Dayeon (Zoey) Ki @dayeonki.bsky.social · 13/06/20251/ Are two #LLMs better than one for equitable cultural alignment? 🌍 We introduce a Multi-Agent Debate framework — where two LLM agents debate the cultural adaptability of a given scenario. #ACL2025 🧵👇 170
Reposted by Dayeon (Zoey) KiVilém Zouhar @zouhar.bsky.social · 02/12/2024Trying to collect all the MT people here. I probably missed many. Ping me! bsky.app/starter-pack... 9268
Dayeon (Zoey) Ki @dayeonki.bsky.social · 21/05/20251/ How can a monolingual English speaker 🇺🇸 decide if an automatic French translation 🇫🇷 is good enough to be shared? Introducing ❓AskQE❓, an #LLM-based Question Generation + Answering framework that detects critical MT errors and provides actionable feedback 🗣️ #ACL2025 112
Reposted by Dayeon (Zoey) KiMyra Cheng @myra.bsky.social · 02/05/2025How does the public conceptualize AI? Rather than self-reported measures, we use metaphors to understand the nuance and complexity of people’s mental models. In our #FAccT2025 paper, we analyzed 12,000 metaphors collected over 12 months to track shifts in public perceptions. 34914
Reposted by Dayeon (Zoey) KiVilém Zouhar @zouhar.bsky.social · 01/05/2025Multilinguality is happening at #NAACL2025 @crystinaz.bsky.social @oxxoskeets.bsky.social @dayeonki.bsky.social @onadegibert.bsky.social 0141
Reposted by Dayeon (Zoey) KiAngel Hsing-Chi Hwang @angelhwang.bsky.social · 18/04/2025Starting my journey on Bluesky with a topic that I care deeply about: AI tools can support creators in various ways, but disclosing AI use may risk devaluing creative work. Check out our abstract here: angelhwang.github.io/doc/ic2s2_AI... Inspired by our past work: arxiv.org/abs/2411.13032arxiv.org"It was 80% me, 20% AI": Seeking Authenticity in Co-Writing with Large Language ModelsGiven the rising proliferation and diversity of AI writing assistance tools, especially those powered by large language models (LLMs), both writers and readers may have concerns about the impact of th... 1275
Dayeon (Zoey) Ki @dayeonki.bsky.social · 17/04/2025🚨 New Paper 🚨 1/ We often assume that well-written text is easier to translate ✏️ But can #LLMs automatically rewrite inputs to improve machine translation? 🌍 Here’s what we found 🧵 184
Reposted by Dayeon (Zoey) KiTokenization Workshop (TokShop) @COLM2026 @tokshop.bsky.social · 15/04/2025🚨 NEW WORKSHOP ALERT 🚨 We're thrilled to announce the first-ever Tokenization Workshop (TokShop) at #ICML2025 @icmlconf.bsky.social! 🎉 Submissions are open for work on tokenization across all areas of machine learning. 📅 Submission deadline: May 30, 2025 🔗 tokenization-workshop.github.iotokenization-workshop.github.ioTokenization Workshop @ ICML 2025 1247
Reposted by Dayeon (Zoey) KiShayne Longpre @shaynelongpre.bsky.social · 14/04/2025Thrilled our global data ecosystem audit was accepted to #ICLR2025! Empirically, it shows: 1️⃣ Soaring synthetic text data: ~10M tokens (pre-2018) to 100B+ (2024). 2️⃣ YouTube is now 70%+ of speech/video data but could block third-party collection. 3️⃣ <0.2% of data from Africa/South America. 1/ 1124
Reposted by Dayeon (Zoey) KiZdeněk Kasner @zdenekkasner.cz · 15/04/2025How do LLMs compare to human crowdworkers in annotating text spans? 🧑🤖 And how can span annotation help us with evaluating texts? Find out in our new paper: llm-span-annotators.github.io Arxiv: arxiv.org/abs/2504.08697llm-span-annotators.github.ioLarge Language Models as Span AnnotatorsWebsite for the paper Large Language Models as Span Annotators 1207
Reposted by Dayeon (Zoey) KiHelsinki NLP @helsinki-nlp.bsky.social · 18/03/2025Call for participation: We just opened the registration for this year's MT Marathon in August in Helsinki, Finland: blogs.helsinki.fi/language-tec..., featuring: - Ayodele Awokoya - Wilker Aziz - Marta Costa-Jussa - Barry Haddow - Amit Moryosse - Sara Papi - Jörg Tiedemann - Marco Turchiblogs.helsinki.fi 053
