MilaNLP Lab
@milanlp.bsky.social
The Milan Natural Language Processing Group #NLProc #AI
milanlproc.github.io
milanlproc.github.io
#TBT #NLProc
'SAFETYKIT: Measuring Safety in Open-domain Conversational Systems' by Dinan et al. (2022) introduces taxonomy for AI safety, assesses tools' limits.
#AIsafety
'SAFETYKIT: Measuring Safety in Open-domain Conversational Systems' by Dinan et al. (2022) introduces taxonomy for AI safety, assesses tools' limits.
#AIsafety
aclanthology.org
January 29, 2026 at 4:22 PM
For our weekly reading group, Afshin Karimi presented "What Social Media Use Do People Regret? An Analysis of 34K Smartphone Screenshots with Multimodal LLM". Super interesting discussion!
Paper: dl.acm.org/doi/10.1145/...
#NLProc
Paper: dl.acm.org/doi/10.1145/...
#NLProc
January 29, 2026 at 3:09 PM
For our weekly reading group, Afshin Karimi presented "What Social Media Use Do People Regret? An Analysis of 34K Smartphone Screenshots with Multimodal LLM". Super interesting discussion!
Paper: dl.acm.org/doi/10.1145/...
#NLProc
Paper: dl.acm.org/doi/10.1145/...
#NLProc
Reposted by MilaNLP Lab
New year, new job? If that is your current mantra, check the open postdoc positions with Debora Nozza and me at our lab. Deadline is January 31st.
milanlproc.github.io/open_positio...
milanlproc.github.io/open_positio...
Postdoctoral Researcher – NLP (2 positions) | MilaNLP Lab @ Bocconi University
Two Postdoctoral Researcher positions – Deadline January 31st, 2026
milanlproc.github.io
January 19, 2026 at 4:13 PM
New year, new job? If that is your current mantra, check the open postdoc positions with Debora Nozza and me at our lab. Deadline is January 31st.
milanlproc.github.io/open_positio...
milanlproc.github.io/open_positio...
#MemoryModay #NLProc
@gattanasio.cc et al. study asks 'Is It Worth the (Environmental) Cost?' analyzing continuous training for language models. Balances benefits, environmental impacts, for responsible use. #AI #Sustainability
arxiv.org/pdf/2210.07365
@gattanasio.cc et al. study asks 'Is It Worth the (Environmental) Cost?' analyzing continuous training for language models. Balances benefits, environmental impacts, for responsible use. #AI #Sustainability
arxiv.org/pdf/2210.07365
arxiv.org
January 26, 2026 at 5:10 PM
#MemoryModay #NLProc
@gattanasio.cc et al. study asks 'Is It Worth the (Environmental) Cost?' analyzing continuous training for language models. Balances benefits, environmental impacts, for responsible use. #AI #Sustainability
arxiv.org/pdf/2210.07365
@gattanasio.cc et al. study asks 'Is It Worth the (Environmental) Cost?' analyzing continuous training for language models. Balances benefits, environmental impacts, for responsible use. #AI #Sustainability
arxiv.org/pdf/2210.07365
Reposted by MilaNLP Lab
The deadline is approaching! Join the team :)
⏳ Deadline approaching! We’re hiring 2 fully funded postdocs in #NLP.
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
January 20, 2026 at 10:27 AM
The deadline is approaching! Join the team :)
Reposted by MilaNLP Lab
🚨(Software) Update:
In my PhD, I had a side project to fix an annoying problem: when you ask 5 people to label the same thing, you often get different answers. But in ML (and lots of other analyses), you still need a single aggregated answer. Using the majority vote is easy–but often wrong.
1/N
In my PhD, I had a side project to fix an annoying problem: when you ask 5 people to label the same thing, you often get different answers. But in ML (and lots of other analyses), you still need a single aggregated answer. Using the majority vote is easy–but often wrong.
1/N
GitHub - dirkhovy/MACE: Multi-Annotator Competence Estimation tool
Multi-Annotator Competence Estimation tool. Contribute to dirkhovy/MACE development by creating an account on GitHub.
github.com
January 20, 2026 at 10:12 AM
🚨(Software) Update:
In my PhD, I had a side project to fix an annoying problem: when you ask 5 people to label the same thing, you often get different answers. But in ML (and lots of other analyses), you still need a single aggregated answer. Using the majority vote is easy–but often wrong.
1/N
In my PhD, I had a side project to fix an annoying problem: when you ask 5 people to label the same thing, you often get different answers. But in ML (and lots of other analyses), you still need a single aggregated answer. Using the majority vote is easy–but often wrong.
1/N
Reposted by MilaNLP Lab
However, disagreement isn’t just noise—it’s information. It can mean an item is genuinely hard—or someone wasn’t paying attention. If only you knew whom to trust…
That summer, Taylor Berg-Kirkpatrick, Ashish Vaswani, and I built MACE (Multi-Annotator Competence Estimation).
