mlsanityData to 14 Oct 2026

What Makes Good Few-shot Examples for Vision-Language Models?

arXiv 2405.13532 ↗First posted 22 May 2024cs.CVcited by 1

On the map: Multimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP

Abstract (arXiv)

Despite the notable advancements achieved by leveraging pre-trained vision-language (VL) models through few-shot tuning for downstream tasks, our detailed empirical study highlights a significant dependence of few-shot learning outcomes on the careful selection of training examples - a facet that has been previously overlooked in research. In this study, we delve into devising more effective strategies for the meticulous selection of few-shot training examples, as opposed to relying on random sampling, to enhance the potential of existing few-shot prompt learning methodologies. To achieve this, we assess the effectiveness of various Active Learning (AL) techniques for instance selection, such as Entropy and Margin of Confidence, within the context of few-shot training. Furthermore, we introduce two innovative selection methods - Representativeness (REPRE) and Gaussian Monte Carlo (Montecarlo) - designed to proactively pinpoint informative examples for labeling in relation to pre-trained VL models. Our findings demonstrate that both REPRE and Montecarlo significantly surpass both random selection and AL-based strategies in few-shot training scenarios. The research also underscores that these instance selection methods are model-agnostic, offering a versatile enhancement to a wide array of few-shot training methodologies.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 13 of its 26 references that are in the map. First posted on arXiv 22 May 2024.

Builds on

13 of 26 references.

Closest ancestors · 1

Same direction, newest first

  1. 2021Learning to Prompt for Vision-Language Models2 Sep

Other references · 12

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2013Describing Textures in the Wildcited by 2,125
  3. 2013Fine-Grained Visual Classification of Aircraftcited by 1,726
  4. 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,496
  5. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  6. 2020A Survey of Deep Active Learningcited by 554
  7. 2021Oriented R-CNN for Object Detectioncited by 231
  8. 2020General Multi-label Image Classification with Transformerscited by 127
  9. 2021On Training Instance Selection for Few-Shot Neural Text Generationcited by 26
  10. 2020Learning to Select Base Classes for Few-shot Classificationcited by 11
  11. 2023Deeply Coupled Cross-Modal Prompt Learningcited by 11
  12. 2023Enhancing Vision-Language Pre-Training with Jointly Learned Questioner and Dense Captionercited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Learning to Select Visual In-Context Demonstrationscited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2023Active Prompt Learning in Vision Language Models
  2. 2024Low-Rank Few-Shot Adaptation of Vision-Language Models
  3. 2025Vocabulary-free few-shot learning for Vision-Language Models
  4. 2024Making Large Vision Language Models to be Good Few-shot Learners
  5. 2025Rethinking Few-Shot Adaptation of Vision-Language Models in Two Stages
  6. 2024Active Learning for Vision-Language Models
  7. 2022Rethinking Task Sampling for Few-shot Vision-Language Transfer Learning
  8. 2024Few-Shot Recognition via Stage-Wise Retrieval-Augmented Finetuning