# Where A Small Language Model Helps in Invoice Categorisation, Understood Through Embedding Geometry

Source: [arXiv](https://arxiv.org/abs/2608.18033v1)  
Feed7 permalink: https://feed7.dev/p/2608-18033v1-0gvjv1y  
Published: 2026-08-18T17:28:01.000Z  
Trust: Needs Review (needs_review)

## Why Included

A single-GPU SBERT beat the reported zero-shot LLM and vendor baseline for invoice coding, suggesting narrow, private classifiers can outperform broader models with modest local data.

## Source Summary

A fine-tuned SBERT reached **0.96 accuracy** on invoice classification, above the study’s zero-shot LLM and vendor baseline. For new-client generalization it reached **0.9 F1** with roughly **100 client-specific invoices**, using one GPU.

## Practical Implication

For narrow classification with sensitive data, benchmark a small in-house encoder before defaulting to a hosted general model. Inspect embedding clusters and test raw versus human-friendly structured inputs rather than assuming extra formatting helps.

## Agent-Ready Context

A fine-tuned SBERT reached **0.96 accuracy** on invoice classification, above the study’s zero-shot LLM and vendor baseline. For new-client generalization it reached **0.9 F1** with roughly **100 client-specific invoices**, using one GPU.

For narrow classification with sensitive data, benchmark a small in-house encoder before defaulting to a hosted general model. Inspect embedding clusters and test raw versus human-friendly structured inputs rather than assuming extra formatting helps.

The results concern one financial corpus and task, so they do not establish a general small-model advantage. Vendor identity strongly shaped the embedding space, and the material does not report deployment cost or performance on other accounting datasets.

## Context Map

- Layer: model
- Domains: data
- Topics: model-selection

## Uncertainty

- The results concern one financial corpus and task, so they do not establish a general small-model advantage. Vendor identity strongly shaped the embedding space, and the material does not report deployment cost or performance on other accounting datasets.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.