Volunteering at NAACL LatinX in AI (LXAI) #NAACL2022 with the awesome Nikita Soni Janet Liu ⛷️ ⭐️💃🏽🎉🖥👾🤖💥 Come meet us at room 507!

Pickup your #LatinXinAI swag at #Naacl2022 in room 507 Sauk today! lxnlp NAACL #LatinXinNLP LatinX in AI (LXAI)

Curso exprés de literatura...

"How did you end up working in Accademia?"

Meanwhile, inside the box, Schrödinger's cat plans its revenge.

In the resort town of Les 2 Alpes, in the northern French Alps there is a bike race which starts at a 3,400m-high glacier. 1/2

Nobody: Mexican taqueros:

Kostiantyn K. presents a position paper on how an interdisciplinary perspective (computational, NLP, vis) is necessary for visual text analytics Nicole Sultanum José Angel Daza Vasiliki Simaki M. Skeppstedt Barbara Plank @jdfaviz @nargesmahyar arxiv.org/abs/2209.11534

El jueves 10/11 participo en una mesa para platicar sobre el PLN y su impacto en el desarrollo científico. Abordaré la alianza de las humanidades con el PLN para explorar nuevas rutas del conocimiento hacia las grandes preguntas alrededor de nuestra sociedades 🫂

Finally out! Our EMNLP 2022 paper explores techniques for dealing with abbreviations on domain-specific data for Slovenian, comparing different baselines, and using a language model for abbreviation identification and expansion: arxiv.org/pdf/2211.02429… 🙂

Sentence embeddings (e.g., SBERT) are powerful -- but we just don't know what is crammed into a %&!$# vector 😵‍💫. 💥So in our new paper, we use Abstract Meaning Representation (AMR) to make sentence embeddings more explainable! #AACL2022 #nlproc #MachineLearning (1/3)

Presenting right now our INTAVIA work on abbreviation identification and expansion for Slovenian @ #emnlp2022 feel free to come by and ask questions 🙂

Interesting new technologies to follow in 2023 technologyreview.com/2023/01/09/106…

🚨Excited to share our new paper, “REFINER: Reasoning Feedback on Intermediate Representations” 🔥 Can LLMs enhance their reasoning capabilities by utilizing feedback on their intermediate reasoning steps? Yes, in some cases! #NLProc 🔗 debjitpaul.github.io/refiner/ 🧵1/n

As #InTaVia prepares for 𝑎𝑛𝑜𝑡ℎ𝑒𝑟 𝑐𝑜𝑛𝑓𝑒𝑟𝑒𝑛𝑐𝑒 (𝐋𝐣𝐮𝐛𝐥𝐣𝐚𝐧𝐚, 𝟐𝟓-𝟐𝟔 𝐒𝐞𝐩𝐭𝐞𝐦𝐛𝐞𝐫, hosted by the Institute of Cultural History ZRC SAZU: 𝐦𝐨𝐫𝐞 𝐭𝐨 𝐜𝐨𝐦𝐞), it is time to look back and take stock of the past few months.

José Angel Daza shows recent work on how to make Wikipedia a data source for INTAVIA users by processing WP articles and importing data for any existing person or group, including those who commonly escape the European / Eurocentric view of national biography projects.

Aquí pueden encontrar una lista de instituciones públicas 🏫 para estudiar IA 🤖 en una maestría; 🔴🔴 la opción en la UNAM se vence pronto: 12 de febrero; la iré actualizando periodicamente turing.iimas.unam.mx/~ivanvladimir/…

Había estado muy confundido 😵‍💫 con el avance regulatorio de IA en México; así que me puse a leer las iniciativas de ley e hice un post al respecto 🤓 (spoiler 🚨) en resumen no hay regulación de la IA en México... todavía -- pero descubrí varias cosas: turing.iimas.unam.mx/~ivanvladimir/…