2/N
That summer, Taylor Berg-Kirkpatrick, Ashish Vaswani, and I built MACE (Multi-Annotator Competence Estimation).
2/N
January 20, 2026 at 10:14 AM
However, disagreement isn’t just noise—it’s information. It can mean an item is genuinely hard—or someone wasn’t paying attention. If only you knew whom to trust…
That summer, Taylor Berg-Kirkpatrick, Ashish Vaswani, and I built MACE (Multi-Annotator Competence Estimation).
2/N
That summer, Taylor Berg-Kirkpatrick, Ashish Vaswani, and I built MACE (Multi-Annotator Competence Estimation).
2/N
Reposted by MilaNLP Lab
MACE estimates:
1. Annotator reliability (who’s consistent?)
2. Item difficulty (which examples spark disagreement?)
3. The most likely aggregate label (the latent “best guess”)
That “side project” ended up powering hundreds of annotation projects over the years.
3/N
1. Annotator reliability (who’s consistent?)
2. Item difficulty (which examples spark disagreement?)
3. The most likely aggregate label (the latent “best guess”)
That “side project” ended up powering hundreds of annotation projects over the years.
3/N
January 20, 2026 at 10:15 AM
MACE estimates:
1. Annotator reliability (who’s consistent?)
2. Item difficulty (which examples spark disagreement?)
3. The most likely aggregate label (the latent “best guess”)
That “side project” ended up powering hundreds of annotation projects over the years.
3/N
1. Annotator reliability (who’s consistent?)
2. Item difficulty (which examples spark disagreement?)
3. The most likely aggregate label (the latent “best guess”)
That “side project” ended up powering hundreds of annotation projects over the years.
3/N
Reposted by MilaNLP Lab
I always wanted to revisit it, port it from Java to Python & extend to continuous data, but never found the time.
Last week, I played around with Cursor – and got it all done in ~1 hour. 🤯
If you work with any response data that needs aggregation, give it a try—and let me know what you think!
4/N
Last week, I played around with Cursor – and got it all done in ~1 hour. 🤯
If you work with any response data that needs aggregation, give it a try—and let me know what you think!
4/N
January 20, 2026 at 10:17 AM
I always wanted to revisit it, port it from Java to Python & extend to continuous data, but never found the time.
Last week, I played around with Cursor – and got it all done in ~1 hour. 🤯
If you work with any response data that needs aggregation, give it a try—and let me know what you think!
4/N
Last week, I played around with Cursor – and got it all done in ~1 hour. 🤯
If you work with any response data that needs aggregation, give it a try—and let me know what you think!
4/N
Reposted by MilaNLP Lab
If you are curious about the theoretical background, see
Hovy, D., Berg-Kirkpatrick, T., Vaswani, A., & Hovy E. (2013). Learning Whom to Trust With MACE. In: Proceedings of NAACL-HLT. ACL.
aclanthology.org/N13-1132.pdf
And for even more details:
aclanthology.org/Q18-1040.pdf
N/N
Hovy, D., Berg-Kirkpatrick, T., Vaswani, A., & Hovy E. (2013). Learning Whom to Trust With MACE. In: Proceedings of NAACL-HLT. ACL.
aclanthology.org/N13-1132.pdf
And for even more details:
aclanthology.org/Q18-1040.pdf
N/N
aclanthology.org
January 20, 2026 at 10:20 AM
If you are curious about the theoretical background, see
Hovy, D., Berg-Kirkpatrick, T., Vaswani, A., & Hovy E. (2013). Learning Whom to Trust With MACE. In: Proceedings of NAACL-HLT. ACL.
aclanthology.org/N13-1132.pdf
And for even more details:
aclanthology.org/Q18-1040.pdf
N/N
Hovy, D., Berg-Kirkpatrick, T., Vaswani, A., & Hovy E. (2013). Learning Whom to Trust With MACE. In: Proceedings of NAACL-HLT. ACL.
aclanthology.org/N13-1132.pdf
And for even more details:
aclanthology.org/Q18-1040.pdf
N/N
Reposted by MilaNLP Lab
I included some test cases on GitHub, will look if I still have the ones we used in the paper.
January 20, 2026 at 11:11 AM
I included some test cases on GitHub, will look if I still have the ones we used in the paper.
Reposted by MilaNLP Lab
Found and added under data/
January 20, 2026 at 11:21 AM
Found and added under data/
Reposted by MilaNLP Lab
🚀 We’re opening 2 fully funded postdoc positions in #NLP!
Join the MilaNLP team and contribute to our upcoming research projects.
🔗 More details: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects.
🔗 More details: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
December 18, 2025 at 3:29 PM
🚀 We’re opening 2 fully funded postdoc positions in #NLP!
