Ece Takmaz

@barbaraplank.bsky.social

20 Sept

I will be attending EMNLP in Budapest! 😊🇭🇺 Here is where you might find me 👀

Ece Takmaz reposted

Barbara Plank

Group picture MaiNLP/CIS with friends at EMNLP 2026 🙌 We look forward to a great conference 🤩 @mainlp.bsky.social @cislmu.bsky.social @munichcenterml.bsky.social @lmu.de

Ece Takmaz reposted

MaiNLP lab, LMU Munich

@mainlp.bsky.social

MaiNLP is headed to #EMNLP2026 with a number of posters and talks! See you in Budapest ✨

The 16th edition of the Workshop on Cognitive Modeling and Computational Linguistics (CMCL) will be co-located with COLING 2027! Stay tuned ✨ @cmclworkshop.bsky.social

@grambelli.bsky.social @byungdoh.bsky.social Tatsuki Kuribayashi

babylm.github.ioBabyLM 4 at EMNLP 2026

I’m happy to share that I’m starting a new position as Interim Professor for AI and Computational Linguistics at @lmu.de 😊🇩🇪🎉

Ece Takmaz reposted

abrsvn.bsky.social

@abrsvn.bsky.social

18 Sept

New paper at babylm.github.io on "Relational Attention for Data-Efficient Language Modeling" with @ecekt.bsky.social and Jakub Dotlačil, building on the seminal work of @taylorwwebb.bsky.social , @awni.bsky.social and others. Paper: arxiv.org/abs/2609.20530. Code: github.com/abrsvn/babyl...

24 Aug

Our workshop MINT @ EMNLP 2026 welcomes ARR-reviewed submissions on multimodal interaction in face-to-face dialogue! Deadline: August 31 🍃 @emnlpmeeting.bsky.social mintworkshop.github.io/2026/

cpl2026.sites.uu.nlComputational Psycholinguistics Meeting 2026 | Universiteit Utrecht

27 June

Computational Psycholinguistics Meeting (CPL 2026) has an updated submission deadline! (July 17) ☀️

Ece Takmaz@ecekt.bsky.social

After the very nice 15th edition of CMCL last Saturday 🌴, I'm happy to announce that the 2nd edition of the Computational Psycholinguistics Meeting (CPL 2026) will take place again in Utrecht in December! We are looking forward to receiving abstracts ✨ cpl2026.sites.uu.nl

mintworkshop.github.ioRedirecting…

26 June

MINT deadline extended to July 22nd!

Ece Takmaz@ecekt.bsky.social

🍃2nd Call for Papers!😊 EMNLP 2026 Workshop on Multimodal Interaction in Face-to-Face Dialogue (MINT) welcomes work on multimodal communication, speech, gesture, gaze, facial expressions, datasets, eval methods, and applications! @emnlpmeeting.bsky.social mintworkshop.github.io

mintworkshop.github.ioRedirecting…

21 May

cpl2026.sites.uu.nlComputational Psycholinguistics Meeting 2026 | Universiteit Utrecht

18 May

Ece Takmaz reposted

Emiel van Miltenburg

@evanmiltenburg.bsky.social

10 May

The International conference on Natural Language Generation (INLG) just issued the first Call For Papers: 2026.inlgmeeting.org/calls.html #nlproc

2026.inlgmeeting.orgINLG2026The 19th International Natural Language Generation Conference is scheduled to be held in Utrecht, the Netherlands from October 17 to 21, 2026.

10 May

If you are attending LREC this week, check out CMCL (15th Workshop on Cognitive Modeling and Computational Linguistics) on May 16! We have a great program with lots of nice papers, and @claires012345.bsky.social and @coryshain.bsky.social as invited speakers! 💫🌴 I will be following it virtually😊

2026.inlgmeeting.orgINLG2026The 19th International Natural Language Generation Conference is scheduled to be held in Utrecht, the Netherlands from October 17 to 21, 2026.

