Politehnica Bucharest NLP Group
@upb-nlp.bsky.social
The Natural Language Processing Group at the National University of Science and Technology Politehnica Bucharest
nlp.upb.ro
#NLProc
nlp.upb.ro
#NLProc
We also presented our work on multiple-choice question generation at AIED 2025: YMCQ: Reasoning-Enhanced MCQ Generation, Andreea Dutulescu, Ruseti Stefan, Denis Iorga, Mihai Dascalu, and Danielle S. McNamara.
July 27, 2025 at 9:35 AM
We also presented our work on multiple-choice question generation at AIED 2025: YMCQ: Reasoning-Enhanced MCQ Generation, Andreea Dutulescu, Ruseti Stefan, Denis Iorga, Mihai Dascalu, and Danielle S. McNamara.
Presented our work on automatic scoring of students responses at EDM 2025: One Model to Score Them All: Unified Scoring of Learning
Strategies with LLMs - Andreea Dutulescu, Stefan Ruseti, Mihai Dascalu and Danielle McNamara.
Strategies with LLMs - Andreea Dutulescu, Stefan Ruseti, Mihai Dascalu and Danielle McNamara.
July 24, 2025 at 5:54 AM
Presented our work on automatic scoring of students responses at EDM 2025: One Model to Score Them All: Unified Scoring of Learning
Strategies with LLMs - Andreea Dutulescu, Stefan Ruseti, Mihai Dascalu and Danielle McNamara.
Strategies with LLMs - Andreea Dutulescu, Stefan Ruseti, Mihai Dascalu and Danielle McNamara.
We went on the waitinglist as soon as we saw the first announcement 🤩 looking forward to this
Last week, we launched a waitlist to move builders on @hf.co from LFS to Xet. This was made possible through months of hard work and staged migrations to test our infrastructure in real-time.
This post provides an inside look into the day of our first migrations and the weeks after.
This post provides an inside look into the day of our first migrations and the weeks after.
Xet is on the Hub
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
March 22, 2025 at 10:26 AM
We went on the waitinglist as soon as we saw the first announcement 🤩 looking forward to this
Really appreciate studies that tame the unnecessary hype
Lets replace search with "AI" then! Totally logical if you ask me. Even more worth it when you know they're exponentially overtaking the airline industry in their carbon footprint.
Study: www.cjr.org/tow_center/w...
Study: www.cjr.org/tow_center/w...
March 15, 2025 at 7:24 PM
Really appreciate studies that tame the unnecessary hype
Really nice angle - to use AI tu support people that are not able to be professionally represented in court 👏
Courthouse AI is "one of the most compelling but also underdeveloped areas of public sector AI," said Stanford Law professor David Engstrom at a recent Stanford HAI seminar. Here, he describes how AI can support self-represented litigants in civil cases: hai.stanford.edu/news/harness...
Harnessing AI to Improve Access to Justice in Civil Courts | Stanford HAI
Stanford law professor advocates for public-sector AI tools to support self-represented litigants in civil cases.
hai.stanford.edu
March 15, 2025 at 7:18 PM
Really nice angle - to use AI tu support people that are not able to be professionally represented in court 👏
This is a really important task, especially nowadays. Not only for research, but also for general use
Check our preprint and code:
📄Paper: arxiv.org/abs/2502.20503
💻Code: github.com/UKPLab/arxiv2025-misleading-visualizations
Joint work of @tongletj.bsky.social, Marie-Francine Moens, Tinne Tuytelaars, and @igurevych.bsky.social (5/5)
📄Paper: arxiv.org/abs/2502.20503
💻Code: github.com/UKPLab/arxiv2025-misleading-visualizations
Joint work of @tongletj.bsky.social, Marie-Francine Moens, Tinne Tuytelaars, and @igurevych.bsky.social (5/5)
Protecting multimodal large language models against misleading visualizations
We assess the vulnerability of multimodal large language models to misleading visualizations - charts that distort the underlying data using techniques such as truncated or inverted axes, leading read...
arxiv.org
March 10, 2025 at 3:30 PM
This is a really important task, especially nowadays. Not only for research, but also for general use
Bit of a tight deadline but good to know! NeurIPS abstracts are due on May 11th.
The NeurIPS Call for Papers is now live. Abstracts are due May 11th AoE, with full papers due May 15th AoE. neurips.cc/Conferences/...
Please read about key changes to Dataset and Benchmarks submissions this year in our blog post: blog.neurips.cc/2025/03/10/n...
Please read about key changes to Dataset and Benchmarks submissions this year in our blog post: blog.neurips.cc/2025/03/10/n...
NeurIPS 2025 Call for Papers
Submit at: https://openreview.net/group?id=NeurIPS.cc/2025/Conference
neurips.cc
March 10, 2025 at 3:29 PM
Bit of a tight deadline but good to know! NeurIPS abstracts are due on May 11th.
Awesome stuff 👏🤩
In 2013, at AKBC 2013 and other workshops, I gave a talk titled “Texts are Knowledge”. This was well before there were any transformer LLMs—indeed before the invention of attention—and my early neural NLP ideas were rudimentary.
🔮 Nevertheless, the talk was quite prophetic!
🔮 Nevertheless, the talk was quite prophetic!
March 4, 2025 at 8:56 AM
Awesome stuff 👏🤩
Looking forward to try it out 👀
Following Skyreel from SkyworkAI, Alibaba now released its open source GenAI model Wan 2.1.
github.com/SkyworkAI/Sk...
github.com/Wan-Video/Wa...
github.com/SkyworkAI/Sk...
github.com/Wan-Video/Wa...
GitHub - SkyworkAI/SkyReels-V1: SkyReels V1: The first and most advanced open-source human-centric video foundation model
SkyReels V1: The first and most advanced open-source human-centric video foundation model - SkyworkAI/SkyReels-V1
github.com
March 3, 2025 at 3:39 PM
Looking forward to try it out 👀
#NLProc #HelloWorld #research
February 27, 2025 at 9:51 AM
#NLProc #HelloWorld #research