DeepStress: Stress-Testing Deep Search Agents
WAB @ COLM 2026 · 2026
I build and study systems that search, read and reason over large collections of text. What interests me is the path an agent takes, not only where it lands. I work on ways to measure that path, and on training models that stay reliable when the evidence they find is not.
Keywords: Search Agents, Deep Research, Dynamic Environments
layer 01
WAB @ COLM 2026 · 2026
EvalLLM 2025 · 2025
INLG 2023 — Generation Challenges · 2023
LaTeCH-CLfL 2023 @ EACL · 2023
layer 03