Ana Ezquerro

Países bombardeados por Estados Unidos en los últimos dos años: 8 Países bombardeados por Israel: 7 Países bombardeados por Irán: 1 (el propio Israel como autodefensa) Pues de alguna manera quien amenaza la seguridad mundial es el último. Absoluta vergüenza. No somos tontos.

No one is illegal on stolen data

.@anaezquerro_ presenting our work on graph parsing as tagging at #EMNLP2025 !!!

Excited to announce that my Conversational SEO paper got accepted at #neurips25 D&B!! 🎉🥳 Do you want to know how do to SEO on conversational search engines such as Perplexity? Take a look to our paper and thread 👇 or come to our poster in 🇺🇸!

We introduce PiCSAR (Probabilistic Confidence Selection And Ranking)💡: A simple training-free method for scoring samples based on probabilistic confidence, selecting a reasoning chain with the highest confidence from multiple sampled responses. ✏️PiCSAR is generalisable across LRMS and LLMs of different model families and scales. 📈PiCSAR achieves huge and consistent performance boost across models and datasets (eg. +10.18% on MATH500, +11.7% on AIME2024, +9.81% on AIME2025, +7.67% on GPQA). 💡By using PiCSAR on only 6 samples, we achieve higher accuracy than self-consistency on 16/32 samples. ⚡️We also observe that when using PiCSAR, confidence can be estimated using a much smaller model (eg. reducing 72b to 8b model) to improve computational efficiency. A huge thank you to my amazing collaborators: @zhengzhao97 @aryopg @soheeyang_ @cyruskwan1997 @zodiacJRH @WendaLi8 @PMinervini @e_giunchiglia Shay Cohen

Today, Adrián Gude, Roi Santos and @anaezquerro_ have presented "LyS at SemEval 2025 Task 8: Zero-Shot Code Generation for Tabular QA" at @semeval_tableQA @SemEvalWorkshop at #ACL2025NLP @lysresearch @citicresearch @UDC_gal

Huge congrats to the NSA team for their ACL2025 Best Paper win! 🏆🏆🏆 We've open-sourced a 3rdparty impl to help you integrate this breakthrough. Play with it now! Get the code: 🔗 NSA: github.com/fla-org/native… 🔗 Also in FLA: github.com/fla-org/flash-… @Xinyu2ML @SonglinYang4

Tomorrow (Tuesday) at 14:00 we present "Hierarchical Bracketing Encodings for Dependency Parsing as Tagging" at #ACL2025NLP. Meet @anaezquerro_ @eusondavid @carlosgr_nlp at Room 1.86 @lysresearch @citicresearch @UDC_gal 🔗Paper aclanthology.org/2025.acl-long.…

Si Tranvías siguiera operando en A Coruña en el año 2300 con las carracas del 2000🤣🤣

@rociooo_sr no se puede 🫠

Dios la puta IA hasta en la sopa como quito el maldito simbolito de was??? Qué no necesito ayuda para hablar con mis amigos

🚨IT'S HAPPENING🚨 a woman who cannot speak now speaks through her brain, in real time, with her own voice. no typing, delay, or sounds made. just neural intent to streaming speech this isn’t prediction. it’s embodiment 1/

Se nos acabó el celo 🤣🤣🤣

Currently at the Data-oriented Area Studies Unit End-Year Workshop 2025 on "Data-Oriented Approaches to the Social Sciences and Humanities", at @kucseas (Kyoto University), honored to give a talk at 16:40 on "Contrasting linguistic patterns in human and LLM-generated news text".

Twitter used to be my favorite place on the Internet. I've derived enormous value from it in the past 16 years. Not true anymore. Most of the people I enjoyed reading have left. My feed, which used to feature art and science and technology and humor, has become constant political propaganda -- on the opposite side of the Enlightenment values I believe in. (I like the rule of law, free speech, free markets, and democracy. I like science and reject obscurantism. I can see that Putin is a dictator, not a genius role model, and that it is Russia that is invading Ukraine, not the other way around.) I have been coming here less and less as a result -- it only brings me negative emotions. I never thought that would be possible, but I might eventually stop coming altogether.

Python library for creating math animations

Merry Christmas to everyone but especially to the people doing open-source

Today at 14:00, we'll be presenting "Dependency Graph Parsing as Sequence Labeling," by @anaezquerro_, @eusondavid, and @carlosgr_nlp at #EMNLP2024! Join us at the poster session in Riverfront Hall. 🔗 Paper: aclanthology.org/2024.emnlp-mai…

🌴 On our way to Miami for #EMNLP2024 to present two papers! 📄 - Revisiting Supertagging for Faster HPSG Parsing (@OlgaZamaraeva, @carlosgr_nlp) - Dependency Graph Parsing as Sequence Labeling (w/ @anaezquerro_, @eusondavid, @carlosgr_nlp)

I'll present our work at #EMNLP2024 in the in-person poster session B (Riverfront Hall) on November 12, 14:00 - 15:30. Come, let’s discuss how to elicit syntactic trees from RoBERTa vectors via structured contrastive hashing! aclanthology.org/2024.emnlp-mai…