Arianna Muti
@arimuti.bsky.social
Postdoc at Bocconi University in the group of MilaNLP. Researching online misogyny, hate speech, implicit bias and personalized NLP.
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
Paper: aclanthology.org/2025.finding...
#EMNLP2025
Proud to present our #EMNLP2025 papers!
Catch our team across Main, Findings, Workshops & Demos 👇
Catch our team across Main, Findings, Workshops & Demos 👇
November 5, 2025 at 5:24 PM
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
Paper: aclanthology.org/2025.finding...
#EMNLP2025
Reposted by Arianna Muti
📚 For today’s reading group @arimuti.bsky.social presented Emergent Misalignment: Narrow Finetuning Can Produce Broadly Misaligned LLMs (Betley et al., 2025).
🧩 arxiv.org/abs/2502.17424
#NLProc #AIAlignment #LLMs
🧩 arxiv.org/abs/2502.17424
#NLProc #AIAlignment #LLMs
October 9, 2025 at 12:07 PM
📚 For today’s reading group @arimuti.bsky.social presented Emergent Misalignment: Narrow Finetuning Can Produce Broadly Misaligned LLMs (Betley et al., 2025).
🧩 arxiv.org/abs/2502.17424
#NLProc #AIAlignment #LLMs
🧩 arxiv.org/abs/2502.17424
#NLProc #AIAlignment #LLMs
Reposted by Arianna Muti
📣 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?
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?
September 29, 2025 at 2:54 PM
📣 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?
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?
Reposted by Arianna Muti
📖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
"Culture is Everywhere: A Call for Intentionally Cultural Evaluation" by Oh et al.
Paper: arxiv.org/pdf/2509.01301
#NLProc
September 29, 2025 at 9:26 AM
📖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
"Culture is Everywhere: A Call for Intentionally Cultural Evaluation" by Oh et al.
Paper: arxiv.org/pdf/2509.01301
#NLProc
Reposted by Arianna Muti
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
Check the paper at clic2025.unica.it/wp-content/u...
#NLProc #clicit25
September 26, 2025 at 10:19 AM
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
Check the paper at clic2025.unica.it/wp-content/u...
#NLProc #clicit25
Reposted by Arianna Muti
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
At our last seminar’s lab, Jan Batzner presented The Brief History of LLM Sycophancy as We Know It.
#NLProc #sycophancy #LLM
September 25, 2025 at 11:08 AM
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
At our last seminar’s lab, Jan Batzner presented The Brief History of LLM Sycophancy as We Know It.
#NLProc #sycophancy #LLM
Reposted by Arianna Muti
Reposted by Arianna Muti
📖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
presented two papers on #sycophancy in #LLMs.
Papers: arxiv.org/pdf/2505.13995, arxiv.org/pdf/2310.13548
#NLProc
September 11, 2025 at 10:50 AM
📖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
presented two papers on #sycophancy in #LLMs.
Papers: arxiv.org/pdf/2505.13995, arxiv.org/pdf/2310.13548
#NLProc
Reposted by Arianna Muti
We're back with the reading group! 🚀
Today @paul-rottger.bsky.social presented The Levers of Political Persuasion with Conversational AI by Hackenburg et al. (2025)
Paper: arxiv.org/pdf/2507.13919
#NLProc
Today @paul-rottger.bsky.social presented The Levers of Political Persuasion with Conversational AI by Hackenburg et al. (2025)
Paper: arxiv.org/pdf/2507.13919
#NLProc
September 4, 2025 at 11:05 AM
We're back with the reading group! 🚀
Today @paul-rottger.bsky.social presented The Levers of Political Persuasion with Conversational AI by Hackenburg et al. (2025)
Paper: arxiv.org/pdf/2507.13919
#NLProc
Today @paul-rottger.bsky.social presented The Levers of Political Persuasion with Conversational AI by Hackenburg et al. (2025)
Paper: arxiv.org/pdf/2507.13919
#NLProc
Reposted by Arianna Muti
"The "r" in "woman" stands for rights. Auditing LLMs in Uncovering Social Dynamics in Implicit Misogyny" (lead by @arimuti.bsky.social and w. @deboranozza.bsky.social Alberto Barron-Cedeño and Chris Emmery)
August 22, 2025 at 8:32 AM
"The "r" in "woman" stands for rights. Auditing LLMs in Uncovering Social Dynamics in Implicit Misogyny" (lead by @arimuti.bsky.social and w. @deboranozza.bsky.social Alberto Barron-Cedeño and Chris Emmery)
Reposted by Arianna Muti
July 30, 2025 at 8:45 AM
Reposted by Arianna Muti
🎉 The @milanlp.bsky.social lab is excited to present 15 papers and 1 tutorial at #ACL2025 & workshops! Grateful to all our amazing collaborators, see everyone in Vienna! 🚀
July 16, 2025 at 12:11 PM
🎉 The @milanlp.bsky.social lab is excited to present 15 papers and 1 tutorial at #ACL2025 & workshops! Grateful to all our amazing collaborators, see everyone in Vienna! 🚀
Reposted by Arianna Muti
🎓For today's lab seminar, we had the pleasure to host
@miriamschirmer.bsky.social with her presentation on Measuring and Reducing the Psychological Impact of Online Harm and Tiago Pimentel with How Much Does Tokenisation Impact Language Models?