Reposted by Dayeon (Zoey) KiOna de Gibert @onadegibert.bsky.social · 18/03/2025Come to Helsinki for the 18th MT Marathon! Sponsored by EAMT @ufal-cuni.bsky.social 085
Reposted by Dayeon (Zoey) KiBarry Haddow @bazril.bsky.social · 28/02/2025** New parallel data set ** . We've just released HPLT v2.0, a parallel data set of 50 languages paired with English, 380M sentence pairs in total. Extracted from the Internet Archive and Common Crawl hplt-project.org/datasets/v2.0hplt-project.orgHPLT - High Performance Language TechnologiesA space that combines petabytes of natural language data with large-scale model training 143
Reposted by Dayeon (Zoey) KiAndrea Piergentili @apierg.bsky.social · 19/03/2025Brilliant and necessary work by Pombal et al. about metric interference in MT system development and evaluation: arxiv.org/abs/2503.08327 Are we developing better systems or are we just gaming the metrics? And how do we address this? Super (m)interesting! 👀arxiv.orgAdding Chocolate to Mint: Mitigating Metric Interference in Machine TranslationAs automatic metrics become increasingly stronger and widely adopted, the risk of unintentionally "gaming the metric" during model development rises. This issue is caused by metric interference (Mint)... 0101
Reposted by Dayeon (Zoey) KiYixiao Song @yixiaosong.bsky.social · 12/03/2025Introducing 🐻 BEARCUBS 🐻, a “small but mighty” dataset of 111 QA pairs designed to assess computer-using web agents in multimodal interactions on the live web! ✅ Humans achieve 85% accuracy ❌ OpenAI Operator: 24% ❌ Anthropic Computer Use: 14% ❌ Convergence AI Proxy: 13% 1115
Reposted by Dayeon (Zoey) KiSiyuan Song @siyuansong.bsky.social · 12/03/2025New preprint w/ @jennhu.bsky.social @kmahowald.bsky.social : Can LLMs introspect about their knowledge of language? Across models and domains, we did not find evidence that LLMs have privileged access to their own predictions. 🧵(1/8) 26116
Reposted by Dayeon (Zoey) KiMiriam Posner @miriamposner.com · 06/03/2025OK, every year I try to explain to my students how LLMs work, and every year I have to do a big trawl for good resources and activities. Here's this year's haul of *introductory* materials. (In-class activities + visualizations, not so much readings.) 66771191
Reposted by Dayeon (Zoey) KiTom Kocmi @kocmitom.bsky.social · 11/03/2025Big news from WMT! 🎉 We are expanding beyond MT and launching a new multilingual instruction shared task. Our goal is to foster truly multilingual LLM evaluation and best practices in automatic and human evaluation. Join us and build the winning multilingual system! www2.statmt.org/wmt25/multil...www2.statmt.orgMultilingual Instruction Shared Task 1127
Reposted by Dayeon (Zoey) KiArtjoms Šeļa @artjomshl.bsky.social · 11/03/2025self-insert, but if you are looking for something multilingual and public domain, we have PoeTree: a collection of poetry corpora with Python & R access points (can get data directly into your jupyter notebook) : versologie.cz/poetree/versologie.czPoeTree. Poetry Treebanks in 10 languagesPoeTree is a standardized collection of poetry corpora comprising over 330,000 poems in ten languages (Czech, English, French, German, Hungarian, Italian, Portuguese, Russian, Slovenian, Spanish). 1112
Reposted by Dayeon (Zoey) KiAaron Mueller @amuuueller.bsky.social · 11/03/2025Lots of work coming soon to @iclr-conf.bsky.social and @naaclmeeting.bsky.social in April/May! Come chat with us about new methods for interpreting and editing LLMs, multilingual concept representations, sentence processing mechanisms, and arithmetic reasoning. 🧵 1206
