Lightnews — Scholar-powered news

Giuseppe Attanasio @gattanasio.cc · Sep 5

There's too little political debate in my BSky feed.

I really really really really hope the Flotilla will succeed in its goals. I wish I would be able to do more for the cause 🍉

1

Reposted by Giuseppe Attanasio

Global Movement To Gaza UK @gmtguk.bsky.social · Aug 31

Uniting 44 nations, the Global Sumud Flotilla sails to support Gaza and challenge injustice.

Let’s make our voices heard.

#GlobalSolidarity #GlobalSumudFlotilla #StandWithGaza#SailToGaza #BreakTheSiege #GlobalMovementToGaza

25 37

Reposted by Giuseppe Attanasio

Gender Bias in NLP Workshop @genderbiasnlp.bsky.social · Aug 1

🔴 LIVE NOW: The first keynote of the Gender Bias in NLP workshop!

@a-lauscher.bsky.social is telling A Fairy Tale of Gender in Language Technology.

#ACL2025 #GeBNLP #NLP

4 10

Giuseppe Attanasio @gattanasio.cc · Jul 29

👀 Interested in gender bias in NLP and MT evaluation?

I strongly encourage you to check out our poster presentations held by @manoszar.bsky.social!

👉🏻 Poster Session 3, today (July 29) from 16:00–17:30

OR at the Gender Bias Workshop, Friday (Aug 1) during Poster Session 2, 14:00–14:45.

#acl2025

3

Giuseppe Attanasio @gattanasio.cc · Jul 15

The first author is here! @manoszar.bsky.social 🙌

Giuseppe Attanasio @gattanasio.cc · Jul 14

‼️ Results are consistent across different SOTA metrics, languages, source types, and gender ambiguous vs. non-ambiguous human entities.

Joint work w/ Emmanouil Zaranis, @swetaagrawal.bsky.social, and @andre-t-martins.bsky.social

1

Giuseppe Attanasio @gattanasio.cc · Jul 14

As neural metrics are a pillar for #MT, being extensively used for evaluation but also improving translation, we'd want them to be fair.

🚨 Our #ACL2025 paper shows they consistently, unduly favor masculine-inflected translations, or gendered forms, over neutral ones.

arxiv.org/pdf/2410.10995

Figure 1 of the paper shows that the metrics score the masculine forms higher than the feminine or neutral ones.

1 8 11

Giuseppe Attanasio @gattanasio.cc · Jul 7

Just wanted to mention that besides me rattling on about speech, social biases, and NLP in the coziest of environments, that TV starred in a blazing karaoke night just a few hours earlier 🎙️

Thanks again for the invitation @milanlp.bsky.social :)

MilaNLP Lab @milanlp.bsky.social · Jul 7

Last week we held our 1st MilaNLP retreat by beautiful Lago Maggiore! ⛰️🌊
We shared research ideas, stories (academic & beyond), and amazing food. It was a great time to connect outside of the usual lab working days, and most importantly, strengthen our bonds as a team. #ResearchLife #NLProc

2 9

Reposted by Giuseppe Attanasio

Ana Marasović @anamarasovic.bsky.social · Jul 4

After all these years of publishing, still surprised by how much the luck of drawing meaningfully engaged reviewers and an AC, matters in peer review

1 20

Reposted by Giuseppe Attanasio

GITT 2025 @gitt-workshop.bsky.social · Jun 23

Good morning @mtsummit2025.bsky.social ! 🌞 We're ready to kick off #GITT2025 in a few minutes. We'll be live skywriting (is that the tweeting equivalent of 'tweeting'? It is now) using our hashtag. Looking forward! 🤩

1 5 7

Giuseppe Attanasio @gattanasio.cc · Jun 19

Extremely sorry that it happened. Grateful you decided to share it publicly. Hopeful it's reversible and, if not, that it'll steer a conv around unsubstantiated reasons to prevent scientists do science

1 1

Giuseppe Attanasio @gattanasio.cc · Jun 16

(totally unbiased opinion): Cool paper!

MilaNLP Lab @milanlp.bsky.social · Jun 16

#MemoryMonday #NLProc 'Interpretability Informs Gender Bias Mitigation in Machine Translation'- Attanasio et al. (2023). Their new research tackles gender bias in machine translations using a simple, few-shot learning method for fairness.

A Tale of Pronouns: Interpretability Informs Gender Bias Mitigation for Fairer Instruction-Tuned Machine Translation

Giuseppe Attanasio, Flor Miriam Plaza del Arco, Debora Nozza, Anne Lauscher. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.

aclanthology.org

4 7

Giuseppe Attanasio @gattanasio.cc · Jun 9

Stiamo facendo questa cosa... Una mano è sempre ben accetta :) 🙏

AILC-NLP @ailc-nlp.bsky.social · Jun 9

Stiamo vivendo una rivoluzione tecnologica, ma quanto conosciamo realmente l'uso dell'IA nella nostra quotidianità? Aiuta a scoprirlo compilando questo breve questionario (10 min): bit.ly/sondaggio_ai...

1

Reposted by Giuseppe Attanasio

Adam Sternbergh @sternbergh.bsky.social · Jun 5

must be incredibly frustrating and disheartening to have federal funding that was promised to you for important work suddenly and arbitrarily ripped away

Best of Dying Twitter @bestofdyingtwit.bsky.social · Jun 5

“In light of the President's statement about cancellation of my government contracts, @SpaceX will begin decommissioning its Dragon spacecraft immediately”

620 8.2K 36K

Reposted by Giuseppe Attanasio

Beatrice Savoldi @bsavoldi.bsky.social · Jun 3

🔍 Stiamo studiando come l'AI viene usata in Italia e per farlo abbiamo costruito un sondaggio!

