Filling crucial language learning gaps

Inside Praktika's conversational approach to language learning

How Praktika uses GPT-4. 1 and GPT-5.

Preserving languages for the future

How Iceland is using GPT-4 to preserve its language.

No Optimal Language Set Exists for Multilingual Instruction Tuning: Insights from a Linguistically-Informed Study

arXiv:2410. 07809v2 Announce Type: replace-cross Abstract: Multilingual instruction tuning (MIT) is challenged by the curse of multilinguality, data scarcity, and high computational cost.

Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLM

arXiv:2609. 17435v1 Announce Type: new Abstract: We submit M\'eTRON-FR, a 125M GPT-2 pretrained on 92.

A Deepdive into Aya Expanse: Advancing the Frontier of Multilinguality

Introducing The World's Largest Open Multilingual Language Model: BLOOM

Language models can explain neurons in language models

We use GPT-4 to automatically write explanations for the behavior of neurons in large language models and to score those explanations. We release a dataset of these (imperfect) explanations and scores for every neuron in GPT-2.

Towards Stress-Aware Sentence-Level Filipino G2P With Weakly-Supervised ByT5 Fine-Tuning

arXiv:2609.09974v1 Announce Type: new Abstract: Grapheme-to-phoneme conversion (G2P) refers to the task of converting a sequence of graphemes to a corresponding sequence of phonemes. While Filipino G...

$\mu^2$-Bench: A Multilingual Machine Unlearning Benchmark

arXiv:2609. 20945v1 Announce Type: new Abstract: Undesired information such as harmful content and private data propagates through Multilingual Large Language Models (LLMs) via direct training and indirect cross-linguistic spread.

GPT-5.3 Instant: Smoother, more useful everyday conversations

Arkios: An Open Bilingual English-Nepali Language Model Trained From Scratch, with a Devanagari-Aware Tokenizer

arXiv:2608.30092v1 Announce Type: cross Abstract: We present Arkios, a 1.04B-parameter dense transformer pretrained from scratch on 150B tokens of bilingual English-Nepali text, using a custom single...

Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence

The paper investigates whether training small decoder-only transformers on code‑switched text can induce cross‑lingual alignment. Using two 100‑million‑word multilingual corpora—one a mix of English, Dutch, and Chinese BabyBabelLM data, and another generated by inserting word‑ and sentence‑level code‑switching via an LLM—the authors find that code‑switched training aligns representations of parallel text, especially across different scripts, and that this alignment persists when later training on monolingual documents. A curriculum that progresses from word‑level code‑switching to sentence‑level code‑switching and finally to monolingual data yields models that outperform baselines on the BabyLM evaluation suite, demonstrating that code‑switching curriculum learning is an effective data augmentation strategy for multilingual pretraining.