Judged Useless, Queried Anyway: Tool-Using Agents Rarely Turn Their Own...
An agent whose tool keeps returning nothing useful should stop relying on it. In a retrieval environment with controlled source failures, we separate how agents judge results...
Demos
- Judged Useless, Queried Anyway: Tool-Using Agents Rarely Turn Their Own... Claude Haiku 4.5 by bennidict23.
Models
- Claude Haiku 4.5 Anthropic.