# Tree-Ising preprint reports cheaper weight repairs that improve math accuracy

_Published Wednesday, November 11, 2026 at 1:33 AM EST · AI, Science · Latest · Tier 2 — Notable_

Researchers report in a preprint that Tree-Ising repairs language model weights with four forward-pass requests and zero LLM tokens. On Qwen3.5-0.8B, the method scored 54.6% on MATH-500, compared with 51.7% for an earlier agent-generated repair under the same evaluation protocol.

Tree-Ising turns selected changes to stored weights into separate optimization problems solved on a GPU. Its simulated annealing recipe cost about $0.90 in GPU time and changed 13% to 17% fewer weights than the agent repair. On Qwen3.5-2B, the same recipe increased greedy GSM8K accuracy from 21.6% to 69.4%. The demonstrated repairs concern selected layers in these two models.

## Sources

- [cs.LG updates on arXiv.org](https://arxiv.org/abs/2611.14210)

---
Canonical: https://techandbusiness.org/newswire/Xv5xAWLkZ4RTFVp3zoYAQ2
Published: 2026-11-11T06:33:40.158Z
Story chronology: 2026-11-11T05:00:00.000Z
Retrieved: 2026-11-11T08:46:40.999Z
Publisher: Tech & Business (techandbusiness.org)