# Measuring Product Quality Using Images: The CLIP Q-Score and an Application to Real Estate

arXiv: https://arxiv.org/abs/2608.01544 · HTML: https://mlsanity.com/paper/2608.01544/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 19 references that are in the map.
- First posted on arXiv 2 Aug 2026.

## Builds on (most cited first, 4 of 4)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [What Image Features Boost Housing Market Predictions?](https://mlsanity.com/paper/2107.07148/) (2021)
- [Necessary but Not Sufficient: Temperature Control and Reproducibility in LLM-as-Judge Safety Evaluations](https://mlsanity.com/paper/2606.26185/) (2026)

## Built on by (most cited first, 1 of 1)

- [Representation Risk in Pretrained Image Encoders](https://mlsanity.com/paper/2609.35470/) (2026)

## Abstract (arXiv)

The CLIP Q-score is a novel, safe, fully reproducible, and computationally efficient method for extracting objective product quality metrics from visual data using contrastive language-image pre-training. We introduce the technique and provide an extensive application to real estate data from an online platform (~500,000 images). Our open-source metric aligns with LLM assessments and proves to be a powerful predictor of housing market prices for both sales and rentals. We also show that a higher CLIP Q-store is associated with better liquidity (reduced time on the market), especially for properties on sale.