Skip to content
Trending storyDeveloping

LLMs can read and edit neural network weights directly

1 article1 sourcesince Nov 2Last article 2h ago ·

Overview

AISummary of 1 article

Elvis Saravia reports that LLMs can read and write neural network weights, which he calls a more direct route to recursive self-improvement than current pipelines that write code to launch training jobs.

Rikiya Takehi says GPT-6 Astra changed only 0.016% of the weights of Qwen3.5-0.8B without gradients or training, and that the edits raised math performance up to 6x and instruction following 3x. These claims come from social media posts, and the report provides no independent evaluation or methods detail.

Written by AI from the articles below · updated Nov 2, 3:49 PM ET

Check the sources:

Latest developmentNov 2, 3:31 PM ET
LLMs can read and edit neural network weights directly

Article timeline

The articles in this story. Times are ET.

Nov 2
  1. elvisX
    LLMs can read and edit neural network weights directly

    AIA report finds LLMs can read and write neural network weights, a capability Elvis Saravia calls a more native route to recursive self-improvement than current pipelines that write code to launch training jobs. In a related post, Rikiya Takehi says GPT-6 Astra changed only 0.016% of Qwen3.5-0.8B's weights without gradients or training, improving math up to 6x and instruction following 3x.

Heat Trend

Not enough continuous observations to show a trend yet.