# Few-Shot Classification & Segmentation Using Large Language Models Agent

Source: [Few-Shot Classification & Segmentation Using Large Language Models Agent](<https://www.sota2.com/research/paper/68fa713a1e209131cea4fea7>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:00:40.638Z

Authors: Tian Meng, Yang Tao, Wuliang Yin

Published: 2023-11-19T00:33:41.000Z

## Abstract

The task of few-shot image classification and segmentation (FS-CS) requires the classification and segmentation of target objects in a query image, given only a few examples of the target classes. We introduce a method that utilises large language models (LLM) as an agent to address the FS-CS problem in a training-free manner. By making the LLM the task planner and off-the-shelf vision models the tools, the proposed method is capable of classifying and segmenting target objects using only image-level labels. Specifically, chain-of-thought prompting and in-context learning guide the LLM to observe support images like human; vision models such as Segment Anything Model (SAM) and GPT-4Vision assist LLM understand spatial and semantic information at the same time. Ultimately, the LLM uses its summarizing and reasoning capabilities to classify and segment the query image. The proposed method's modular framework makes it easily extendable. Our approach achieves state-of-the-art performance on the Pascal-5i dataset.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2311.12065>)

[PDF](<https://arxiv.org/pdf/2311.12065.pdf>)
