News for hackers. Stuff that matters.
Re:DeepSeek R2 launch stalled as CEO balks at progress by jamesblonde on Saturday June 28, 2025 @05:19AM (#44403318)
Rumour was that DeepSeek used the outputs of the thinking steps in OpenAI's reasoning model (o1 at the time) to traing DeepSeek's Large Reasoning Model R1.
Re:DeepSeek R2 launch stalled as CEO balks at progress by orbital-decay on Saturday June 28, 2025 @06:06AM (#44403543)
More like a direct (and extremely dubious) accusation without proof from Altman. In reality those two models have as little in common as possible, and o1 reasoning chain wasn't available anyway.
Re:DeepSeek R2 launch stalled as CEO balks at progress by dachworker on Saturday June 28, 2025 @07:04AM (#44403796)
Maybe they also do that, but I work with a class of problems* that no other model has managed to crack, except for R1 and that is still the case today.

Remember that DeepSeek is the offshoot of a hedge fund that was already using machine learning extensively, so they probably have troves of high quality datasets and source code repos to throw at it. Plus, they might have higher quality data for the Chinese side of the internet.

* Of course I won't detail my class of problems else my benchmark would quickly stop being useful. I'll just say that it is a task at the undergraduate level of CS, that requires quite a bit of deductive reasoning.

Re:DeepSeek R2 launch stalled as CEO balks at progress by Art9681 on Sunday June 29, 2025 @02:04PM (#44415044)
This is completely irrelevant without knowing if you are effectively prompting each model. Your workflow may just be suitable for a particular model and not others. And tuning a workflow for each model is tedious. I seriously doubt there is ANY class of problem DSR1 can solve that OAI's third tier model can't at this point (o4-mini).
Re:DeepSeek R2 launch stalled as CEO balks at progress by WiSaGaN on Saturday June 28, 2025 @07:04AM (#44403801)
Deepseek published thinking trace before OpenAI did, not after.
Re:DeepSeek R2 launch stalled as CEO balks at progress by msgodel on Saturday June 28, 2025 @06:27AM (#44403641)
I don't think so. They came up with a new RL algorithm that's just better.
Re:DeepSeek R2 launch stalled as CEO balks at progress by Art9681 on Sunday June 29, 2025 @02:11PM (#44415096)
Better how? DeepSeek has never held the top spot in any aggregated benchmark. The Chinese bot armies are certainly better at convincing the internet they are trailing western models, despite the fact that in practical use this is not the case and at this rate, likely will never be. If AI progress is exponential, so is falling behind. What doesnt change however, is holding your ace card and dropping it when the time is right. China is competing with the public western models. Western SOTA labs are competing with their previous unreleased SOTA model. Publicly, China is ~3 months behind. But in reality, they are much further behind, and will never catch up.

Be mindful of what this means. A kid in his garage fine tuning a model can "catch up" to SOTA models for most use cases. For actual "frontier" work that requires SOTA levels of intelligence, there are only 3 companies in the race. None of them are from China or Europe.

Re:DeepSeek R2 launch stalled as CEO balks at progress by tw1984 on Saturday June 28, 2025 @08:07AM (#44404082)
OpenAI used literally all available text owned by the entire human race to train o1/o3.

so what?