Published November 16, 2026 | Version 1.0

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

Authors/Creators

  • 1. ROR icon University of Padua

Description

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. The record contains the note, the two pre-registered protocols, the code, the vectors u and v of all arms, and the per-problem results.

Files

just_sort_it.pdf

Files (660.3 kB)

Name Size
md5:0a61b8cd8be59d3bcff47d04d6e8425c
220.7 kB Preview Download
md5:db159c85e0579815ae11bfe97401221a
439.6 kB Preview Download

Additional details