QUEST trains open deep research agents on 8K synthetic tasks from unified rubric trees, approaching or surpassing closed-source agents on eight benchmarks while releasing all artifacts.
- If search snippets reveal candidate authoritative URLs, you may reference them inside uncertain.need, but do NOT add them to visited_sources unless visited
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks
QUEST trains open deep research agents on 8K synthetic tasks from unified rubric trees, approaching or surpassing closed-source agents on eight benchmarks while releasing all artifacts.