# What Matters for Grocery Product Retrieval with Open Source Vision Language Models

arXiv: https://arxiv.org/abs/2605.18029 · HTML: https://mlsanity.com/paper/2605.18029/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Document understanding and multimodal RAG, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 33 of its 39 references that are in the map.
- First posted on arXiv 18 May 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Document understanding and multimodal RAG](https://mlsanity.com/mllm/document-understanding-and-multimodal-rag/)

## Builds on (most cited first, 30 of 33)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT](https://mlsanity.com/paper/2004.12832/) (2020)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [PaLI: A Jointly-Scaled Multilingual Language-Image Model](https://mlsanity.com/paper/2209.06794/) (2022)
- [Passage Re-ranking with BERT](https://mlsanity.com/paper/1901.04085/) (2019)
- [DataComp: In search of the next generation of multimodal datasets](https://mlsanity.com/paper/2304.14108/) (2023)
- [Perception Encoder: The best visual embeddings are not at the output of the network](https://mlsanity.com/paper/2504.13181/) (2025)
- [Demystifying CLIP Data](https://mlsanity.com/paper/2309.16671/) (2023)
- [SmolVLM: Redefining small and efficient multimodal models](https://mlsanity.com/paper/2504.05299/) (2025)
- [ELEVATER: A Benchmark and Toolkit for Evaluating Language-Augmented Visual Models](https://mlsanity.com/paper/2204.08790/) (2022)
- [The Efficiency Misnomer](https://mlsanity.com/paper/2110.12894/) (2021)
- [MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training](https://mlsanity.com/paper/2311.17049/) (2023)
- [An Inverse Scaling Law for CLIP Training](https://mlsanity.com/paper/2305.07017/) (2023)
- [Meta CLIP 2: A Worldwide Scaling Recipe](https://mlsanity.com/paper/2507.22062/) (2025)
- [Products-10K: A Large-scale Product Recognition Dataset](https://mlsanity.com/paper/2008.10545/) (2020)
- [ViTamin: Designing Scalable Vision Models in the Vision-Language Era](https://mlsanity.com/paper/2404.02132/) (2024)
- [RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification](https://mlsanity.com/paper/2006.12634/) (2020)
- [NLLB-CLIP -- train performant multilingual image retrieval model on a budget](https://mlsanity.com/paper/2309.01859/) (2023)
- [Domain invariant hierarchical embedding for grocery products recognition](https://mlsanity.com/paper/1902.00760/) (2019)

## Abstract (arXiv)

Multimodal product retrieval (MPR) underpins checkout-free retail and automated inventory systems, yet it demands fine-grained SKU discrimination that standard vision-language benchmarks fail to capture. We present the first systematic zero-shot evaluation of 190 open-source VLMs on the MPR task of the GroceryVision Challenge, isolating pre-training data, architecture, and input resolution. Our analysis yields three actionable findings. (1) Data quality trumps scale. Switching from raw web-scrapes to filtered datasets delivers up to 16.6% accuracy gains, exceeding the benefit of doubling model parameters. (2) Efficient models can win. MobileCLIP-B (150M parameters) outperforms 351M counterparts trained on noisy data. We introduce textitsemantic power density (φ), an efficiency metric that penalizes sub-threshold accuracy. (3) A precision gap persists. State-of-the-art models achieve 94.5% Recall@5 but suffer a 17.5% drop at Recall@1, revealing that contrastive embeddings cluster categories effectively but fail to rank visually similar SKUs. Code and evaluation scripts are available at urlhttps://github.com/upeee/openmpr.