Search-to-World: Evaluation of 3D World Delivery from User Request through Web Search

2026-09-07

AI Summary

本文解决了从用户请求到3D世界生成的问题，通过引入Search-to-World任务及WorldSearcher工具，系统评估了从网络搜索内容转换成3D世界的全过程。

Abstract

Agentic systems can interpret user requests, search the live web, and use external tools, but their ability to transform retrieved web content into a usable 3D world has not been systematically evaluated. No established end-to-end pipeline or benchmark exists for this capability. We introduce Search-to-World, an end-to-end evaluation task covering request understanding, web visual-content retrieval, and 3D-world delivery. We define Observed Retrieval Rate (ORR) and World Delivery Rate (WDR) to distinguish observing relevant content from successfully delivering a request-aligned, perceptually acceptable world. We also present WorldSearcher, a reuse-then-reconstruction harness that connects existing search agents to world delivery: it first retrieves reusable 3D worlds and, when none are available, reconstructs a world from video. A structured recovery controller revises temporal grounding, replaces source videos, or reformulates queries after failure. Using WorldSearcher, we benchmark representative models on Search-to-World and study supervised fine-tuning (SFT) for recovery subagents. Results show that delivery depends on the underlying agentic model, and that relevant-content observation does not ensure world delivery. Jointly training recovery agents improves delivery success and action efficiency. Search-to-World makes agentic 3D-world delivery measurable, while WorldSearcher provides a practical evaluation harness with recovery capabilities.

Problem

• 3D world delivery
• web search
• user request
• end-to-end evaluation
• agentic systems

Innovation

• Search-to-World
• WorldSearcher
• recovery controller
• Observed Retrieval Rate (ORR)
• World Delivery Rate (WDR)