Method

Vision-language model

Multimodal LLMs for image understanding, VQA and document parsing.

23 cases from 17 companies apply Vision-language model. 3 were checked in an online A/B test; 9 report results in production.

Typical tasks: Match entities, Evaluate models, Build an assistant over internal docs, Label data with LLMs and 5 more. Companies: Yandex, DoorDash, Instacart, LY Corporation, Avito, Etsy and 11 more. Often combined with: OCR, Embedding similarity matching, LLM-as-a-judge, RAG, LLM classification / extraction and 59 more.

Building Food Metadata with LLM Juries, Context Optimization & Multimodal AIfood-delivery2026-07
AI Agents for the Frontline: 7-Eleven's GenAI Maintenance Assistantretail2026-06
Shaping product understanding with contrastive reinforcement learningmarketplace2026-05
Applying OCR to Automating Accounting Documentstelecom2026-05
Building RAG Systems That Work Accurately: From Prototype to Thousands of Users a Dayretail2026-05
From print to digital: Making weekly flyers shoppable at Instacart through computer vision and LLMsretail2026-02
PinLanding: Turn Billions of Products into Instant Shopping Collections with Multimodal AIe-commerce2026-01
Building Multimodal Product Embeddings to Improve Matching at Yandex Marketmarketplace2025-11
Scaling Catalog Attribute Extraction with Multi-modal LLMse-commerce2025-08
Advancing Menu Content with AI: How DoorDash uses AI to generate menu descriptionsfood-delivery2025-06
Choosing an automatic quality metric for inpainting models by correlation with human ratingssocial2025-06
Evaluating AI-generated images (The basics)general2025-06
Description Generation One Year Later. What We Managed to Understand Over the Yearclassifieds2025-05
Product Applications of Yandex VLMsearch-engine2025-05
YandexART: Humans and Neural Networks in Evaluating Generative Modelssearch-engine2025-05
High-Speed Multimodal Moderation: LLMs and VLMs in Real Productionmarketplace2025-05
Labeling Product Cards with LLMse-commerce2025-05
A Case Study in Recognizing and Structuring Data from Medical Formshealthtech2025-05
Snap Ads Understanding: From Pixels to Wordsadtech2025-04
Multimodality and RAG, or How to Sit on Two Chairsgeneral2024-11
Extracting Product Attributes from PDFs using PAE Frameworkretail2024-06
Auto-Generating Product Descriptions in Fashionclassifieds2024-05
Marketplace Product Matching in Detail: Models, Approaches and Tips for Building an Effective Matcher for Tens of Millions of Productsmarketplace2024-05