xbench-DeepSearch
Emerging6papers using it
2025first seen
'XBench-DeepSearch' is a benchmark dataset that contains complex question-answer pairs generated from authentic web sources, used to evaluate the deep reasoning capabilities of multi-turn agents in long-horizon interactions.
Papers using xbench-DeepSearch (6)
- Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel ExecutionSlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward GatingBeyond Turn Limits: Training Deep Search Agents with Dynamic Context WindowTongyi DeepResearch Technical ReportMiroFlow: Towards High-Performance and Robust Open-Source Agent Framework for General Deep Research TasksWebAnchor: Anchoring Agent Planning to Stabilize Long-Horizon Web Reasoning