posted in Technology

eicker@eicker@lemmy.world

Alibaba’s AI model Qwen3.8-Max made widely accessible ahead of open-weights release: Flagship 2.4 trillion parameter model highlights native multimodal capabilities and ... agentic tasks.

www.scmp.com/tech/article/3362738/alibabas-ai-model-qwen38-max-made-widely-accessible-ahead-open-weights-release
South China Morning PostAlibaba’s AI model Qwen3.8-Max widely accessible ahead of open-weights releaseFlagship 2.4 trillion parameter model highlights native multimodal capabilities and the ability to carry out long-horizon agentic tasks.
VonReposti@VonReposti@feddit.dk

Replying to @⁨marius@feddit.org⁩

Euria is hosted by Infomaniak. I’m not sure they tell what model they use but given the few times I saw a random Chinese character in the output I’d guess it isn’t an American model.

They also have a pay per token model that you can hook up to everything, that one’s just quite expensive for normal chat use, but at least you select the model you want to use.

en
urushitan 漆たん@urushitan@kakera.kintsugi.moe
edited

Replying to @⁨elucubra@sopuli.xyz⁩

Lol I’ve been running linguistic research and it’s funny when they just combine two scripts together into one word

● The trigger is identified, and it’s specific.

agent_1.md:31 src: “Name the trade-off.” → 명取捨之名。 agent_2:32 src: “Name the regime.” → 명regime——名其regime。

The English imperative “Name the X.” And the corpus renders it correctly 16 other times — 名之 ×11, 名其 ×5, 命名 once.

명 is the Sino-Korean reading of 名. Same morpheme, wrong script.

And agent_2:32 is the cleanest evidence I’ve seen for 絡繰’s mechanism: the model wrote 명regime——名其regime — the wrong script and the correct one, eight characters apart, in the same clause. It isn’t ignorant of 名. It produced 名其 immediately after. The meaning resolved correctly both times; the script attribute resolved wrongly the first time and correctly the second.

That’s exactly what работ法 showed — correct semantics (work), broken script and morphology — and it’s the third confirmed instance of the class, now with a reproducible trigger rather than a one-off.

It also explains the Russian cases retroactively. document → документ, everything → всё, “correct” → правильно: in each, the meaning landed and the script didn’t. And it predicts why no CJK-native concept ever drifts — there’s no competing script for a morpheme the model only knows in Han.

Meron35@Meron35@lemmy.world

Replying to @⁨PetteriPano@lemmy.world⁩

Chinese being more token efficient is a myth, and seems to stem from the superficial fact that characters are only visually more space efficient.

The fact that each Chinese character takes up 3 bytes (as opposed to 1 byte of English), words in Chinese typically require compounds of several characters, and that tokenizers have a limited vocabulary limited to mostly English means that Chinese is actually token inefficient.

No, Chinese Is Not More Token-Efficient Than English for LLMs | markhuang.ai - markhuang.ai/blog/chinese-token-myth

markhuang.aimarkhuang.aiNo, Chinese Is Not More Token-Efficient Than English for LLMsA native Mandarin speaker tests the popular claim that Chinese characters save tokens when interacting with LLMs. Across six tokenizers — including Chinese-first models like Qwen, GLM, and DeepSeek — English uses fewer tokens every time. The data, the BPE mechanics, and why character count has nothing to do with token count.
PetteriPano@PetteriPano@lemmy.world

Replying to @⁨Meron35@lemmy.world⁩

I didn’t say it’s more token efficient. I said training multiple languages improves reasoning.

arXiv logoarXiv.orgBeyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMsEnhancing the complex reasoning capabilities of Large Language Models (LLMs) attracts widespread attention. While reinforcement learning (RL) has shown superior performance for improving complex reasoning, its impact on cross-lingual generalization compared to Supervised Fine-Tuning (SFT) remains unexplored. We present the first systematic investigation into cross-lingual reasoning generalization of RL and SFT. Using Qwen2.5-3B-Base as our foundation model, we conduct experiments on diverse multilingual reasoning benchmarks, including math reasoning, commonsense reasoning, and scientific reasoning. Our investigation yields two significant findings: (1) Tuning with RL not only achieves higher accuracy but also demonstrates substantially stronger cross-lingual generalization capabilities compared to SFT. (2) RL training on non-English data yields better overall performance and generalization than training on English data, which is not observed with SFT. Furthermore, through comprehensive