SyncAI.news, a Varaisys broadcasting
Just sort it! A note on the LP behind rank-one anti-loop edits
MF

Matteo Fischetti

· 1 min read

ResearcharXiv cs.CL

Just sort it! A note on the LP behind rank-one anti-loop edits

arXiv:2611.21432v1 Announce Type: cross Abstract: Weight-space programming removes the thinking loops of a small reasoning model by a rank-one edit W' = W + u v^T of an MLP output matrix, where u is a loop direction and v is the solution of a sparse Linear Program (LP) over the neuron activations. We investigate whether the LP is needed. We replace it by a sort: we take the 8 neurons whose activation has the smallest coefficient of variation, and give them equal shares of the push that the LP solution realizes. On Qwen3.5-0.8B the sorted edit scores 67.0% on 1000 problems (MATH-500 and GSM8K) against 67.4% for the LP edit and 9.5% for the unedited model; the paired difference is -0.4 points, with 95% confidence interval [-2.9, +2.1], inside a non-inferiority margin of 5 points declared before the runs. On three further sampling seeds the difference is -0.6 points, interval [-2.0, +0.8], and -0.8 points, interval [-2.3, +0.6], when the dose is set without the LP. The sort does not solve the LP, but the neurons it shares with the LP solution carry most of its mean push. The outcome is that, on this model, the edit acts as a nearly constant bias along u.

Original source

This story was published by arXiv cs.CL and written by Matteo Fischetti. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News