Reposted by Dayeon (Zoey) KiNishant Balepur @nbalepur.bsky.social · 11/03/2025🚨 Our team at UMD is looking for participants to study how #LLM agent plans can help you answer complex questions 💰 $1 per question 🏆 Top-3 fastest + most accurate win $50 ⏳ Questions take ~3 min => $20/hr+ Click here to sign up (please join, reposts appreciated 🙏): preferences.umiacs.umd.edu 023
Reposted by Dayeon (Zoey) KiKathy @kathaem.bsky.social · 03/03/2025Happy to say that our paper "Beyond Literal Token Overlap: Token Alignability for Multilinguality" will be presented at #NAACL2025! This is work with @tomlim.bsky.social, @jlibovicky.bsky.social, and Alex Fraser. arxiv.org/abs/2502.06468 #newpaper #NLP #NLProcarxiv.orgBeyond Literal Token Overlap: Token Alignability for MultilingualityPrevious work has considered token overlap, or even similarity of token distributions, as predictors for multilinguality and cross-lingual knowledge transfer in language models. However, these very li... 1102
Reposted by Dayeon (Zoey) KiCatherine Arnett @catherinearnett.bsky.social · 07/03/2025✨New pre-print✨ Crosslingual transfer allows models to leverage their representations for one language to improve performance on another language. We characterize the acquisition of shared representations in order to better understand how and when crosslingual transfer happens. 2367
Reposted by Dayeon (Zoey) KiJess Hamrick @jhamrick.bsky.social · 07/03/2025This is a really neat use case for AI—checking whether claims are actually supported by the given citations. 1256
Reposted by Dayeon (Zoey) KiKarolina Stańczak @karstanczak.bsky.social · 04/03/2025📢New Paper Alert!🚀 Human alignment balances social expectations, economic incentives, and legal frameworks. What if LLM alignment worked the same way?🤔 Our latest work explores how social, economic, and contractual alignment can address incomplete contracts in LLM alignment🧵 12713
Reposted by Dayeon (Zoey) KiCohere Labs @cohereforai.bsky.social · 04/03/2025We hope Aya enables researchers and developers throughout the world to build upon this technology, ask deeper questions about multilingual AI, and develop tools that can support their communities. Learn more: cohere.com/blog/aya-vis...cohere.comAya Vision: Expanding the Worlds AI Can SeeOur state-of-the-art open-weights vision model offers a foundation for AI-enabled multilingual and multimodal communication globally. Today, Cohere For AI, Cohere’s open research arm, is proud to an... 021
Reposted by Dayeon (Zoey) KiMarzieh Fadaee @mziizm.bsky.social · 04/03/2025✨👓 Aya Vision is here 👓✨ A multilingual, multimodal model designed to understand across languages and modalities (text, images, etc) to bridge the language gap and empower global users! 142
Reposted by Dayeon (Zoey) KiTom Kocmi @kocmitom.bsky.social · 01/03/2025Huge shoutout to colleagues at Google & Unbabel for extending our WMT24 testset to 55 languages in four domains, this is game changer! 🚀 I really hope it puts the final nail in the coffin of FLORES or WMT14. The field is evolving, legacy testsets can't show your progress arxiv.org/abs/2502.124...arxiv.orgWMT24++: Expanding the Language Coverage of WMT24 to 55 Languages & DialectsAs large language models (LLM) become more and more capable in languages other than English, it is important to collect benchmark datasets in order to evaluate their multilingual performance, includin... 0146
Reposted by Dayeon (Zoey) KiKayo Yin @kayoyin.bsky.social · 28/02/2025Induction heads are commonly associated with in-context learning, but are they the primary driver of ICL at scale? We find that recently discovered "function vector" heads, which encode the ICL task, are the actual primary mechanisms behind few-shot ICL! arxiv.org/abs/2502.14010 🧵👇 1227
Reposted by Dayeon (Zoey) KiTejas Srinivasan @tejassrinivasan.bsky.social · 27/02/2025People are increasingly relying on AI assistance, but *how* they use AI advice is influenced by their trust in the AI, which the AI is typically blind to. What if they weren’t? We show that adapting AI assistants' behavior to user trust mitigates under- and over-reliance! arxiv.org/abs/2502.13321 182
