EconBase
← All papers

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

Fabian Slonimczyk, Danila Karapsin

arXiv 2 Aug 2026 · Econometrics

arXiv:2608.01544 · PDF · Extracted main text

Abstract

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 ($\sim500,000$ 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.

Citation extraction

19
references
23
in-text mentions
19
distinct cited
0
self-citations
7,645
main-text words

appendix boundary found by appendix_command · 98% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Radford, Alec and Kim, Jong Wook and Hallacy, Chris and Ramesh, Adit… (2021) Learning Transferable Visual Models From Natural Language Supervision0.64441100%
2Hessel, Jack and Holtzman, Ari and Forbes, Maxwell and Bras, Ronan L… (2021) CLIPScore: A Reference-free Evaluation Metric for Image Captioning0.51121100%
3Ash, Elliott and Durante, Ruben and Grebenschikova, Maria and Schwar… (2021) Visual Representation and Stereotypes in News Media0.40511100%
4Atil, Berk and Aykent, Sarp and Chittams, Alexa and Fu, Lisheng and… (2025) Non-Determinism of “Deterministic” LLM System Settings in Hosted Environments0.40511100%
5Deng, Lin (2025) Real Estate Valuation with Multi-Source Image Fusion and Enhanced Machine Learning Pipeline0.40511100%
6Dzyabura, Daria and El Kihal, Siham and Hauser, John R. and Ibragimo… (2023) Leveraging the Power of Images in Managing Product Return Rates0.40511100%
7Gorin, Clément and Heblich, Stephan and Zylberberg, Yanos (2025) State of the Art: Economic Development Through the Lens of Paintings0.40511100%
8Gunko, Maria and Bogacheva, Polina and Medvedev, Andrey and Kashnits… (2018) Path-Dependent Development of Mass Housing in Moscow, Russia0.40511100%
9Jean, Neal and Burke, Marshall and Xie, Michael and Davis, Matthew a… (2016) Combining Satellite Imagery and Machine Learning to Predict Poverty0.40511100%
10Kostic, Zona and Jevremovic, Aleksandar (2020) What Image Features Boost Housing Market Predictions?0.40511100%

Showing the top 10 of 19 scored citations.