List your company

Machine Translation on ZHUANGBENCH (test)

48.3BLEU (easy)GPT-4
-1.82811.18624.237.214Feb 29, 2024

Evaluation results

MethodLinks
GPT-4Evaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0248.343.2353122.821.831.929.1
Qwen-72B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0243.639.430.12818.719.927.326.4
Qwen-14B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0234.33121.620.89.19.719.517.8
GPT-3.5Evaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0234.331.817.11816.317.520.120.5
Qwen-72B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0231.158.113.943.49.940.416.445.1
GPT-4Evaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0230.757.315.145.47.441.715.746.1
Llama-2-7B-chatEvaluation Protocol=Finetune2024.022954.75.934.2125.56.533.2
Qwen-14B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0228.255.810.642.54.834.912.641.7
Llama-2-70B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0226.425.412.513.96.28.712.713.8
GPT-3.5Evaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0225.753.811.342.67.639.613.343.5
Qwen-72B-chatEvaluation Protocol=DIPMT, Few-shot Context=3-shot2024.022321.319.817.99.810.416.315.1
Llama-2-13B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0222.820.69.111.567.510.711.6
Qwen-7B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0222.425.211.715.55.58.511.414.3
Llama-2-13B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0221.551.67.137.7332.4938.2
Llama-2-70B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0221.550.78.241.32.434.69.340.6
Llama-2-7B-chatEvaluation Protocol=Finetune2024.0221.219.410.111.25.87.611.311.2
Qwen-14B-chatEvaluation Protocol=DIPMT, Few-shot Context=3-shot2024.0220.618.117165.97.712.712.8
Llama-2-7B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0219.819.39.310.94.36.59.710.7
Qwen-7B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0217.648.45.1372.831.57.636.9
Llama-2-7B-chatEvaluation Protocol=DIPMT++, Few-shot Context=3-shot2024.0214.847.44.433.51.1285.933.9
Qwen-7B-chatEvaluation Protocol=DIPMT, Few-shot Context=3-shot2024.0212.813.910.211.23.35.78.39.3
Qwen-72B-chatEvaluation Protocol=DIPMT, Few-shot Context=3-shot2024.029.236.64.531.73.129.45.131.7
Qwen-14B-chatEvaluation Protocol=DIPMT, Few-shot Context=3-shot2024.028.435.44.230.82.126.14.429.6
Qwen-7B-chatEvaluation Protocol=DIPMT, Few-shot Context=3-shot2024.023.928.2124.40.723.31.824.8
mT5-largeEvaluation Protocol=Finetune2024.023.3210.8150.312.51.115.1
mT5-largeEvaluation Protocol=Finetune2024.022.54.60.42.11.11.91.32.6
GPT-4Evaluation Protocol=Direct, Few-shot Context=3-shot2024.020.93.90.83.71.73.91.43.9
GPT-3.5Evaluation Protocol=Direct, Few-shot Context=3-shot2024.020.63.80.53.20.23.40.33.5
GPT-3.5Evaluation Protocol=Direct, Few-shot Context=3-shot2024.020.2140.315.80.517.20.316.1
mT5-baseEvaluation Protocol=Finetune2024.020.21.20.21.40.21.30.21.3
mT5-baseEvaluation Protocol=Finetune2024.020.17.50.27.907.90.17.8
GPT-4Evaluation Protocol=Direct, Few-shot Context=3-shot2024.020.16.20.17.20.17.207