Anastasiia Sedova

Thrilled to suddenly spot my name on the list of Outstanding Reviewers for #EMNLP2024! 😍 #nlproc

@ViennaNLP group at #EMNLP in Miami! 🌴 #EMNLP2024 #nlproc

Stop by at our #EMNLP poster today and say hi! 🗓️ Thursday, Nov 14, 10:30-12:00 📌 Jasmine Also, find me to chat anytime about model trustworthiness, interpretability, noisy learning, or just live in beautiful Vienna! #EMNLP2024

@BaruaJosh Hi! Have you presented your poster yet? I’d love to chat!

Do LLMs encode knowledge of concept variation across languages? Can they use this knowledge to resolve ambiguity in translation? Our #EMNLP2024 paper finds a big performance gap between closed- and open-weight LLMs, but lexical rules can help transfer knowledge across models! 🧵

📰Check out the paper: arxiv.org/pdf/2407.17125… And come talk to me at EMNLP in Miami!🌴 A joint work with: @barbara_plank and Robert Litschko and Diego Frassinelli and Benjamin Roth @emnlpmeeting #EMNLP2024 #NLPRoc (5/5)🧵

Our findings: 🫧Each model has a preferred interpretation for entity types and is strongly biased toward it 🫧Models favor popular interpretations (Amazon → company) and struggle with alternative ones 🫧Models fail to reconfirm the knowledge they have just provided (4/5)🧵

🚨How well can LLMs perform under lexical ambiguity? Excited to present our #EMNLP2024 paper, where we analyze large language models' performance and self-consistency when prompted with ambiguous entities. A joint work with the awesome @MaiNLPlab! (1/5)🧵

😅

@KatyaArtemov Hi! Cannot dm you, but want to chat :)

On my way to #iclr in the beautiful Vienna! Ping me if you want to chat about data collection, data quality and benchmarking for LLMs, and artificial text detection:)

The research group Data Mining and Machine Learning at the University of Vienna is looking for a Postdoctoral Researcher in Natural Language Processing. For more details see: jobs.univie.ac.at/job/Postdoctor…

Just in time for the holidays, we are releasing some new software today from Apple machine learning research. MLX is an efficient machine learning framework specifically designed for Apple silicon (i.e. your laptop!) Code: github.com/ml-explore/mlx Docs: ml-explore.github.io/mlx/build/html…

Summary of Gemini's 60-page technical report. 1. Written in Jax and trained using TPUs. The architecture, while not explained in details, seems similar to Flamigo's. 2. Gemini Pro's performance is similar to GPT-3.5 and Gemini Ultra is reported to be better than GPT-4. Nano-1 (1.8B params) and Nano-2 (3.25B params) are designed to run on-device. 3. 32K context length. 4. Very good at understanding vision and speech. 5. Coding ability: the big jump in HumanEval compared to GPT-4 (74.4% vs. 67%), if true, is awesome. However, the Natural2Code benchmark (no leakage on the Internet) shows a much smaller gap (74.9% vs. 73.9%). 6. On MMLU: using COT@32 (32 samples) to show that Gemini is better than GPT-4 seems forced. In 5-shot setting, GPT-4 is better (86.4% vs. 83.7%). 7. No information at all on the training data, other than they ensured "all data enrichment workers are paid at least a local living wage." Full report: storage.googleapis.com/deepmind-media…

🇸🇬 Excited to be at #EMNLP2023 in Singapore! I love making new friends, so let's grab a coffee and chat about the terrific future of NLP✨

@erazumovskaia Женя! Вау, ничего себе :) однозначно пинг на кофе!

Just got to Singapore for #EMNLP2023! 🇸🇬 Looking forward to seeing old and new friends here :) I’ll be around throughout the conference! Please ping me about multilinguality and dialogue systems! Coffee and food recommendations are extremely welcome ☕️

Last week, Ben and @andst_link, @anasedova and @vassilikiKou had an invited tutorial about weak supervision at @Aalborg_uni in Copenhagen 🇩🇰 Big thanks to @johannesbjerva for the invitation!

@hanna_schrader @emnlpmeeting Thank you 🤗

@anasedova @emnlpmeeting Congrats, Nastja! 🥳

Our paper "ULF: Unsupervised Labeling Function Correction using Cross-Validation for Weak Supervision" has been accepted as a main conference paper at @emnlpmeeting #EMNLP2023 🎉 updated preprint coming out soon😊

When looking forward to #EMNLP2023 today with everyone, I also have something to share about past events 😊 two weeks ago, I had a fantastic time at #ECMLPKDD 2023 conference presenting our recent work on gradient-based outlier removal as both an oral presentation and a poster!