You might like
How nice 🥳🎉 - we got an outstanding paper at #naacl2025 with our paper on literary translation (evaluation) with LLMs aclanthology.org/2025.naacl-lon…. Congrats to @yyrrss2021 @andyweizhao @egere14
1/3 📢📢We're very thrilled to announce #DeTikZify, a vision language model trained on #TikZ code to automatically convert sketches and existing scientific figures into editable, semantics-preserving TikZ programs. A video is worth a 1000 words: youtube.com/watch?v=hDbTSi…
youtube.com
YouTube
DeTikZify: Synthesizing Graphics Programs for Scientific Figures and...
We're happy to announce that our paper "Cross-lingual Cross-temporal Summarization (CLCTS)" by @yyrrss2021, Jihed Ouni and @egere14 is now published at the Computational Linguistics journal direct.mit.edu/coli/article/d….
🌟 Exciting times ahead! The NLLG group is now part of #UniMannheim, working with miraculous Simone Ponzetto and the rest of the wonderful @dwsunima group. Can't wait to see what we cook up together — it's definitely more than just coffee! ☕😉🚀
📢📢 We did a little analysis on the most "influential" (= most cited) papers from the first half of 2023 (only NLP and ML, all from arXiv). LLMs clearly dominate. LLaMA > GPT-4 > PaLM 2 > ... Arxiv link: arxiv.org/pdf/2308.04889… Medium link: medium.com/p/affb3cdc5c6a
📢📢📢 The Eval4NLP workshop will take place this year at AACL 2023. Special focus: Evaluation of/with LLMs. Including a shared task on Prompting LLMs as Explainable Metrics. 📢📢📢 Direct submission deadline: 25.08. Webpage: eval4nlp.github.io/2023/index.html CFP: eval4nlp.github.io/2023/cfp.html
Highlights on Day 3 @ #ACL2023. - 📢📢📢 @ccccahhh presents MENLI accepted by TACL at the virtual poster session on 12 July (17:00 PM - 18:30 PM CEST). -🎉🎉🎉 @jbelouadi @egere14 & ByGPT5 received an honorable mention. Check for more details: 2023.aclweb.org/program/best_p…
Happy to announce our new paper MENLI accepted by TACL by @ccccahhh & @egere14！ We challenge NLG evaluation metrics in adversarial conditions with correctness/adequacy errors. Our NLI-based metrics are much more robust to adversarial attacks！ 👉Preprint: arxiv.org/abs/2208.07316
Happy to announce our new paper UScore at #eacl2023 by @Jonas62744761 & @egere14 the first fully unsupervised evaluation metric for MT, expanding the applicability of MT evaluation by addressing more languages with promising results! arxiv.org/abs/2202.10062
👉 #hiring: New #PhDposition in our great team @NL2GBielefeld with @egere14. Very cool interdisciplinary research project, #NLProc + #socialsciences. Application deadline 23.02.2023. More about the opening: nl2g.github.io/positions @unibielefeld @CITEC_Bielefeld
Here we are! Welcome to NLLG group @unibielefeld. We do #NLProc, text generation + evaluation, digital humanities, #socialsciences. @egere14 @Jonas62744761 @ChrLeiter @yyrrss2021 @ Daniil
United States Trends
- 1. #BUNCHITA 1,307 posts
- 2. #SmackDown 42.7K posts
- 3. Tulane 3,649 posts
- 4. Aaron Gordon 2,371 posts
- 5. Giulia 14K posts
- 6. Supreme Court 181K posts
- 7. Russ 12.9K posts
- 8. #OPLive 2,218 posts
- 9. Connor Bedard 2,216 posts
- 10. #TheLastDriveIn 3,181 posts
- 11. #BostonBlue 4,184 posts
- 12. Podz 2,568 posts
- 13. Caleb Wilson 5,408 posts
- 14. Northwestern 4,834 posts
- 15. Scott Frost N/A
- 16. Rockets 20.1K posts
- 17. Memphis 15.7K posts
- 18. Frankenstein 72.6K posts
- 19. Zach Lavine N/A
- 20. Isaiah Hartenstein N/A
Something went wrong.
Something went wrong.