Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on TruthfulQA (BLEU)
Loading...
49
BLEU Score
Original
19.88
27.44
35
42.56
Nov 14, 2024
BLEU Score
Updated 8d ago
Evaluation Results
Method
Method
Links
BLEU Score
Original
Backbone Model=Llama-3...
2024.11
49
SVD-LLM V2
Backbone Model=Llama-3...
2024.11
46
FiPS
Backbone Model=Llama-3...
2024.11
46
SVD-LLM
Backbone Model=Llama-3...
2024.11
45
ASVD
Backbone Model=Llama-3...
2024.11
37
Original
Backbone Model=Llama-7...
2024.11
30
SVD-LLM V2
Backbone Model=Llama-7...
2024.11
27
FiPS
Backbone Model=Llama-7...
2024.11
27
SVD-LLM
Backbone Model=Llama-7...
2024.11
24
ASVD
Backbone Model=Llama-7...
2024.11
21
Feedback
Search any
task
Search any
task