Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
• 24
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
WebHarbor: Docking Real Websites for Evolving GUI Agent Environments
ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost
Articles
microsoft/SynthDocQA
Preview • Updated • 1.02k • 6
microsoft/SchGen_dataset
Viewer • Updated • 8.42k • 720 • 10
microsoft/CABRA
Updated • 1.76k • 9
microsoft/sigmacollab
Preview • Updated • 1.08k • 6
microsoft/rushes
Viewer • Updated • 57.2k • 811 • 4
microsoft/ProgramDistill
Updated • 4.37k • 14
microsoft/roboeval-data-for-rho
Viewer • Updated • 2.5k • 646 • 1
microsoft/libero-data-for-rho
Franka Panda • Updated • 1.72k episodes • 8.77k • 4
microsoft/BusyBox
Viewer • Updated • 11.3k • 1.51k • 4
microsoft/WorkflowPerturb
Viewer • Updated • 44.9k • 166 • 3
microsoft/XL-DocBench
Viewer • Updated • 2.98k • 860 • 9
microsoft/RHELM
Viewer • Updated • 1.31k • 4.41k • 16
microsoft/OfficeComprehensionBenchmark
Viewer • Updated • 1.32k • 1.92k • 8
microsoft/Dayhoff-MMseqs2
Updated • 115 • 3
microsoft/SciFormaData-700K
Viewer • Updated • 1.38M • 2.49k • 1
microsoft/benchpress-score-matrix
Viewer • Updated • 995 • 394 • 2
microsoft/SciFormaBench
Viewer • Updated • 2k • 212 • 2
microsoft/MuseVLA-dataset
Viewer • Updated • 1.74k • 1.18k • 2
microsoft/Echoverse
Viewer • Updated • 722 • 53 • 9
microsoft/Orchard
Viewer • Updated • 110k • 833 • 24
microsoft/RESOURCE2SKILL
Updated • 1.62k • 18
microsoft/NOTSOFAR
Viewer • Updated • 50.6k • 39.3k • 27
microsoft/AVGen-Bench
Viewer • Updated • 3.01k • 5.3k • 7
microsoft/GridSFM_US_power_grid
Updated • 1.03k • 10
microsoft/WebTailBench
Preview • Updated • 645 • 20
microsoft/delulu-fim-benchmark
Viewer • Updated • 1.95k • 335 • 3
microsoft/PatientSafetyBench
Viewer • Updated • 466 • 146 • 10
microsoft/synthetic-computers-at-scale
Viewer • Updated • 98 • 213 • 23
microsoft/World-R1
Viewer • Updated • 6.48k • 151 • 10
microsoft/msr-acc-tae25
Viewer • Updated • 72.6k • 165 • 10