Multi-Hop Reasoning

Overview

Multi-hop reasoning goes beyond one-step answers to trace multi-layered fact relationships. It uncovers indirect links and reveals new opportunities that traditional AI often misses.

Why it Matters

Real-world questions often require multiple steps of reasoning.

Example:

"What's the total market cap of companies where Company A's CEO previously worked?" → Identify Company A's CEO → Find previous employment → Look up current market cap of each company → Calculate sum

How it Works

1. Analyze the question to identify required reasoning steps.
2. Search for relevant FactBlocks at each step.
3. Track relationships between steps.
4. Derive final conclusions and present the reasoning path.

Benefits

• Handle Complex Questions — Answer questions that can't be resolved with a single search
• Discover Hidden Connections — Identify relationships not explicitly stated
• Reasoning Transparency — Clearly show the path to conclusions