential existential threat of large language models to online survey research" by @
@seanjwestwood.bsky.social
.
8 days ago
0
8
3
#MemoryModay
#NLProc
' 'State of Profanity Obfuscation in NLP Scientific Publications' probes bias in non-English papers.
@deboranozza.bsky.social
&
@dirkhovy.bsky.social
(2023) propose 'PrOf' to aid authors & improve access.
loading . . .
The State of Profanity Obfuscation in Natural Language Processing Scientific Publications
Debora Nozza, Dirk Hovy. Findings of the Association for Computational Linguistics: ACL 2023. 2023.
https://aclanthology.org/2023.findings-acl.240
11 days ago
0
4
2
#TBT
#NLProc
Hessenthaler et al.'s 2022 work delves into AI's link with fairness & energy reduction in English NLP models, challenging bias reduction theories.
#AI
#sustainability
loading . . .
Bridging Fairness and Environmental Sustainability in Natural Language Processing
Marius Hessenthaler, Emma Strubell, Dirk Hovy, Anne Lauscher. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. 2022.
https://aclanthology.org/2022.emnlp-main.533
15 days ago
0
5
2
#MemoryModay
#NLProc
'Measuring Harmful Representations in Scandinavian Language Models' uncovers gender bias, challenging Scandinavia's equity image.
loading . . .
Measuring Harmful Representations in Scandinavian Language Models
Samia Touileb, Debora Nozza. Proceedings of the Fifth Workshop on Natural Language Processing and Computational Social Science (NLP+CSS). 2022.
https://aclanthology.org/2022.nlpcss-1.13
18 days ago
0
4
2
#TBT
#NLProc
"Explaining Speech Classification Models" by Pastor et al. (2024) makes speech classification more transparent! 🔍 Their research reveals which words matter most and how tone and background noise impact decisions.
loading . . .
Explaining Speech Classification Models via Word-Level Audio Segments and Paralinguistic Features
Eliana Pastor, Alkis Koudounas, Giuseppe Attanasio, Dirk Hovy, Elena Baralis. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long...
https://aclanthology.org/2024.eacl-long.136
22 days ago
0
4
2
reposted by
MilaNLP Lab
Arianna Muti
23 days ago
LLMs require social knowledge to understand implicit misogyny, yet they mostly fail. If you want to know more, come check my poster from 12.30 to 13.30! Paper:
aclanthology.org/2025.finding...
#EMNLP2025
add a skeleton here at some point
0
6
2
reposted by
MilaNLP Lab
Debora Nozza
23 days ago
Feeling a little sad not to be in Suzhou for
#EMNLP2025
, but so proud of all the amazing work from our MilaNLP Lab! 💫 Honored to have received the Outstanding Senior Area Chair Award! Check out our papers 👇
add a skeleton here at some point
0
10
2
#MemoryModay
#NLProc
'Universal Joy: A Data Set and Results for Classifying Emotions Across Languages' by Lamprinidis et al. (2021) explores how AI research affects our planet.
loading . . .
Universal Joy A Data Set and Results for Classifying Emotions Across Languages
Sotiris Lamprinidis, Federico Bianchi, Daniel Hardt, Dirk Hovy. Proceedings of the Eleventh Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis. 2021.
https://aclanthology.org/2021.wassa-1.7
25 days ago
0
6
2
For our weekly lab seminar it was a pleasure to have Valerio Capraro talking about The Economics of Language.
#NLProc
28 days ago
0
4
1
Proud to present our
#EMNLP2025
papers! Catch our team across Main, Findings, Workshops & Demos 👇
28 days ago
12
11
6
#TBT
#NLProc
Explore 'Wisdom of Instruction-Tuned LLM Crowds' by Plaza et al. LLM labels outperform single models in tasks & languages. But few-shot can't top zero-shot. Supervised models rule.
loading . . .
Wisdom of Instruction-Tuned Language Model Crowds. Exploring Model Label Variation
Flor Miriam Plaza-del-Arco, Debora Nozza, Dirk Hovy. Proceedings of the 3rd Workshop on Perspectivist Approaches to NLP (NLPerspectives) @ LREC-COLING 2024. 2024.
https://aclanthology.org/2024.nlperspectives-1.2
29 days ago
0
2
2
Great session today in our lab reading group. Thanks to Emanuele Moscato for presenting the article “Universities are embracing AI: will students get smarter or stop thinking?” from
@naturemagazine.bsky.social
. Article:
www.nature.com/articles/d41...
