# Personalizing Retrieval using Joint Embeddings or "the Return of Fluffy"

arXiv: https://arxiv.org/abs/2510.05411 · HTML: https://mlsanity.com/paper/2510.05411/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 31 references that are in the map.
- First posted on arXiv 6 Oct 2025.

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [Robust fine-tuning of zero-shot models](https://mlsanity.com/paper/2109.01903/) (2021)
- [DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning](https://mlsanity.com/paper/2204.04799/) (2022)
- [ClipCap: CLIP Prefix for Image Captioning](https://mlsanity.com/paper/2111.09734/) (2021)
- [Zero-Shot Video Question Answering via Frozen Bidirectional Language Models](https://mlsanity.com/paper/2206.08155/) (2022)
- [What does CLIP know about a red circle? Visual prompt engineering for VLMs](https://mlsanity.com/paper/2304.06712/) (2023)
- [DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images](https://mlsanity.com/paper/1901.07973/) (2019)
- [X-Pool: Cross-Modal Language-Video Attention for Text-Video Retrieval](https://mlsanity.com/paper/2203.15086/) (2022)
- [Vision-by-Language for Training-Free Compositional Image Retrieval](https://mlsanity.com/paper/2310.09291/) (2023)
- ["This is my unicorn, Fluffy": Personalizing frozen vision-language representations](https://mlsanity.com/paper/2204.01694/) (2022)
- [Text-Only Training for Image Captioning using Noise-Injected CLIP](https://mlsanity.com/paper/2211.00575/) (2022)
- [CompoDiff: Versatile Composed Image Retrieval With Latent Diffusion](https://mlsanity.com/paper/2303.11916/) (2023)
- [MDMMT: Multidomain Multimodal Transformer for Video Retrieval](https://mlsanity.com/paper/2103.10699/) (2021)
- [CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor](https://mlsanity.com/paper/2312.07661/) (2023)
- [Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models](https://mlsanity.com/paper/2404.12387/) (2024)
- [Test-Time Adaptation with CLIP Reward for Zero-Shot Generalization in Vision-Language Models](https://mlsanity.com/paper/2305.18010/) (2023)
- [Composing Ensembles of Pre-trained Models via Iterative Consensus](https://mlsanity.com/paper/2210.11522/) (2022)
- [Meta-Personalizing Vision-Language Models to Find Named Instances in Video](https://mlsanity.com/paper/2306.10169/) (2023)
- [Evaluation of Test-Time Adaptation Under Computational Time Constraints](https://mlsanity.com/paper/2304.04795/) (2023)

## Abstract (arXiv)

The goal of this paper is to be able to retrieve images using a compound query that combines object instance information from an image, with a natural text description of what that object is doing or where it is. For example, to retrieve an image of "Fluffy the unicorn (specified by an image) on someone's head". To achieve this we design a mapping network that can "translate" from a local image embedding (of the object instance) to a text token, such that the combination of the token and a natural language query is suitable for CLIP style text encoding, and image retrieval. Generating a text token in this manner involves a simple training procedure, that only needs to be performed once for each object instance. We show that our approach of using a trainable mapping network, termed pi-map, together with frozen CLIP text and image encoders, improves the state of the art on two benchmarks designed to assess personalized retrieval.