Michele Papucci

Last week I presented my latest work at the Findings Poster Session of ACL 2025 in Vienna! If you missed it, check it out 💥 michelepapucci.github.io/blog/findings-…

@ItaliaNLP_Lab @CLiC_it_conf @AILC_NLP 🔥🔥

🎉 Great news! We got 9 papers accepted at CLiC-it 2025! Looking forward to presenting them this year in Cagliari! 🇮🇹 #CLiCit2025 #NLProc @CLiC_it_conf @AILC_NLP

We are at the Lectures on Computational Linguistics in Milan organized by @AILC_NLP ! 🔥 @mpapucci_ and @workerplacemint are presenting their PhD work during the poster session! 🥳 #NLProc

@koksing456 @pdrndr @AlessioMiaschi @gpuccetti92 Gemma is going to be uploaded here: huggingface.co/andreapdr/LID-…

@koksing456 @pdrndr @AlessioMiaschi @gpuccetti92 We've yet to upload the models, but the Llama one will be uploaded on HuggingFace in the next days, here: huggingface.co/andreapdr/LID-…

🧵1/ Machine-Generated Text (MGT) detection is failing. Our paper, accepted at Findings of ACL 2025, shows that LLMs can easily fool generated-text detectors. arxiv.org/abs/2505.24523 @pdrndr, Cristiano Ciaccio, @AlessioMiaschi, @gpuccetti92, Felice dell'Orletta, Andrea Esuli

8/ TL:DR; 🚨 State-of-the-art Detectors today are too shallow 📉 A bit of style alignment makes them crumble 🧠 We need stronger benchmarks 🛠 We develop a way to create hard, in-domain texts for making and evaluating the next generation of more robust and reliable MGT Detectors

7/ What about Humans? Human performance was unaffected: they performed poorly in detecting machine-generated text (around 50% accuracy in a binary task) both before and after our alignment.

@mntssys and I are excited to announce circuit-tracer, a library that makes circuit-finding simple! Just type in a sentence, and get out a circuit showing (some of) the features your model uses to predict the next token. Try it on @neuronpedia: shorturl.at/SUX2A

4) Findings: Stress-testing Machine Generated Text Detection: Shifting Language Models Writing Style to Fool Detectors (@pdrndr, @mpapucci_, Ciaccio C., @AlessioMiaschi, @gpuccetti92, Dell'Orletta F. and Esuli A.)

Exciting news!🔥 We got 4 papers accepted at #ACL2025NLP: 2 at the main conference and 2 in the Findings! See you in Vienna! 🎉 @aclmeeting #NLPRoc Details in the thread 👇

And now is the turn of Chiara Fazzone, presenting “SimilEx: the First Italian Dataset for Sentence Similarity with Natural Language Explanations” (with @AlzettaChiara, Dell’Orletta F. and Venturi G.)! 🔥 @CLiC_it_conf @AILC_NLP #NLProc

[1/4] 🎉 Excited to announce that our paper "🥞TEXT-CAKE: Challenging Language Models on Local Text Coherence" (@lucadini_, F. Dell'Orletta, D. Brunato, @tommaso_caselli) has been accepted at COLING 2025 (@coling2025)! 📝✨ #NLProc #COLING2025