iiinova — where your digital self keeps evolving

Ondra Platek @oplatek.bsky.social · 14/07/2026

FP8 safetensors huggingface.co/bottlecapai/...

Ondra Platek @oplatek.bsky.social · 14/07/2026

bottlecapai/ThinkingCap-Qwen3.6-27B and quants! 50% less thinking tokens on average, and over 90% less in best cases. Achieved via finetuning Qwen3.6-27B (Qwen Team, 2026) with state-of-the-art finetuning algorithms on a curated set of domains with different difficulty. I love fp8 quant 😁

420

Ondra Platek @oplatek.bsky.social · 12/05/2026

Still listening to radio stations? Someone? I still have a station I like to listen -- Radio Beat [1] . I just want it in an app. So the stream plays nicely with the car navigation. No adds, with playback. I crave nostalgia without the negatives 🤣 [1] www.radiobeat.cz/cs/71/on-lin...

radiobeat.cz

ON-LINE VYSÍLÁNÍ | Rádio BEAT: První bigbít u nás

Oficiální internetové stránky rozhlasové stanice Radio BEAT, která Vám přináší to nejlepší z klasického rocku

110

Reposted by Ondra Platek

Institute of Formal and Applied Linguistics @ufal.mff.cuni.cz · 28/03/2026

LLMs as Span Annotators: A Comparative Study of LLMs and Humans by @zdenekkasner.cz, @zouharvi.bsky.social, @patuchen.bsky.social, @ivankartac.bsky.social, K. Onderková, @oplatek.bsky.social, Dimitra Gkatzia, @saad.me.uk , @tuetschek.bsky.social & Simone Ballocu aclanthology.org/2026.mme-mai...

1134

Reposted by Ondra Platek

Ivan Kartáč @ivankartac.bsky.social · 24/03/2026

New preprint: arxiv.org/abs/2603.20133 Does performance on reasoning benchmarks transfer to real-world settings such as task-oriented dialogue? Not necessarily: our new benchmark tests LLMs on problems framed in both standalone and dialogue settings, and shows that dialogue makes reasoning harder.

1176

Ondra Platek @oplatek.bsky.social · 29/01/2026

Endure: Mind body and the curiously elastic limits of human performance by Alex Hutchinson. Good theory before to learn anything from any Dummies series book

000

Ondra Platek @oplatek.bsky.social · 29/01/2026

Interesting did not know that arxiv requires peer-reviewed papers arxiv.org/abs/2601.17036

030

Ondra Platek @oplatek.bsky.social · 29/01/2026

Our Pulse News App -- our first app is live on iOS today. We also announce star investors cap. See the announcement video to see our plans. If you are LLM guru at research or product building, join us to make the plans a reality youtu.be/cA8-XJeoZAQ?...

youtu.be

BottleCapAI Announcement, Seed Round, Pulse: Community News

YouTube video by BottleCapAI

330

Ondra Platek @oplatek.bsky.social · 29/01/2026

Thank you to all coauthors to push the article to acceptance!

000

Ondra Platek @oplatek.bsky.social · 29/01/2026

In general using LLM annotations for self-improvement is great future direction

120

Ondra Platek @oplatek.bsky.social · 29/01/2026

Annotations using LLM are pretty under researched topics. I would be interested how it performs on coding tasks and if it helps during CoT for coding tasks

100

Ondra Platek @oplatek.bsky.social · 29/01/2026

Well done @zdenekkasner.bsky.social et al! LLMs as Span Annotators: A Comparative Study of LLMs and Humans is accepted to multilingual-multicultural-evaluation.github.io 🎉 See paper arxiv.org/abs/2504.08697

282

Ondra Platek @oplatek.bsky.social · 29/01/2026

What you plan to read on vacation?

