paper-with-me

홈 › Papers

Meta-Personalizing Vision-Language Models to Find Named Instances in Video

2023-06-16 · CVPR 2023 1 · Chun-Hsiao Yeh, Bryan Russell, Josef Sivic, Fabian Caba Heilbron, Simon Jenni

Large-scale vision-language models (VLM) have shown impressive results for language-guided search applications. While these models allow category-level queries, they currently struggle with personalized searches for moments in a video where a specific object instance such as ``My dog Biscuit'' appears. We present the following three contributions to address this problem. First, we describe a method to meta-personalize a pre-trained VLM, i.e., learning how to learn to personalize a VLM at test time to search in video. Our method extends the VLM's token vocabulary by learning novel word embeddings specific to each instance. To capture only instance-specific features, we represent each instance embedding as a combination of shared and learned global category features. Second, we propose to learn such personalization without explicit human supervision. Our approach automatically identifies moments of named visual instances in video using transcripts and vision-language similarity in the VLM's embedding space. Finally, we introduce This-Is-My, a personal video instance retrieval benchmark. We evaluate our approach on This-Is-My and DeepFashion2 and show that we obtain a 15% relative improvement over the state of the art on the latter dataset.

📄 PDF Abstract BibTeX arXiv:2306.10169

Code (1)

danielchyeh/this-is-my 공식 구현 pytorch

Tasks

RetrievalWord Embeddings

Similar Papers 제목 키워드 기반

Small-Large Collaboration: Training-efficient Concept Personalization for Large VLM using a Meta Personalized Small VLM

2025-08-10 · Sihan Yang, Huitong Ji, Shaolin Lu, Jiayi Chen 외 arxiv

Personalizing Vision-Language Models (VLMs) to transform them into daily assistants has emerged as a trending research direction. However, leading companies like OpenAI continue to increase model size and develop complex…

Enhanced Meta-Learning for Cross-lingual Named Entity Recognition with Minimal Resources

2019-11-14 · Qianhui Wu, Zijia Lin, Guoxin Wang, Hui Chen 외

For languages with no annotated resources, transferring knowledge from rich-resource languages is an effective solution for named entity recognition (NER). While all existing methods directly transfer from source-learned…

Cross-Lingual NERCross-Lingual TransferMeta-Learningnamed-entity-recognition+4

JudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in Hebrew

2026-04-20 · Itay Razumenko, Arnon Sturm, Nir Grinberg arxiv

Despite significant advances in large language models, personalizing them for individual decision-makers remains an open problem. Here, we introduce a synthetic-organic supervision pipeline that transforms raw judicial d…

parameter-efficient fine-tuningSemantic Similarity

Evaluating Approaches to Personalizing Language Models

2020-05-01 · LREC 2020 5 · Milton King, Paul Cook

In this work, we consider the problem of personalizing language models, that is, building language models that are tailored to the writing style of an individual. Because training language models requires a large amount …

Language ModelingLanguage Modelling

Decomposed Meta-Learning for Few-Shot Named Entity Recognition

2022-04-12 · Findings (ACL) 2022 5 · Tingting Ma, Huiqiang Jiang, Qianhui Wu, Tiejun Zhao 외

Few-shot named entity recognition (NER) systems aim at recognizing novel-class named entities based on only a few labeled examples. In this paper, we present a decomposed meta-learning approach which addresses the proble…

Entity TypingFew-shot NERMeta-LearningNamed Entity Recognition+1