# Small language models improve reasoning by sharing hints, preprint finds

_Published Tuesday, September 29, 2026 at 4:06 AM EDT · AI, Science · Latest · Tier 2 — Notable_

A research preprint reports that small language models working together solved more reasoning problems when they shared hints than when they independently generated answers for a vote. In the researchers' TalkMesh system, the most confident agent broadcasts a hint, less confident agents revise their answers, and each keeps a revision only if it scores higher.

In one test using Qwen3.5-0.8B on GSM8K, accuracy rose from 0.568 with independent sampling to 0.705 with the mesh. The authors also tested a defense against agents sending false hints: it retained 0.507 accuracy when four of eight agents colluded, compared with 0.000 for majority voting.

## Sources

- [cs.LG updates on arXiv.org](https://arxiv.org/abs/2609.30578)

---
Canonical: https://techandbusiness.org/newswire/_0PGYEYByG2tVDicxXsd56
Published: 2026-09-29T08:06:38.828Z
Story chronology: 2026-09-29T04:00:00.000Z
Retrieved: 2026-09-29T10:15:42.366Z
Publisher: Tech & Business (techandbusiness.org)