Join the MilaNLP team and contribute to our upcoming research projects.
🔗 More details: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects.
🔗 More details: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Reposted by MilaNLP Lab
For today's reading group, Serena Pugliese presented the paper “Adversarial Poetry as a Universal Single-Turn Jailbreak Mechanism in Large Language Models" by Piercosma Bisconti et al. (2025).
Paper: arxiv.org/pdf/2511.15304
#NLProc
#LLMs #jailbreaking
Paper: arxiv.org/pdf/2511.15304
#NLProc
#LLMs #jailbreaking
December 18, 2025 at 11:32 AM
For today's reading group, Serena Pugliese presented the paper “Adversarial Poetry as a Universal Single-Turn Jailbreak Mechanism in Large Language Models" by Piercosma Bisconti et al. (2025).
Paper: arxiv.org/pdf/2511.15304
#NLProc
#LLMs #jailbreaking
Paper: arxiv.org/pdf/2511.15304
#NLProc
#LLMs #jailbreaking
Reposted by MilaNLP Lab
We're also back with the lab's seminar! Today we had Eleonora Mancini presenting her doctoral research "Multimodal AI for Human Expression Understanding".
#NLP #multimodality #speech
#NLP #multimodality #speech
January 16, 2026 at 4:47 PM
We're also back with the lab's seminar! Today we had Eleonora Mancini presenting her doctoral research "Multimodal AI for Human Expression Understanding".
#NLP #multimodality #speech
#NLP #multimodality #speech
Reposted by MilaNLP Lab
January 20, 2026 at 11:15 AM
Reposted by MilaNLP Lab
⏳ Deadline approaching! We’re hiring 2 fully funded postdocs in #NLP.
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
January 19, 2026 at 5:24 PM
⏳ Deadline approaching! We’re hiring 2 fully funded postdocs in #NLP.
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Reposted by MilaNLP Lab
Thank you @belindazli.bsky.social for the great talk "Solving the Specification Problem through Interaction” at our weekly seminar!
#NLProc
#NLProc
January 23, 2026 at 4:26 PM
Thank you @belindazli.bsky.social for the great talk "Solving the Specification Problem through Interaction” at our weekly seminar!
#NLProc
#NLProc
Reposted by MilaNLP Lab
This week at reading group 📚
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
January 23, 2026 at 1:35 PM
This week at reading group 📚
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
Thank you @belindazli.bsky.social for the great talk "Solving the Specification Problem through Interaction” at our weekly seminar!
#NLProc
#NLProc
January 23, 2026 at 4:26 PM
Thank you @belindazli.bsky.social for the great talk "Solving the Specification Problem through Interaction” at our weekly seminar!
#NLProc
#NLProc
This week at reading group 📚
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
January 23, 2026 at 1:35 PM
This week at reading group 📚
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
@pranav-nlp.bsky.social presented "You Cannot Sound Like GPT": Signs of language discrimination and resistance in computer science publishing.
Paper: arxiv.org/abs/2505.08127
#NLProc
January 20, 2026 at 11:15 AM
⏳ Deadline approaching! We’re hiring 2 fully funded postdocs in #NLP.
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
January 19, 2026 at 5:24 PM
⏳ Deadline approaching! We’re hiring 2 fully funded postdocs in #NLP.
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
Join the MilaNLP team and contribute to our upcoming research projects (SALMON & TOLD)
🔗 Details + how to apply: milanlproc.github.io/open_positio...
⏰ Deadline: Jan 31, 2026
We're also back with the lab's seminar! Today we had Eleonora Mancini presenting her doctoral research "Multimodal AI for Human Expression Understanding".
#NLP #multimodality #speech
#NLP #multimodality #speech
January 16, 2026 at 4:47 PM
We're also back with the lab's seminar! Today we had Eleonora Mancini presenting her doctoral research "Multimodal AI for Human Expression Understanding".
#NLP #multimodality #speech
#NLP #multimodality #speech
Holidays over, reading group resumes 📖
Today Henning Hoffmann presented the paper "Music for All: Representational Bias and Cross-Cultural Adaptability of Music Generation Models"
Paper: arxiv.org/pdf/2502.07328
#NLProc
Today Henning Hoffmann presented the paper "Music for All: Representational Bias and Cross-Cultural Adaptability of Music Generation Models"
Paper: arxiv.org/pdf/2502.07328
#NLProc
January 15, 2026 at 11:50 AM
Holidays over, reading group resumes 📖
Today Henning Hoffmann presented the paper "Music for All: Representational Bias and Cross-Cultural Adaptability of Music Generation Models"
Paper: arxiv.org/pdf/2502.07328
#NLProc
Today Henning Hoffmann presented the paper "Music for All: Representational Bias and Cross-Cultural Adaptability of Music Generation Models"
Paper: arxiv.org/pdf/2502.07328
#NLProc