Pinterest Uses VLM for Search Relevance Eval, Cutting Labeling Turnaround by 20x
_reachsumit · x · 2026-08-04
Pinterest deploys a VLM-based relevance evaluation pipeline for Search, fine-tuning Qwen3-VL on human-annotated query-Pin pairs, reducing labeling turnaround by over 20x. The method validates alignment with human annotations, improves evaluation efficiency, and enables expanding query sets and optimizing sampling, significantly reducing MDEs in online experiments. Paper at RecSys'26 Industry track.
More from Apps
- Hands-on with uncensored AI image generators: FLUX local, Krea vs Wan for speed and fidelity — Fun_Walk_4965 · 2026-08-04
- Google AI Overviews May Cut Search Ad CTR by Up to 40% — emmanuelvivier · 2026-08-04
- Google AI Overviews may cut search ad click-through rates by 20-40% — emmanuelvivier · 2026-08-04
- OpenAI launches ChatGPT Work agent for hours-long complex projects — emmanuelvivier · 2026-08-04
- Trick Grok into overnight mode to deliver results by morning — yunta_tsai · 2026-08-04
- Google Launches Managed Agents in Gemini API with MCP Support — emmanuelvivier · 2026-08-04