#NLProc
29 days ago
0
4
1
reposted by
MilaNLP Lab
Paul Röttger @ EMNLP
about 1 month ago
LLMs are good at simulating human behaviours, but they are not going to be great unless we train them to. We hope SimBench can be the foundation for more specialised development of LLM simulators. I really enjoyed working on this with
@tiancheng.bsky.social
et al. Many fun results 👇
add a skeleton here at some point
0
8
3
reposted by
MilaNLP Lab
Paul Röttger @ EMNLP
30 days ago
There’s plenty of evidence for political bias in LLMs, but very few evals reflect realistic LLM use cases — which is where bias actually matters. IssueBench, our attempt to fix this, is accepted at TACL, and I will be at
#EMNLP2025
next week to talk about it! New results đź§µ
add a skeleton here at some point
1
32
11
For our last Thursday Reading Group,
@taniseceron.bsky.social
presented "Helpful, harmless, honest? Sociotechnical limits of AI alignment and safety through Reinforcement Learning from Human Feedback" by A. D. Lindström et al. (2025) Paper:
link.springer.com/article/10.1...
#NLProc
about 1 month ago
0
5
1
#MemoryModay
#NLProc
'Dense Node Representation for Geolocation' by Fornaciari &
@dirkhovy.bsky.social
reveals efficient geolocation methods using node2vec & doc2vec models. Greater network size, less parameters.
loading . . .
Dense Node Representation for Geolocation
Tommaso Fornaciari, Dirk Hovy. Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). 2019.
https://aclanthology.org/D19-5529
about 1 month ago
0
3
2
reposted by
MilaNLP Lab
Debora Nozza
about 1 month ago
Over the past two days, I participated in the
@erc.europa.eu
Workshop on Data Access under DSA Article 40. An enriching experience that deepened my understanding of the DSA's implications for research and enabled me to connect with exceptional media researchers.
erc.europa.eu/news-events/...
loading . . .
ERC Workshop on data access under the Digital Services Act (DSA) Article 40 (opening session)
The Digital Services Act (DSA) is an European legislation that specifies a set of rules to make the digital space safer and more trustworthy for users.
https://erc.europa.eu/news-events/events/erc-workshop-data-access-under-digital-services-act-dsa-article
0
11
4
#TBT
#NLProc
'Classist Tools: Social Class Correlates with Performance in NLP' by Curry et al. (2024) explores AI's hidden energy problem, and how machine learning impacts environmental sustainability.
about 1 month ago
0
4
1
reposted by
MilaNLP Lab
WASSA 2026
about 1 month ago
🚀 We are pleased to announce the First Call for Papers for
#WASSA2026
This year, we introduce a Special Track on Multilinguality and Social Bridges between High- & Lesser-Resourced Languages/Communities. 🌍 🗓️ Deadlines: Dec 17 (direct) and Jan 2 (ARR). 🔗
workshop-wassa.github.io/2026/call-fo...
loading . . .
Call for Papers
Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis
https://workshop-wassa.github.io/2026/call-for-papers/
0
3
4
#MemoryModay
#NLProc
'Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection' - Attanasio et al. Explores reliability of interpretability in hate speech detection.
loading . . .
Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection
Giuseppe Attanasio, Debora Nozza, Eliana Pastor, Dirk Hovy. Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP. 2022.
https://aclanthology.org/2022.nlppower-1.11
about 1 month ago
0
4
3
It was a pleasure to have
@rochellechoenni.bsky.social
presenting "Brittle but Steerable: Aligning Cultural Values in Multilingual Language Models" at our lab seminar.
#NLProc
about 1 month ago
0
7
1
We’re delighted to welcome Eve Fleisig to our
@milanlp.bsky.social
lab as a visiting PhD student! ✨
about 1 month ago
0
8
1
#TBT
#NLProc
'Geolocation with Attention-Based Multitask Learning Models' by Tommaso Fornaciari,
@dirkhovy.bsky.social
(2019) reveals how online political talks can become one-sided.
