Re:Astra can read and write weights to improve Qwen
by amluto on Monday November 02, 2026 @03:58PM
(#50343646)
The getting-the-car-over-the-barrier one looks particularly simple. It’s hard to see from the animation, but the obvious solution is to back up then drive forward. I’m not remotely surprised that Astra can figure that out, and I’m also not particularly surprised that Astra can come up with weights that do this.
Re:Astra can read and write weights to improve Qwen
by itkovian_ on Monday November 02, 2026 @03:42PM
(#50343415)
Also no doubt they’ve coined a term with weight space programming and this will expand into an active sub field
Re:Astra can read and write weights to improve Qwen
by direwolf20 on Monday November 02, 2026 @08:51PM
(#50346968)
Weight Space Programming will be an amazing discovery when they learn that instead of telling an LLM what to do in English, you can create a much smaller LLM that is tuned to your specific task. Maybe it'll even turn out that the input and output can be directly connected to the problem task and don't need to be in English!
Re:Astra can read and write weights to improve Qwen
by root_axis on Tuesday November 03, 2026 @02:17AM
(#50349027)
> I’d say this is fortunate because if llms could do the latter, rsi would be immediate.
It doesn't follow that because a model could improve the capabilities of a model many orders of magnitude smaller, that it could then recursively improve it's own capabilities. There's also a conservation of information issue with the idea.
- the primary experiments are on small policy networks - they are compressed dynamic system models. The observation and action space can be read off. It’s not surprising to me that these types of networks can be used to identify the env they were trained on
- the exps on actual llms remove looping behaviour and that’s it - from article “but it could not add an ability the model lacked”
I’d say this is fortunate because if llms could do the latter, rsi would be immediate. It may be possible we simply need bigger and better edit models, but at least from the experiments there doesn’t seem to be any indication that is imminently about to work.
Conceptually incredibly interesting non the less.