Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification
IntermediateYuxuan Wan, Tianqing Fang et al.Jan 22arXiv
DeepVerifier is a plug-in checker that helps Deep Research Agents catch and fix their own mistakes while they are working, without retraining.
#Deep Research Agents#verification asymmetry#rubrics-based feedback