---
title: "PMCE: Probabilistic Multi-Granularity Semantics with Caption-Guided Enhancement for Few-Shot Learning"
authors:
  - "Jiaying Wu"
  - "Can Gao"
  - "Jinglu Hu"
  - "Hui Li"
  - "Xiaofeng Cao"
  - "Jingcai Guo"
doi: "10.24963/ijcai.2026/206"
url: "https://ijcai.org/proceedings/2026/206"
pdf: "https://ijcai.org/proceedings/2026/0206.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "1848-1856"
track: "Main Track"
keywords:
  - "Computer Vision: Transfer, low-shot, semi- and un- supervised learning"
  - "Machine Learning: Learning sparse models"
  - "Machine Learning: Open-World/Open-Set/OOD Learning"
  - "Machine Learning: Foundation models"
  - "Machine Learning: Cost-sensitive learning"
---

# PMCE: Probabilistic Multi-Granularity Semantics with Caption-Guided Enhancement for Few-Shot Learning

Jiaying Wu, Can Gao, Jinglu Hu, Hui Li, Xiaofeng Cao, Jingcai Guo

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 1848-1856. DOI: [10.24963/ijcai.2026/206](https://doi.org/10.24963/ijcai.2026/206)

## Abstract

Few-shot learning aims to recognize novel categories from limited labeled samples, where prototypes estimated from 1--5 supports per class are often unreliable. Semantic-based approaches alleviate this by introducing class-level priors, but they often ignore instance-level cues and rarely optimize queries under the inductive protocol. We propose PMCE, a Probabilistic framework that leverages Multi-granularity semantics with Caption-guided Enhancement for few-shot classification. On base classes, we build a knowledge bank with class-wise visual statistics and class-name embeddings. At test time, the semantic embedding of a novel class retrieves a few similar base classes whose visual priors are aggregated into a class-specific prior and combined with the support-based prototype via MAP estimation. Simultaneously, we train a lightweight enhancer on base classes that fuses frozen BLIP captions with visual features, and apply it to both supports and queries without using query-set statistics or novel labels. A simple caption-consistency regularizer further improves robustness to noisy captions. Experiments on four standard benchmarks with ResNet-12 and Swin-T backbones show that PMCE outperforms state-of-the-art few-shot baselines, achieving up to 7.71% gains over the strongest competitor on MiniImageNet in the challenging 1-shot setting. Our code is available at https://github.com/channa419/PMCE.