Hacker News Reader

Comment by constantlm | Hacker News Reader

So, unlike what leaders of western labs labs would like you to believe (that Kimi is just the result of distillation attacks), they are introducing new and novel approaches.

View on HN

Even if they are distilling, I don't particularly care. Anthropic and others have been distilling copyrighted material by to build these models, largely without permission.

Exactly,let Anthropic get back to training on the New York times.

No! It's just not fair!! What about our financial bubble!

Moreover, Distillation is a misused term here. Distillation means training a smaller student model using a larger teacher model to completely mimic its behavior. As in you take a base model and create its smaller “turbo” version. Like distillation in Chemistry it means it’s %100 purified version of its teacher.

"Kimi is largely a byproduct of distillation" and "Kimi is introducing new and novel approaches" are not mutually exclusive, and I'm not sure it's clear from the paper how much of the improvement comes from the new approaches. So I wouldn't take the new approaches to be much evidence about whether the distillation attacks occurred.

> distillation attacks

Like the distillation “attacks” Anthropic conducted against literally millions and millions of different creators?

Scanning google is an ‘attack’

Reading Claude's outputs is an 'attack'

Or against all those old books that they ingested and then literally destroyed

Even Anthropic hasn't claimed "Kimi is largely a byproduct of distillation"

There's been some distillation.

Just like how Elon said in court they distill to make Grok.

I thought there is discussion circulating around whether the main reason Kimi is impressive is due to distillation. Though possibly if this occurred, this was just one component of their training pipeline and not a majority

"attack"

Reply on news.ycombinator.com