10 May

Some of my fellow organizers are planning to be there! @byungdoh.bsky.social @grambelli.bsky.social @jixingli.bsky.social Tatsuki Kuribayashi, Philipp Wicke, Ryo Yoshida

Ece Takmaz reposted

INLG 2026

@inlg.bsky.social

3 Apr

📍It's official! #INLG2026 is coming to Utrecht, Netherlands, Oct 17-21! Hosted with support from Utrecht University and the local NLP community. Follow us here and check 2026.inlgmeeting.org for updates -- hope to see you there!

mintworkshop.github.ioRedirecting…

30 Mar

We are announcing a very new workshop! 🌿 MINT @ EMNLP 2026, 1st Workshop on Multimodal Interaction in Face-to-Face Dialogue! ✨ mintworkshop.github.io Raquel Fernández, Diego Frassinelli, @esamghaleb.bsky.social, Bulat Khaertdinov, @asliozyurek.bsky.social Zerrin Yumak @emnlpmeeting.bsky.social

26 Feb

CMCL deadline extended to Feb 28 AoE!

Ece Takmaz@ecekt.bsky.social

The submission deadline for CMCL is coming up in less than a month! (Feb 25) CMCL will be co-located with LREC and take place on May 16!🌴https://sites.google.com/view/cmclworkshop/cfp

28 Jan

The submission deadline for CMCL is coming up in less than a month! (Feb 25) CMCL will be co-located with LREC and take place on May 16!🌴https://sites.google.com/view/cmclworkshop/cfp

Ece Takmaz@ecekt.bsky.social

Cognitive Modeling and Computational Linguistics (CMCL) workshop will be co-located with LREC 2026 in Palma, Mallorca!🌴Stay tuned for more details!✨ @byungdoh.bsky.social Tatsuki Kuribayashi @grambelli.bsky.social Philipp Wicke, Jixing Li, Ryo Yoshida @cmclworkshop.bsky.social

18 Jan

I have just returned from a week-long visit to Bielefeld University! Thank you very much for hosting me Sina Zarrieß and @ozgealacam.bsky.social 😊 @clausebielefeld.bsky.social

CLAUSE - Computational Linguistics @ Bielefeld University

18 Jan

I had so many inspiring conversations with lovely colleagues and I am already looking forward to visiting again in the future! Many thanks to @simeonjunker.bsky.social, @bbunzeck.bsky.social, @manarali.bsky.social, @hbuschme.bsky.social, Clara Lachenmaier, Lisa Gottschalk, Emilie Sitter, Yu Wang ✨

Ece Takmaz reposted

@clausebielefeld.bsky.social

13 Jan

This week we’re having @ecekt.bsky.social as our guest in Bielefeld. She gave a highly timely talk on language+vision models, how they process images under noise conditions, and about how to train a highly effective multimodal BabyLM with model merging. 🗣️👀💻

21 Dec 2025

Photos from the Computational Psycholinguistics Meeting in Utrecht, many thanks to everyone who joined us in making this a memorable event! ✨

21 Dec 2025

21 Dec 2025

scholar.google.comscholar.google.com

15 Dec 2025

The CfP for CMCL is out!🌴 We are looking forward to receiving many interesting submissions! ✨ (Deadline: February 25, 2026) sites.google.com/view/cmclwor...

Ece Takmaz@ecekt.bsky.social

9 Dec 2025

Many thanks to @dnliu.bsky.social for inviting me, and to the members of the group for their insightful questions! 😊✨

AI4LT@ai4lt.bsky.social

Last week @ecekt.bsky.social gave a great talk on her recent work "Model Merging to Maintain Language-Only Performance in Developmentally Plausible Multimodal Models"! Thank you so much for joining us, Ece! 🌟

4 Dec 2025

The program of the Computational Psycholinguistics Meeting 2025 at Utrecht University is out, packed with a lot of very interesting talks! The registration is full, but there is a waiting list, if you would like to attend✨ cpl2025.sites.uu.nl/schedule/