#SocialMedia
loading . . .
Geolocation with Attention-Based Multitask Learning Models
Tommaso Fornaciari, Dirk Hovy. Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). 2019.
https://aclanthology.org/D19-5528
about 1 month ago
0
4
2
#MemoryModay
#NLProc
' 'State of Profanity Obfuscation in NLP Scientific Publications' probes bias in non-English papers. @debora_nozza & Dirk Hovy (2023) propose 'PrOf' to aid authors & improve access.
aclanthology.org/2023.finding...
loading . . .
The State of Profanity Obfuscation in Natural Language Processing Scientific Publications
Debora Nozza, Dirk Hovy. Findings of the Association for Computational Linguistics: ACL 2023. 2023.
https://aclanthology.org/2023.findings-acl.240
about 2 months ago
0
4
2
Thanks Clara Meister for presenting "Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization" at our lab seminar.
#NLProc
#tokenization
#fairness
about 2 months ago
0
5
1
#TBT
#NLProc
"Explaining Speech Classification Models" by Pastor et al. (2024) makes speech classification more transparent! 🔍 Their research reveals which words matter most and how tone and background noise impact decisions.
loading . . .
Explaining Speech Classification Models via Word-Level Audio Segments and Paralinguistic Features
Eliana Pastor, Alkis Koudounas, Giuseppe Attanasio, Dirk Hovy, Elena Baralis. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long...
https://aclanthology.org/2024.eacl-long.136
about 2 months ago
0
6
3
📚 For today’s reading group
@arimuti.bsky.social
presented Emergent Misalignment: Narrow Finetuning Can Produce Broadly Misaligned LLMs (Betley et al., 2025). đź§©
arxiv.org/abs/2502.17424
#NLProc
#AIAlignment
#LLMs
about 2 months ago
0
10
2
reposted by
MilaNLP Lab
Flor Plaza
about 2 months ago
📢 Are you interested in a PhD in
#NLProc
to study and improve how AI model emotions and social signals? 🚨Exciting news:🚨 I’m hiring a PhD candidate at LIACS,
@unileiden.bsky.social
. 📍 Leiden, The Netherlands 📅 Deadline: 17 Nov 2025 👉 Position details and application link:
tinyurl.com/5x5v6zsa
loading . . .
PhD Candidate in Emotionally and Socially Aware Natural Language Processing
The Faculty of Science and the Leiden Institute of Advanced Computer Science (LIACS) are looking for a:PhD Candidate in Emotionally and Socially Aware Natural Language Processing (1.0fte)Project descr...
https://tinyurl.com/5x5v6zsa
0
9
9
#MemoryModay
#NLProc
'Measuring Harmful Representations in Scandinavian Language Models' uncovers gender bias, challenging Scandinavia's equity image.
#MachineLearning
loading . . .
Measuring Harmful Representations in Scandinavian Language Models
Samia Touileb, Debora Nozza. Proceedings of the Fifth Workshop on Natural Language Processing and Computational Social Science (NLP+CSS). 2022.
https://aclanthology.org/2022.nlpcss-1.13
about 2 months ago
0
3
2
#TBT
#NLProc
Explore 'Wisdom of Instruction-Tuned LLM Crowds' by Plaza et al. LLM labels outperform single models in tasks & languages. But few-shot can't top zero-shot. Supervised models rule.
loading . . .
Wisdom of Instruction-Tuned Language Model Crowds. Exploring Model Label Variation
Flor Miriam Plaza-del-Arco, Debora Nozza, Dirk Hovy. Proceedings of the 3rd Workshop on Perspectivist Approaches to NLP (NLPerspectives) @ LREC-COLING 2024. 2024.
https://aclanthology.org/2024.nlperspectives-1.2
about 2 months ago
0
2
2
reposted by
MilaNLP Lab
Tanise Ceron
about 2 months ago
📣 New Preprint! Have you ever wondered what the political content in LLM's training data is? What are the political opinions expressed? What is the proportion of left- vs right-leaning documents in the pre- and post-training data? Do they correlate with the political biases reflected in models?