100

Ondra Platek @oplatek.bsky.social · 29/01/2026

Dr.LLM: Adaptive Layer Routing for Efficient Inference Avoiding computation is pretty obvious way how to be more efficient. The question is how to maintain quality. MoE and router is an efficient approach but suffer from exposure bias fixed thresholds. arxiv.org/abs/2510.12773

320

Ondra Platek @oplatek.bsky.social · 29/01/2026

What do you want to do with the deeper layers

010

Ondra Platek @oplatek.bsky.social · 29/01/2026

Do you think future is MoE or dense models?

100

Ondra Platek @oplatek.bsky.social · 29/01/2026

Qwen3 Embedder is amazing but it is tricky to use. Left padding confused lot of early adopters. However, clearly it is the future. Reuse of pre-trained LLM for everything. Including vector search where BERT-like models ruled. 1/N arxiv.org/abs/2506.05176 huggingface.co/Qwen/Qwen3-E...

510

Ondra Platek @oplatek.bsky.social · 29/01/2026

So would you like to prune them?

000

Ondra Platek @oplatek.bsky.social · 29/01/2026

The no free lunch idea is here again. Saw it in TTS models where Guided Attention worked very well. Idea is simple: For speech conversion prefer guide the attention to focus on narrow context. It learns faster. Looking forward to deep dive on this one - expect similarities arxiv.org/abs/2601.15165

arxiv.org

The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models

Diffusion Large Language Models (dLLMs) break the rigid left-to-right constraint of traditional LLMs, enabling token generation in arbitrary orders. Intuitively, this flexibility implies a solution sp...

110

Ondra Platek @oplatek.bsky.social · 29/01/2026

It can be used in RAG, so for fetching relevant documents in clawbot it makes very much sense

000

Ondra Platek @oplatek.bsky.social · 29/01/2026

The last token pooling is compressing the embeddings too much. However it is engineering marvel. Including tricks like the spherical last checkpoint averaging is awesome.

000

Ondra Platek @oplatek.bsky.social · 07/11/2025

👀 WE ARE LOOKING FOR YOU! (yes, you!) 🚨NoCap test deadline 11/11/2025 🏆 $3K / $2K / $1K prizes Our new team members shared tips&tricks which helped them to go through our test! TRY IT NOW! 👉 github.com/BottleCapAI/...

lnkd.in

This link will take you to a page that’s not on LinkedIn

100

Ondra Platek @oplatek.bsky.social · 26/09/2025

The right words at the right time with the right style for the venue. Proud of our president 🇨🇿 youtu.be/d3wT84egi-g?...

youtu.be

Czech President Delivers Powerful Speech at UN, Slams Russia, Iran, Israel Aggression | AC1G

YouTube video by DRM News

010

Ondra Platek @oplatek.bsky.social · 22/09/2025

Me: Scaling algorithms is just much more fun, than just burning electricity on more and more GPUs.

000

Ondra Platek @oplatek.bsky.social · 22/09/2025

Official BottleCapAI: we believe AI shouldn’t cost tens of millions to train. We are now opening challenge for those who want to help with that! 🏆 $3K / $2K / $1K prizes ⏰ Deadline: 11/11/2025 🖥️ 1 GPU. Your ideas. 👉 Join the NoCap Test: github.com/BottleCapAI/...

github.com

GitHub - BottleCapAI/NoCap-Test: Open Test for BottleCapAI

Open Test for BottleCapAI. Contribute to BottleCapAI/NoCap-Test development by creating an account on GitHub.

110

Reposted by Ondra Platek

Kirill Semenov @kiryukhasemenov.bsky.social · 06/06/2025

📣Take part in 3rd Terminology shared task @WMT!📣 This year: 👉5 language pairs: EN->{ES, RU, DE, ZH}, 👉2 tracks - sentence-level and doc-level translation, 👉authentic data from 2 domains: finance and IT! www2.statmt.org/wmt25/termin... Don't miss an opportunity - we only do it once in two years😏

www2.statmt.org

Terminology Translation Task

032

Ondra Platek @oplatek.bsky.social · 04/05/2025

I just found awesome channel on YouTube. It has 77k subscribers with just 7 videos. Because the videos are just awesome! I wish I could explain stuff that simply! Does anybody know how long take to prepare such video? www.youtube.com/@algorithmic...

youtube.com

Algorithmic Simplicity

This is an educational channel for all things algorithmic, including but not limited to computer science, machine learning, physics, and mathematics.