Reposted by Dayeon (Zoey) Kinaitian @naitian.org · 18/02/2025There's been a lot of work on "culture" in NLP, but not much agreement on what it is. A position paper by me, @dbamman.bsky.social, and @ibleaman.bsky.social on cultural NLP: what we want, what we have, and how sociocultural linguistics can clarify things. Website: naitian.org/culture-not-... 1/n 512234
Reposted by Dayeon (Zoey) KiShayne Longpre @shaynelongpre.bsky.social · 26/02/2025Thrilled to be at #AAAI2025 for our tutorial, “AI Data Transparency: The Past, Present, and Beyond.” We’re presenting the state of transparency, tooling, and policy, from the Foundation Model Transparency Index, Factsheets, the the EU AI Act to new frameworks like @MLCommons’ Croissant. 1/ 1154
Reposted by Dayeon (Zoey) KiAkhila Yerukola @akhilayerukola.bsky.social · 26/02/2025Did you know? Gestures used to express universal concepts—like wishing for luck—vary DRAMATICALLY across cultures? 🤞means luck in US but deeply offensive in Vietnam 🚨 📣 We introduce MC-SIGNS, a test bed to evaluate how LLMs/VLMs/T2I handle such nonverbal behavior! 📜: arxiv.org/abs/2502.17710 1337
Reposted by Dayeon (Zoey) KiChau Minh Pham @chautmpham.bsky.social · 21/02/2025⚠️Current methods for generating instruction-following data fall short for long-range reasoning tasks like narrative claim verification. We present CLIPPER ✂️, a compression-based pipeline that produces grounded instructions for ~$0.5 each, 34x cheaper than human annotations. 1218
Reposted by Dayeon (Zoey) KiNishant Balepur @nbalepur.bsky.social · 24/02/2025🚨 New Position Paper 🚨 Multiple choice evals for LLMs are simple and popular, but we know they are awful 😬 We complain they're full of errors, saturated, and test nothing meaningful, so why do we still use them? 🫠 Here's why MCQA evals are broken, and how to fix them 🧵 24612
Reposted by Dayeon (Zoey) KiStella Li @stellali.bsky.social · 21/02/2025Asking the right questions can make or break decisions in fields like medicine, law, and beyond✴️ Our new framework ALFA—ALignment with Fine-grained Attributes—teaches LLMs to PROACTIVE seek information through better questions through **structured rewards**🏥❓ (co-led with @jiminmun.bsky.social) 👉🏻🧵 1247
Reposted by Dayeon (Zoey) KiHal Daumé III @haldaume3.bsky.social · 10/12/2024starter pack for the Computational Linguistics and Information Processing group at the University of Maryland - get all your NLP and data science here! go.bsky.app/V9qWjEi 12912
Reposted by Dayeon (Zoey) KiMarzieh Fadaee @mziizm.bsky.social · 08/12/2024Aya Expanse technical report is here and I couldn’t be prouder! 🎉 Check out the report: arxiv.org/abs/2412.04261arxiv.orgAya Expanse: Combining Research Breakthroughs for a New Multilingual FrontierWe introduce the Aya Expanse model family, a new generation of 8B and 32B parameter multilingual language models, aiming to address the critical challenge of developing highly performant multilingual ... 043
Reposted by Dayeon (Zoey) KiMarc Marone @marcmarone.com · 23/11/2024I noticed a lot of starter packs skewed towards faculty/industry, so I made one of just NLP & ML students: go.bsky.app/vju2ux Students do different research, go on the job market, and recruit other students. Ping me and I'll add you! 10117654
Reposted by Dayeon (Zoey) KiSara Hooker @sarahooker.bsky.social · 05/12/2024Is MMLU Western-centric? 🤔 As part of a massive cross-institutional collaboration: 🗽Find MMLU is heavily overfit to western culture 🔍 Professional annotation of cultural sensitivity data 🌍 Release improved Global-MMLU 42 languages 📜 Paper: arxiv.org/pdf/2412.03304 📂 Data: hf.co/datasets/Coh... 75912