👉 bit.ly/sondaggio_ai...

(è anonimo, richiede ~10 minuti, e se partecipi o lo fai girare ci aiuti un sacco🙏)

Ci interessa anche raggiungere persone che non si occupano e non sono esperte di AI!

Qualtrics Survey | Qualtrics Experience Management

The most powerful, simple and trusted way to gather experience data. Start your journey to experience management and try a free account today.

bit.ly

1 18 16

Reposted by Giuseppe Attanasio

Tiago Pimentel @tpimentel.bsky.social · May 29

If you're finishing your camera-ready for ACL or ICML and want to cite co-first authors more fairly, I just made a simple fix to do this! Just add $^*$ to the authors' names in your bibtex, and the citations should change :)

github.com/tpimentelms/...

Inline citations with only first author name, or first two co-first author names.

4 23 85

Reposted by Giuseppe Attanasio

David Bau @davidbau.bsky.social · May 28

I am in Boston because I believe in American democracy. I love our freedoms and our culture. I want to be teaching US students.

And I want to live and teach here in my home, where I grew up.

Why are we setting our home on fire?

www.theguardian.com/us-news/202...

Trump administration orders US embassies to stop student visa interviews

Directive could severely delay visa processing and hurt universities that rely on foreign students for revenue

www.theguardian.com

1 2 5

Reposted by Giuseppe Attanasio

Dennis Fucci @dennisfucci.bsky.social · May 19

More great news! 🎉
Our paper “Echoes of Phonetics: Unveiling Relevant Acoustic Cues for ASR via Feature Attribution” has been accepted at #Interspeech2025!

If you’re interested in interpretability for speech models, stay tuned, the preprint is coming soon!

1 1 3

Giuseppe Attanasio @gattanasio.cc · May 17

Yup, on Mac. But that requires interacting with Overleaf's UI, dunno if it's doable easily. Same reason why I wouldn't like a browser extension over... Something coming from @overleaf.com themselves :)

1

Reposted by Giuseppe Attanasio

Margaret Mitchell @mmitchell.bsky.social · May 16

Some of my newer work with many found that GenAI models (like Claude) create fake articles to justify discrimination. Fabricating articles for law isn't surprising...so why use them that way? =/
techcrunch.com/2025/05/15/a...
(Real) article on our work: www.technologyreview.com/2025/04/30/1...

Anthropic’s lawyer was forced to apologize after Claude hallucinated a legal citation | TechCrunch

A lawyer representing Anthropic used Claude to generate citations in a court filing, then it hallucinated.

techcrunch.com

4 17

Giuseppe Attanasio @gattanasio.cc · May 16

Wish I had a single @overleaf.com keyboard shortcut that would 1) append whatever I have in my clipboard to my bib file, 2) parse it, 3) extract the bibkey, and 4) paste it at my current cursor location in the editor.

3 3 26

Reposted by Giuseppe Attanasio

Anne Lauscher @a-lauscher.bsky.social · May 16

If you are interested in the relationship between what is happening INSIDE 🔍 of our models 🤖 and OUTSIDE 🏞️ of our models 🤖, consider submitting to our workshop at COLM. #LLM #NLProc

INTERPLAY Workshop@COLM '25 @interplay-workshop.bsky.social · May 16

🚨🚨 Studying the INTERPLAY of LMs' internals and behavior?

Join our
@colmweb.org
workshop on comprehensivly evaluating LMs.

Deadline: June 23rd
CfP: shorturl.at/sBomu
Page: shorturl.at/FT3fX

We're excited to see your insights and methods!!

See you in Montréal 🇨🇦 #nlproc #interpretability

Call for Papers, Interplay Workshop at COLM: June 23rd - submissions due. July 24th - acceptance notification. October 10th - workshop day.

2 8

Reposted by Giuseppe Attanasio

Gabriele Sarti @gsarti.com · May 15

BlackboxNLP is back! 💥

Happy to be part of the organizing team for this year, and super excited for our new shared task using the excellent MIB Benchmark, check it out! blackboxnlp.github.io/2025/task/

BlackboxNLP @blackboxnlp.bsky.social · May 15

BlackboxNLP, the leading workshop on interpretability and analysis of language models, will be co-located with EMNLP 2025 in Suzhou this November! 📆

This edition will feature a new shared task on circuits/causal variable localization in LMs, details here: blackboxnlp.github.io/2025/task

2 6

Giuseppe Attanasio @gattanasio.cc · May 6

Not related to debiasing techniques per se, but we have been pondering how *actionable* (if you want, used in practice/yielding actual change) bias metrics are -- and what it takes them to be :)
aclanthology.org/2024.emnlp-m...

Metrics for What, Metrics for Whom: Assessing Actionability of Bias Evaluation Metrics in NLP

Pieter Delobelle, Giuseppe Attanasio, Debora Nozza, Su Lin Blodgett, Zeerak Talat. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

aclanthology.org

3

Reposted by Giuseppe Attanasio

Randall Munroe @xkcd.com · Apr 25

PhD Timeline xkcd.com/3081

600 21K 60K