000

Reposted by Ondra Platek

Zdeněk Kasner @zdenekkasner.cz · 15/04/2025

Thanks to all my awesome collaborators! 👉️ Vilém @zouharvi.bsky.social 👉️ Patrícia @patuchen.bsky.social 👉️ Ivan @ivankartac.bsky.social 👉️ Kristýna Onderková 👉️ Ondřej P. @oplatek.bsky.social 👉️ Dimitra @dimitrag.bsky.social 👉️ Saad @saad.me.uk 👉️ Ondřej D. @tuetschek.bsky.social 👉️ Simone Balloccu

153

Reposted by Ondra Platek

Zdeněk Kasner @zdenekkasner.cz · 15/04/2025

How 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/2504.08697

llm-span-annotators.github.io

Large Language Models as Span Annotators

Website for the paper Large Language Models as Span Annotators

1207

Reposted by Ondra Platek

Institute of Formal and Applied Linguistics @ufal.mff.cuni.cz · 07/02/2025

We've been making the media rounds! 👉📺 @hajicjan.bsky.social talked about the new OpenEuroLLM project on Czech TV's Studio 6 www.ceskatelevize.cz/porady/10969... 👉📻 @tuetschek.bsky.social discussed #LLMs on Czech Radio radiozurnal.rozhlas.cz/proc-umela-i...).

ceskatelevize.cz

Čech povede evropský výzkum AI - 3. února 05:59 - Studio 6 | Česká televize

Jan Hajič

073

Reposted by Ondra Platek

Institute of Formal and Applied Linguistics @ufal.mff.cuni.cz · 25/02/2025

Congrats to ✨Šárka Zikánová✨on her promotion to Associate Professor! 🎉 Her research explores syntax, information structure, and discourse relations in Czech and beyond. She contributed to key linguistic corpora and now focuses on psycholinguistic studies of discourse. ufal.mff.cuni.cz/sarka-zikanova

ufal.mff.cuni.cz

Šárka Zikánová | ÚFAL

061

Ondra Platek @oplatek.bsky.social · 16/01/2025

I got 503 error for this link 🤔

100

Reposted by Ondra Platek

Institute of Formal and Applied Linguistics @ufal.mff.cuni.cz · 19/11/2024

Hello Blue Sky! 👋 This is the official account of the Institute of Formal and Applied Linguistics (ÚFAL for short) at the Faculty of Mathematics and Physics, Charles University in Prague 🇨🇿. Here, we will share news from the life of the institute and our members.

0165

Reposted by Ondra Platek

Institute of Formal and Applied Linguistics @ufal.mff.cuni.cz · 28/11/2024

Today is the Open Doors Day🚪at Matfzy, Charles University. @rur.bsky.social and @zdenekkasner.bsky.social chat with high schoolers about #LLMs and #LanguageTechnology... @rudolf_rosa shows a demo of automatic interpreting 🗣️🎙️🌐 by Dominik Macháček and Ondřej Bojar.

093

Ondra Platek @oplatek.bsky.social · 20/11/2024

Hello Bsky! Let me introduce github.com/ufal/factgenie A Span Annotation tool from @ufal-cuni.bsky.social 💡 Annotate errors in the outputs & w/out any references 💪 Use many ready-to-use datasets for data-to-text. MT & NLP friends are coming... 🚀 You get your annotations in minutes.

060

Reposted by Ondra Platek

Zdeněk Kasner @zdenekkasner.cz · 14/11/2024

When a model output gets a BLEU score of 12.34: 😶 When you can *see* the errors in the output: 🤯 Check out factgenie, our new error span annotation tool: github.com/ufal/factgenie (Just matured into v1.0.1! 🎂) #nlp #nlg

062