2
46
14
#MemoryModay
#NLProc
'Universal Joy: A Data Set and Results for Classifying Emotions Across Languages' by Lamprinidis et al. (2021) explores how AI research affects our planet. Tech can be green too!
#SustainableTech
loading . . .
Universal Joy A Data Set and Results for Classifying Emotions Across Languages
Sotiris Lamprinidis, Federico Bianchi, Daniel Hardt, Dirk Hovy. Proceedings of the Eleventh Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis. 2021.
https://aclanthology.org/2021.wassa-1.7
about 2 months ago
0
3
2
đź“–For our last Reading Group
@donyarn.bsky.social
presented "Culture is Everywhere: A Call for Intentionally Cultural Evaluation" by Oh et al. Paper:
arxiv.org/pdf/2509.01301
#NLProc
2 months ago
0
6
2
MilaNLPers at CLiC-it 2025 presenting "Probing Feminist Representations: A Study of Bias in LLMs and Word Embeddings" Check the paper at
clic2025.unica.it/wp-content/u...
#NLProc
#clicit25
2 months ago
0
17
4
#TBT
#NLProc
'Classist Tools: Social Class Correlates with Performance in NLP' by Curry et al. (2024) explores AI's hidden energy problem, and how machine learning impacts environmental sustainability. Tech can be green!
#CleanTech
loading . . .
https://arxiv.org/pdf/2403.04445
2 months ago
0
3
2
What makes LLMs agree even when they shouldn’t? 🤔 At our last seminar’s lab, Jan Batzner presented The Brief History of LLM Sycophancy as We Know It.
#NLProc
#sycophancy
#LLM
2 months ago
0
9
2
#MemoryModay
#NLProc
'Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection' - Attanasio et al. Explores reliability of interpretability in hate speech detection.
loading . . .
Benchmarking Post-Hoc Interpretability Approaches for Transformer-based Misogyny Detection
Giuseppe Attanasio, Debora Nozza, Eliana Pastor, Dirk Hovy. Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP. 2022.
https://aclanthology.org/2022.nlppower-1.11
2 months ago
0
4
2
#TBT
#NLProc
'Geolocation with Attention-Based Multitask Learning Models' by Tommaso Fornaciari,
@dirkhovy.bsky.social
(2019) reveals how online political talks can become one-sided. Breaking out of our bubbles!
#SocialMedia
loading . . .
Geolocation with Attention-Based Multitask Learning Models
Tommaso Fornaciari, Dirk Hovy. Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). 2019.
https://aclanthology.org/D19-5528
2 months ago
0
4
2
đź“–For this Thursday's Reading Group
@elisabassignana.bsky.social
presented "The sociolinguistic foundations of language modeling" by Grieve et al. Paper:
www.frontiersin.org/journals/art...
#NLProc
#LLM
#sociolinguistics
2 months ago
0
11
1
#MemoryModay
#NLProc
'Dense Node Representation for Geolocation' by Fornaciari &
@dirkhovy.bsky.social
reveals efficient geolocation methods using node2vec & doc2vec models. Greater network size, less parameters. /publication/2019_m2v/2019_m2v
loading . . .
Dense Node Representation for Geolocation
Tommaso Fornaciari, Dirk Hovy. Proceedings of the 5th Workshop on Noisy User-generated Text (W-NUT 2019). 2019.
https://aclanthology.org/D19-5529
2 months ago
0
3
2
🎓We're back with Fridays' lab seminars! Today we had Suyash Fulay presenting "Truth, Political Bias, and AI Representation".
#NLProc
3 months ago
0
5
2
#TBT
#NLProc
'MilaNLP @ WASSA: Does BERT Feel Sad When You Cry?' by Fornaciari et al. (2021) indicates that emotion and empathy are not related tasks for prediction.
loading . . .
https://aclanthology.org/2021.wassa-1.29
3 months ago
0
2
2
đź“–For this Thursday's Reading Group
@deboranozza.bsky.social
presented two papers on
#sycophancy
in
#LLMs
. Papers:
arxiv.org/pdf/2505.13995
,
arxiv.org/pdf/2310.13548
#NLProc
3 months ago
0
6
2
Load more
feeds!
log in