I will present two posters in the following sessions. Welcome to have a chat. #EMNLP2025
Latest posts 5
👏I'm excited to share that 3 of our papers have been accepted by #EMNLP2025!
Thank you to all the co-authors🙌 It's impossible to achieve without your help!
See you in Suzhou 😁
Thank you to all the co-authors🙌 It's impossible to achieve without your help!
See you in Suzhou 😁
Wonder how humans and LLMs perceive interestingness in English learning conversations?
Check out our paper at CogSci 25: osf.io/preprints/osf/…
Fri 1st August at 12:00 US/Pacificc time (21:00 CET). Link: teams.microsoft.com/l/meetup-join/…
@pergolagb
Check out our paper at CogSci 25: osf.io/preprints/osf/…
Fri 1st August at 12:00 US/Pacificc time (21:00 CET). Link: teams.microsoft.com/l/meetup-join/…
@pergolagb
Multi-step reasoning #LLMs are often fine-tuned on trajectories that are filtered with process reward models (PRMs) automatically trained on Monte Carlo estimation-generated pseudo labels.
If you are attending #NAACL2025 in person, you are welcome to check out our Safespeech DEMO in Hall 3 (9:00-10:30).
arxiv.org/abs/2503.06534
arxiv.org/abs/2503.06534
arxiv.org SafeSpeech: A Comprehensive and Interactive Tool for Analysing... Detecting toxic language including sexism, harassment and abusive behaviour, remains a critical challenge, particularly in its subtle and context-dependent forms. Existing approaches largely focus...