首页 > AI前沿 > WeightBench: Evaluating Language Models as Weight-Space Programmers

WeightBench: Evaluating Language Models as Weight-Space Programmers

arXiv机器学习 2026-11-14 02:58 11 阅读 查看原文

Recent work shows that large language models (LLMs) can operate directly on the parameters of other neural networks. They can read what a network was trained to do, fix a network that fails, and write working networks from scratch, all without gradient descent. This capability has been termed Weight-Space Programming (WSP).

WEIGHTBENCH

Exploring this new paradigm systematically requires a robust benchmark. We introduce WEIGHTBENCH, a benchmark that evaluates LLMs of increasing capability on 42 WSP tasks across five tracks: reading, fixing and writing weights, recognizing edits that are impossible, and data-free compression. The networks whose weights the LLMs read and edit range from 2K-parameter control policies to 7B-parameter language models.

Experimental Benchmark Results

Experimental benchmark results show that frontier agents can already read small networks and write working control policies without ever running them,