Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Spatial Question Answering on OST-Bench 1,396 QA pairs

65.04Average Score

TRACE

58.860.4262.0463.66Mar 24, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2026.03
65.0491.8557.5261.2210087.369.1524.5632.3574.8742.4226.4410029.0794.5238.06
2026.03
6488.5954.957.1410088.8969.1521.0533.9278.8936.3627.5310011.8597.2638.89
2026.03
63.7588.0444.4463.2710088.8977.1121.0535.8876.8840.9122.331007.7810036.94
2026.03
62.6592.3951.6353.0610085.7163.6821.0534.1276.3840.9128.771009.2691.7840.28
2026.03
62.291.352.2951.0210090.4864.6815.7929.2275.3840.9127.5310015.7484.9341.39
2026.03
61.6989.6748.3748.9810082.5468.6615.7933.3375.3839.3928.7797.1411.389.0439.31
2026.03
60.4295.0586.5866.6796.4377.785752.7334.1260.356.4531.2554.2931.9276.7129.03
2026.03
59.2795.6585.5266.6710076.195350.9125.8855.7853.0331.2371.4335.8569.8618.06
2026.03
59.2682.2493.9965.9696.5577.7852.2462.9627.8452.765031.6471.4324.2672.626.67
2026.03
59.2296.7284.8768.7589.6682.5454.2748.1528.6360.6154.553071.4322.7872.622.68
2026.03
59.294.5483.5566.6793.180.6554.2754.393155.6153.8532.3661.7629.0772.624.58
2026.03
59.0495.0586.1868.7589.6677.7855.7248.1525.460.349.2326.38024.8169.8628.47