AI BriefingQ2D-Web Releases Large-Scale Web Search Retriever Benchmark Based on 190 Million Documents
Key point
Q2D-Web has released a large-scale web search retriever evaluation benchmark featuring 190 million documents and 10 languages.
Details
Q2D-Web is a large-scale benchmark and leaderboard for evaluating web search retriever models, containing 190 million documents and 69,721 queries across 10 languages.
This benchmark uses three independent sets of relevance judgments to minimize bias and enhance reliability, helping retrievers distinguish between relevant data and distractors.
Evaluation submissions are processed via subsampling to reduce resource costs while maintaining the integrity of model rankings.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Step 5 Preview Released: 600B Parameter MoE Model Highlights Agentic Tasks and Financial Analysis PerformancePrivatemode Benchmarks GLM-5.3-Flash as Vision-Capable Decision Model with Parity to Jev but Higher Cost