#NLProc #onlineharms #tokenisation
@miriamschirmer.bsky.social with her presentation on Measuring and Reducing the Psychological Impact of Online Harm and Tiago Pimentel with How Much Does Tokenisation Impact Language Models?
#NLProc #onlineharms #tokenisation
July 11, 2025 at 3:10 PM
🎓For today's lab seminar, we had the pleasure to host
@miriamschirmer.bsky.social with her presentation on Measuring and Reducing the Psychological Impact of Online Harm and Tiago Pimentel with How Much Does Tokenisation Impact Language Models?
#NLProc #onlineharms #tokenisation
@miriamschirmer.bsky.social with her presentation on Measuring and Reducing the Psychological Impact of Online Harm and Tiago Pimentel with How Much Does Tokenisation Impact Language Models?
#NLProc #onlineharms #tokenisation
Reposted by Arianna Muti
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
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
July 7, 2025 at 2:08 PM
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
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
A new paper with Alice Iannantuoni and Matteo Casiraghi
analyzing Italian tweets on feminism, tracking toxicity and shifting connotations over time is being presented today at the LLM for political science workshop at @europsa.bsky.social 📉📈
analyzing Italian tweets on feminism, tracking toxicity and shifting connotations over time is being presented today at the LLM for political science workshop at @europsa.bsky.social 📉📈
June 25, 2025 at 8:23 AM
A new paper with Alice Iannantuoni and Matteo Casiraghi
analyzing Italian tweets on feminism, tracking toxicity and shifting connotations over time is being presented today at the LLM for political science workshop at @europsa.bsky.social 📉📈
analyzing Italian tweets on feminism, tracking toxicity and shifting connotations over time is being presented today at the LLM for political science workshop at @europsa.bsky.social 📉📈
Reposted by Arianna Muti
🎓For today's lab seminar it was a pleasure to have Federico Ruggeri presenting "From data annotation to modelling,
the importance of integrating and extracting textual knowledge in hate speech"
#NLProc
the importance of integrating and extracting textual knowledge in hate speech"
#NLProc
June 20, 2025 at 3:57 PM
🎓For today's lab seminar it was a pleasure to have Federico Ruggeri presenting "From data annotation to modelling,
the importance of integrating and extracting textual knowledge in hate speech"
#NLProc
the importance of integrating and extracting textual knowledge in hate speech"
#NLProc
Reposted by Arianna Muti
📖For this Thursday's Reading Group @taniseceron.bsky.social presented "Generative Echo Chamber? Effects of LLM-Powered Search Systems on Diverse Information Seeking " by Nikhil Sharma, Q. Vera Liao and Ziang Xiao.
Paper: dl.acm.org/doi/pdf/10.1...
#NLProc
Paper: dl.acm.org/doi/pdf/10.1...
#NLProc
June 19, 2025 at 11:25 AM
📖For this Thursday's Reading Group @taniseceron.bsky.social presented "Generative Echo Chamber? Effects of LLM-Powered Search Systems on Diverse Information Seeking " by Nikhil Sharma, Q. Vera Liao and Ziang Xiao.
Paper: dl.acm.org/doi/pdf/10.1...
#NLProc
Paper: dl.acm.org/doi/pdf/10.1...
