Twitxy Saved

Nandan Thakur

@beirmug

Postdoc @MSFTResearch | PhD @uwaterloo | RA @UKPLab l working on search and agents | (beir.ai, miracl.ai, @TREC_RAG and FreshStack)
Bengaluru, India thakur-nandan.github.io Joined July 2016
3.6K Following 3.5K Followers 1.9K Posts

Latest posts 5

Nandan Thakur @beirmug
With agents, one thing I'm seeing nowadays is reproduction happens effortlessly (ofcourse with supervision)!

but back in the days, manually understanding someone's code + making sure you are faithful to the setup + lots of pain and effort.
Alex Krentsel Verified account @AlexKrentsel Nov 2
i've reproduced the results from the MIT blog post today on weight-space programming. but one additional result i have to add: this is a new, step-change emerging capability that's appeared in just the latest set of models.
View quoted post on X
0 replies 1 reposts 15 likes 820 views View on X
Nandan Thakur @beirmug
Wow, woke up to see this amazing direction of work from @rikiyatakehi jayden, @a1zhang, @lateinteraction!
Rikiya Takehi @rikiyatakehi Nov 2
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.

We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weight-sp…
View quoted post on X
1 replies 0 reposts 12 likes 759 views View on X