Fernando Adauto retweeted
🌍 How do #LLMs handle trolley problems across cultures? We test them with 98K dilemmas in 107 languages, grounded in 40M+ human moral judgments.
💡 Spotlight @ICLR2025 in Singapore✈️| Best Paper @pluralistic_ai Workshop #NeurIPS2024
📄 Paper: arxiv.org/abs/2407.02273 🧵👇
Excited to be at #OpenAIDevDay, looking for AI advancements to help innovate and transform Kavak @KavakMexico
1
247
Fernando Adauto retweeted
Excited to share our 5 papers just accepted at #EMNLP2024 and #NeurIPS2024! Thanks to all collaborators, my dear mentors @radamihalcea @mrinmayasachan @bschoelkopf, and talented mentees! [Talent Alert🧑🎓] MSc mentees now open for PhD application: @Jiarui_Liu_ @fer_adauto @giorgiopiatti @YahangQi I can say nothing but very high recommendation of how great they are as rising researchers!🎉
Fernando Adauto retweeted
Para contextualizar el reciente #linchamiento en #Taxco, quizás este artículo sobre linchamientos en América Latina sea de interés. journals.sagepub.com/doi/10.…
Sigue hilo en español abajo:
El linchamiento es una forma de violencia común en América Latina. ¿Qué tan común?
Hoy publicamos la primera base de datos sobre linchamientos cubriendo TODA América Latina.
Encontramos evidencia de 2818 casos de 2010-2019.
HILO 👇
App aquí: css-ethz.github.io/lyla/
Check out our poster today at 10:45 AM (Poster #413 in the big hall) presenting our paper "CLadder: A Benchmark to Assess Causal Reasoning Capabilities of LMs" #NeurIPS
Can LLMs pass 𝐂𝐚𝐮𝐬𝐚𝐥 𝐓𝐮𝐫𝐢𝐧𝐠 𝐓𝐞𝐬𝐭? In our #NeurIPS2023 paper “CLadder: A Benchmark to Assess Causal Reasoning Capabilities of LMs”, we test LLMs on the Pearl's 3 Rungs of Causality, and found that LLMs are *far* from solving them.➡️arxiv.org/abs/2312.04350
2
190
Fernando Adauto retweeted
Dec 10th, 9:00-9:15 (East Foyer), Findings 7 (Posters)
Beyond Good Intentions: Reporting the Research Landscape of NLP for Social Good presented by @fer_adauto and @ZhijingJin
Fernando Adauto retweeted
I and many of our group members are at #EMNLP2023 this week. We have a number of contributions to the conference spanning various topics in NLP, Reasoning and AI in Education. Please drop by if you are interested in any of the following:
Fernando Adauto retweeted
We're rolling out new features and improvements that developers have been asking for:
1. Our new model GPT-4 Turbo supports 128K context and has fresher knowledge than GPT-4. Its input and output tokens are respectively 3× and 2× less expensive than GPT-4. It’s available now to all developers in preview.
2. Assistants API and new tools (Retrieval, Code Interpreter) will help developers build world-class AI assistants within their own apps.
3. The platform is becoming multimodal. GPT-4 Turbo with Vision, DALL·E 3, and text-to-speech are all now available to developers.
Oh… and we’re doubling GPT-4 rate limits. openai.com/blog/new-models-a…
Happy to announce that our paper "Beyond Good Intentions: Reporting the Research Landscape for NLP for Social Good" has been accepted to EMNLP 2023 Findings! #EMNLP2023
Excited for our paper "Beyond Good Intentions: Reporting the Research Landscape of NLP for Social Good (NLP4SG)" at #EMNLP2023 Findings. We propose 3 NLP tasks: (1) detect if a paper relates to NLP4SG , (2) check which UN SDG it addresses, and (3) extract task & methods proposed.
Fernando Adauto retweeted
Conoce a nuestro segundo speaker: @fer_adauto 🥳
Fer hablará de Retrieval-Augmented Generation (RAG) para grandes modelos de lenguaje (LLMs), herramientas y casos de uso.
Últimos lugares en meetup.com/ai-learners/event… 👋
Muy emocionado por presentar mi charla en el próximo meetup de @AI_Learners sobre como potenciar los modelos de lenguaje. "Retrieval-Augmented Generation for LLMs". Registrate en meetup.com/ai-learners/event…
AI Learners está de vuelta🥳
Nuevo Meetup con @AndreMonserratt, @fer_adauto,
@MichDiaz_ y @osanseviero. ¡Nos vemos el 31 de Octubre en @platzi México!
Regístrense 👉 meetup.com/ai-learners/event…
What impact does NLP have on Social Good (NLP4SG)? In our new paper, we introduce a new scientific NLP task and a data visualization that shows an overview of the NLP4SG research landscape. arxiv.org/abs/2305.05471 (See our live demo: nlp4sg.vercel.app💙)🧵#NLProc
We also release our code and data:
📜Paper: arxiv.org/abs/2305.05471
💻Code: github.com/feradauto/nlp4sg
📁Data: Check out our "NLP4SGPapers set" in HuggingFace: huggingface.co/datasets/fera…
Credits to our wonderful team of co-authors: @ZhijingJin @bschoelkopf @Hoper_Tom @mrinmayasachan @radamihalcea from @ETH_en @ETH_AI_Center @MPI_IS @HebrewU @allen_ai @SemanticScholar @ai2_s2research @UMich @michigan_AI
Together with @cristiguzman we developed this app that shows the «Lynching in Latin America» (LYLA) dataset created by @EnzoNussio, @govclayton, and @CSS_ETHZurich. Check: css-ethz.github.io/lyla/ & the thread.
Lynching is a common form of violence in Latin America. How common?
Today, we release the first cross-national dataset on reported «Lynching in Latin America» (LYLA) from 2010-2019:
We found evidence of 2818 cases.
THREAD👇
Check out our app: css-ethz.github.io/lyla/
Fernando Adauto retweeted
Thanks to people in Latin America for their support! @mexevalua @CIDE_MX @CesigItam @causaencomun @CNDH @TheWJP_mx @ideaspaz @LAPOP_Lab @INEGI_INFORMA
Thanks @snsf_ch for funding and @CSS_ETHZurich @ETH_en for hosting the project
The app was created by @fer_adauto @cristiguzman



















