Dolly
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Dolly
27.74Rouge-L
50
dolly_cw
100Accuracy
48
Dolly Eval (test)
29.69ROUGE-L
42
Dolly Closed QA
100ASR
36
Dolly
71.3Score
36
Dolly AC (test)
81.59AUC
33
Dolly
22MP Score
24
Dolly
98.5AP (alpha=0.5)
24
Dolly
30.94Accuracy
24
dolly_cw
99.27Accuracy
24
Dolly
71.4SBERT Similarity
24
Dolly-15K alpha=5.0
35.79Rouge-L
22
Dolly-15K alpha=0.5
35.48Rouge-L
22
Dolly
35.34RougeL
21
Dolly Llama2-13B (test)
75.76Accuracy
21
Dolly Llama2-7B (test)
77.78Acc
21
Dolly
80AUC
20
Dolly AC LLaMA3-8B
83.92Recall
19
Dolly AC LLaMA2-13B
0.974Recall
19
Dolly AC LLaMA2-7B
87.28Recall
19
Dolly Eval
62A Win Count
19
Dolly
0.961Skywork Reward V2 Score
18
Dolly CW
8.88WCS
18
Dolly Out-of-Distribution
49.9GPT-4o Score
17
Dolly databricks 15k (test)
29.7ROUGE-L
14