#NLProc
Reposted by Arianna Muti
🎓For last Friday's lab seminar, it was a pleasure to have
@shehzaadzd.bsky.social presenting "Learning to Adapt and Align with Feedback"
#NLProc
@shehzaadzd.bsky.social presenting "Learning to Adapt and Align with Feedback"
#NLProc
June 16, 2025 at 1:29 PM
🎓For last Friday's lab seminar, it was a pleasure to have
@shehzaadzd.bsky.social presenting "Learning to Adapt and Align with Feedback"
#NLProc
@shehzaadzd.bsky.social presenting "Learning to Adapt and Align with Feedback"
#NLProc
Reposted by Arianna Muti
📖For this Thursday's Reading Group @a-lauscher.bsky.social presented "Aligned but Blind: Alignment Increases Implicit Bias by Reducing Awareness of Race " by Lihao Sun, Chengzhi Mao, Valentin Hofmann and Xuechunzi Bai.
Paper: arxiv.org/abs/2506.002....
#NLProc
Paper: arxiv.org/abs/2506.002....
#NLProc
June 12, 2025 at 12:23 PM
📖For this Thursday's Reading Group @a-lauscher.bsky.social presented "Aligned but Blind: Alignment Increases Implicit Bias by Reducing Awareness of Race " by Lihao Sun, Chengzhi Mao, Valentin Hofmann and Xuechunzi Bai.
Paper: arxiv.org/abs/2506.002....
#NLProc
Paper: arxiv.org/abs/2506.002....
#NLProc
🤖 Insieme ad alcune ricercatrici e ricercatori, abbiamo creato un sondaggio per capire come viene usata l’IA per la lingua italiana da persone non esperte del settore
📊 Dura circa 10 minuti e può partecipare chiunque parli italiano e abbia più di 18 anni
🔗 bit.ly/sondaggio_ai...
📊 Dura circa 10 minuti e può partecipare chiunque parli italiano e abbia più di 18 anni
🔗 bit.ly/sondaggio_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
June 12, 2025 at 11:15 AM
Reposted by Arianna Muti
📖For today's Reading Group @dirkhovy.bsky.social presented "Don’t lie to your friends: Learning what you know from collaborative self-play" by Jacob Eisenstein et al.
Paper: arxiv.org/pdf/2503.14481
#NLProc #AiAgents
Paper: arxiv.org/pdf/2503.14481
#NLProc #AiAgents
May 15, 2025 at 12:55 PM
📖For today's Reading Group @dirkhovy.bsky.social presented "Don’t lie to your friends: Learning what you know from collaborative self-play" by Jacob Eisenstein et al.
Paper: arxiv.org/pdf/2503.14481
#NLProc #AiAgents
Paper: arxiv.org/pdf/2503.14481
#NLProc #AiAgents
Reposted by Arianna Muti
🎓For today's lab seminar, it was a pleasure to have
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
May 9, 2025 at 4:10 PM
🎓For today's lab seminar, it was a pleasure to have
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
@manueltonneau.bsky.social presenting "HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter" ✨
#NLProc #hatespeech
Reposted by Arianna Muti
📖For today's Reading Group Joe Baumann presented "The Leaderboard Illusion" by Shivalika Singh et al.
Paper: arxiv.org/pdf/2504.20879
#NLProc
Paper: arxiv.org/pdf/2504.20879
#NLProc
May 8, 2025 at 1:47 PM
📖For today's Reading Group Joe Baumann presented "The Leaderboard Illusion" by Shivalika Singh et al.
Paper: arxiv.org/pdf/2504.20879
#NLProc
Paper: arxiv.org/pdf/2504.20879
#NLProc
May 2, 2025 at 2:48 PM
Reposted by Arianna Muti
📖For today's Reading Group @arimuti.bsky.social presented "Do LLMs Understand Social Knowledge? Evaluating the Sociability of Large Language Models with the SocKET Benchmark" by Minje Choi et al.
Paper: aclanthology.org/2023.emnlp-m...
#NLProc
Paper: aclanthology.org/2023.emnlp-m...
#NLProc
April 24, 2025 at 11:50 AM
📖For today's Reading Group @arimuti.bsky.social presented "Do LLMs Understand Social Knowledge? Evaluating the Sociability of Large Language Models with the SocKET Benchmark" by Minje Choi et al.
Paper: aclanthology.org/2023.emnlp-m...
#NLProc
Paper: aclanthology.org/2023.emnlp-m...
#NLProc