List your company

Instruction Following on Tulu3 Evaluation Suite pool (test)

92.54ARCTAGS
58.698467.484276.2785.0558Jan 20, 2026

Evaluation results

MethodLinks
TAGSBase Model=Qwen3-4B, DataSize=50K2026.0192.5470.7889.6976.8375.3366.7378.6514.787.3440.9843.0560.85
CaRBase Model=Qwen3-4B, DataSize=50K2026.0191.8668.5179.377.4475.2654.974.5413.547.3338.841.8858.21
IFDBase Model=Qwen3-4B, DataSize=50K2026.0191.5372.478.1776.2275.2357.8675.2315.657.2640.1542.859.02
MIGBase Model=Qwen3-4B, DataSize=50K2026.0191.5372.170.4381.175.4765.876.0714.047.6440.6643.759.88
QDITBase Model=Qwen3-4B, DataSize=50K2026.0191.1972.985.0676.8375.1658.7876.6512.87.2438.8641.3559
IFDBase Model=Qwen3-8B, DataSize=50K2026.0190.8577.6591.3681.7178.7957.1279.5815.167.6242.3244.5662.07
PoolBase Model=Qwen3-8B, DataSize=939K2026.0188.4769.7188.8677.4475.2769.8778.279.697.3444.5242.5460.4
TAGSBase Model=Qwen3-8B, DataSize=50K2026.0186.7874.2291.5182.3278.7566.9180.0820.127.8544.7647.7963.93
#InsTagBase Model=Qwen3-8B, DataSize=50K2026.0185.4270.7890.7580.4978.8160.6377.8110.687.6742.1643.1860.5
RandomBase Model=Qwen3-4B, DataSize=50K2026.0185.4267.7589.2378.0575.2961.1876.159.077.2538.1639.9158.03
#InsTagBase Model=Qwen3-4B, DataSize=50K2026.0184.0766.8886.9678.6675.3563.5975.929.577.3838.2640.5458.23
TAGSBase Model=Qwen3-1.7B, DataSize=50K2026.0182.0354.3180.1467.6863.3148.4365.9810.936.7629.8336.1251.05
RandomBase Model=Qwen3-8B, DataSize=50K2026.0180.6872.2990.7582.9378.664.1478.239.447.5141.7442.0960.16
MIGBase Model=Qwen3-1.7B, DataSize=50K2026.0180.3453.5378.7766.466444.3664.589.696.5427.9234.3449.46
CaRBase Model=Qwen3-8B, DataSize=50K2026.0179.6673.491.3682.3278.9156.1976.9713.297.4942.5243.5760.27
RandomBase Model=Qwen3-1.7B, DataSize=50K2026.0178.6451.6877.2665.8564.3342.1463.324.846.6425.0232.0947.7
IFDBase Model=Qwen3-1.7B, DataSize=50K2026.0178.3153.9254.9767.0764.235.4958.996.216.625.4632.5645.77
#InsTagBase Model=Qwen3-1.7B, DataSize=50K2026.0178.3150.8877.2666.4664.2743.4463.445.096.5126.1732.1247.78
QDITBase Model=Qwen3-8B, DataSize=50K2026.0176.9577.490.6781.7178.8258.2377.314.047.4140.943.0160.16
DEITABase Model=Qwen3-4B, DataSize=50K2026.0176.9567.8381.3578.6675.5663.2273.938.27.2638.1439.6456.78
QDITBase Model=Qwen3-1.7B, DataSize=50K2026.0176.9554.6258.0767.0764.1337.7159.767.086.2624.4931.3945.58
MIGBase Model=Qwen3-8B, DataSize=50K2026.0176.6174.4991.5179.8879.0465.877.8916.657.5943.4945.3561.46
CaRBase Model=Qwen3-1.7B, DataSize=50K2026.0171.8652.773.7761.5964.3638.4560.456.466.4326.4232.446.42
DEITABase Model=Qwen3-1.7B, DataSize=50K2026.0171.5351.4262.5565.8564.2942.3359.665.226.3325.431.3145.48
DEITABase Model=Qwen3-8B, DataSize=50K2026.016071.5992.6585.3778.6160.4474.7814.047.6244.1644.859.79