ResearchBenchmarksTable Storage on Real-world Table Storage (seen prompt)Table Storage on Real-world Table Storage (seen prompt)Follow100Success Rate (Red Tea Bag)APT94.896.1597.598.85Jun 10, 2026Evaluation resultsMethodMethodLinksSuccess Rate (Red Tea Bag)Success Rate (Spoon)Success Rate (Box)Average Success RateAPT2026.06100908585π0.52026.0695958080