I was in Sweden this week! 🇸🇪❄️ Many thanks to Nikolai Ilinykh for inviting me to give a talk at the University of Gothenburg. I enjoyed having inspiring chats and delicious food with Sharid Loáiciga, @asayeed.bsky.social, Simon Dobnik, Hyewon Jang and Chris Howes at CLASP. Much appreciated! 😊🎄

I will be attending EMNLP in China to present our paper with @bylinina.bsky.social (who will be in China, too) and Jakub Dotlacil in the BabyLM workshop! Looking forward to meeting people there! ✨ 😊 #EMNLP2025 @emnlpmeeting.bsky.social lnkd.in/e-Bzz6De

Instead of using the data provided in the BabyLM challenge, I opted for obtaining them from their sources, which added extra layers of filtering and complexity, revealing some discrepancies in the multimodal BabyLM data. I mention these in the paper.

aclanthology.orgModel Merging to Maintain Language-Only Performance in Developmentally Plausible Multimodal ModelsEce Takmaz, Lisa Bylinina, Jakub Dotlacil. Proceedings of the First BabyLM Workshop. 2025.

I hope our findings would be helpful for the future contributors to the multimodal track of the BabyLM challenge! aclanthology.org/2025.babylm-...

The report on the Findings of the Third BabyLM Challenge indicates that the multimodal track received only 1 full submission this year. We submitted our paper to the workshop track instead of the challenge.

Unfortunately, we had limited time and resources to modify the whole evaluation pipeline for our specific multimodal architecture. As a result, we tested our models on a subset of the benchmarks.

How can we mitigate this issue in developmentally plausible multimodal models and maintain language-only performance? We explored model merging, a technique that has been shown to benefit multi-task and multi-language models, reducing the effects of catastrophic forgetting.

We experiment with weighted linear interpolation of language-only and multimodal model weights. Model merging with language-only checkpoints helps alleviate the issue to some extent, benefiting performance in language-only benchmarks and not disrupting accuracy in multimodal tasks heavily.

Previous work, including BabyLM contributions, indicates that multimodal data has limited or no benefits in text-only benchmarks. We reach similar conclusions in our low-resource multimodal scenario.

Our multimodal BabyLM model surpasses previous multimodal baselines and submissions on the leaderboard. Yet, compared to language-only models, it underperforms in grammar-oriented benchmarks, although being exposed to the same language-only data as the language-only models (+ multimodal data).

27 Oct 2025

Just got back from Hawaii, where I presented a workshop paper on image memorability at @iccv.bsky.social 🌺 Coming from multimodal NLP, it was my first time attending a CV conference. Everywhere I looked, there were talks and posters that were incredibly interesting!

arxiv.orgTraces of Image Memorability in Vision Encoders: Activations, Attention Distributions and Autoencoder LossesImages vary in how memorable they are to humans. Inspired by findings from cognitive science and computer vision, this paper explores the correlates of image memorability in pretrained vision encoders...

27 Oct 2025

I felt very much at home at #ICCV2025! Here is the paper: arxiv.org/abs/2509.01453

Ece Takmaz reposted

Jaap Jumelet

@jumelet.bsky.social

15 Oct 2025

🌍Introducing BabyBabelLM: A Multilingual Benchmark of Developmentally Plausible Training Data! LLMs learn from vastly more data than humans ever experience. BabyLM challenges this paradigm by focusing on developmentally plausible data We extend this effort to 45 new languages!

15 Oct 2025

What makes an image memorable? And can we predict image memorability using pretrained vision encoders? We explored activations, attention distributions, image patch uniformity and sparse autoencoder losses using image representations across the layers of CLIP, DINOv2 and SigLIP2.

15 Oct 2025

I will be presenting this work at the @iccv.bsky.social 2025 workshop MemVis: The 1st Workshop on Memory and Vision! 🌺 Work done with Albert Gatt & Jakub Dotlacil arxiv.org/abs/2509.01453