Do you think Anthropic's PR were worried they miss out the positive hype of openAI breaking laws and attacking harmfully two companies. So they ran to look for places to say they also do it? (and pretend remorseful) "Everything you can do I can do" better? apnews.com/article/anth...

Anthropic says its AI models hacked 3 organizations during testing

Anthropic says its AI models hacked into three organizations during testing. This comes just days after OpenAI said its AI models went rogue and hacked into another company.

apnews.com

In September, I will be visiting Ghana.🇬🇭 I always like meeting new people and cultures, so exciting for me Should I meet someone there? I also wanted to give a talk or have 1x1s but got no response from Prof.

This is so easy why don't more link sharing channels just share it with everyone interested?

@colab-links.bsky.social · last wk.

Multilingual cultural dataset 6 european languages https://arxiv.org/abs/2607.21445 many multiple choice hard cultural QA in 288 topics (3k questions) — LChoshen

Sad to see no African representation in the last translation benchmark, despite LLMs and MT being so bad If you know who would be interested or are interested yourself in contributing(coauthoring), please do and share last-translation-benchmark.vilda.net Viz: lvu5.github.io/translation_...

Following recent requests, I made a bit of organizing on tools one can use with papers (searching, looking for trends, etc.) Enjoy https://borgr.github.io/paper_updated/ — LChoshen

OpenEvolve underperforms simple autmated discovery harnesses the rest are insignificant from each other. The best choice changed across model–problem pairs. We ran a controlled study (3m+ rollouts) using repeated budget-matched runs, strong baselines, and statistical hypothesis testing.

Machine translation is solved right? Well almost, we just need to figure this one thing. Share the one thing?

Vilém Zouhar@zouharvi.bsky.social · 3w ago

There are many things machine translation still can't do. Help us steer the next direction by contributing hard-to-translate inputs (and be on a cool paper).

There are many things machine translation still can't do. Help us steer the next direction by contributing hard-to-translate inputs (and be on a cool paper).

Is Gemma 4 multilingual? Not really🤖 A true multilingual LLM should share language-invariant skills like spatial understanding across languages. It doesnt😱 We had LLMs play 2D board games against themselves in different languages🕹️ and found 2D performance is language dependent

What cool posters did you see in acml? (Content and design both work) openreview.net/forum?id=CbJ...

Sota on 89 languages speech models. There's plenty of speech data it appears, so the simplest fine-tuning plus tokenization on public data just improves everything substantially. And it's not even with any tricks or all the data... #conll #acl @catherinearnett.bsky.social @alexwarstadt.bsky.social

Website: vamsin07.github.io/buzzasr-docs/ Tokenizers: github.com/vamsin07/mul... SFT models: github.com/vamsin07/whi...

Sota on 89 languages speech models. There's plenty of speech data it appears, so the simplest fine-tuning plus tokenization on public data just improves everything substantially. And it's not even with any tricks or all the data... #conll #acl @catherinearnett.bsky.social @alexwarstadt.bsky.social

Livetweet @mcxfrank.bsky.social talk: If we care about intelligence and cognition, AI has recently allowed a change; we now have two ways to study them. AI is of course allowing us a lot that we wouldn't dare on our Children (brain surgery, never tell a child about cats...) #acl #conll

A large part of CoNLL reviewers volunteered to do so without submitting a paper. A large sign of confidence. I'll be live tweeting from #acl and #icml so forgive the possibilty high volume

I automatically agree, but then started doubting myself. If we had a billion people working on a project, or if progress is genuinely fast. Then if in 6mo what you did is irrelevant, but was an important step stone on the way, it was worth it.

Want to see all the links CoLab 🐶 is thinking about? bsky.app/profile/cola...

Yet another reason to share the evaluations you've ran with the community, HF will also host it and refer to the source. huggingface.co/blog/eee-com...

Featuring Every Eval Ever Results on Hugging Face Model Pages

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co

What is divisive about evaluations? If you had leading figures sitting and you could ask them anything (eval related) what would you ask? Soon @eval-eval.bsky.social workshop would take place @aclmeeting.bsky.social and I'm in charge of the panel. Will you help me make good use of it?

ACL 2026 Workshop on Evaluating Evaluations (EvalEval)

This workshop focuses on AI evaluation in practice, centering the tensions and collaborations between model developers and evaluation researchers and aims to surface practical insights from across the...

evalevalai.com

Effective language identification based on a tokenizer UnigramLM tokenizer already gives probabilities, testing those to identify a language is fast and effective. Whiceh leads me to wonder, can we identify language during training and affect behavior? arxiv.org/abs/2602.17655 @tpimentel.bsky.social

We pitted LLMs against themselves in different languages across games🎮 Usually when a LLM is worse at a task in some language, we go "it's a language issue". But isn't that circular? If the task is made of language, how do you tell if the model lacks skills, or just struggles to read the question? 🧵

Does anyone in my network know when one should buy B6000 vs B200? Or who to ask?

How do projects from Academia become sustainable? You've got a project, and it succeeds, then you've got technical debt and people that care. What's your options next?

Have a complaint about how arXiv works? ArXiv is ready to listen, open community for open science: www.reddit.com/r/arXivMeta/

While language is obviously solved, Google Home can't understand a word I speak to it. Text or speech to text language and cultural issues are countless. This calls for you to join, to co author and to show those model providers they should care about your language and culture as well.

Multilingual Representation Workshop @ EMNLP 2026@mrl-workshop.bsky.social · 2mo ago

After the enthusiasm of the shared task last year, we are running another shared task to create a community-made, culturally relevant multilingual benchmark! The deadline to contribute is August 1 AoE. See more details below.

Counterfactual futures are the holy Grail of, well of a lot, right? It's predicting the future plus making A/B tests. Humanities come with this question now. Can we now use AI to reimagine history and see how it would unfold differently?

AI language acquisition. Geniuses debated how humans learn language, what their stimulus is, and what kind of rules do they apply. Why is the way AI learns language, less legitimate field of study? I want the theory of AI language acquisition

I've just got hundreds of new citations in Google Scholar Apparently, you can search your name and first letter + last name and find tons of badly parsed versions of your papers. (Yes, citations don't matter, but making camera ready is boring, and I needed an escape,...)

At last, a way to unlock mulitlingual knowledge sharing in LLMs! 🌍 ​By pretraining an English/Arabic model and swapping Arabic for a word-wise, 1-to-1 translation mapping to English, we saw a massive boost in cross-lingual knowledge transfer 🚀

What do you think ICL changes?

Andreas Waldis@tresiwald.bsky.social · 3mo ago

Do instructions affect how LMs process and produce language? ☝️Not the way you think! 😲LMs barely change task information when processing a task sample. Instead, instructions shape how this information is accessed and expressed when producing output tokens. #interpretability #nlproc (1/🧵)