Where Do the Tokens Go? Understanding and Reducing Costs in LLM Agents for Vulnerability Discovery
The study analyzed how different AI agents vary in success and cost when finding vulnerabilities using code. It found that token consumption is dominated by understanding code and reasoning, while certain existing efficiency methods fail to consistently reduce cost without sacrificing success. The research introduces AVRI, an interface using a Bidirectional Evidence Trace, which successfully reduced costs for both agents while maintaining or improving effectiveness.