LLM Response Scenarios
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
LLM Response Scenarios ambiguous lie (test)
68Accuracy
2
LLM Response Scenarios ambiguous truthful reply (test)
85Truthfulness Accuracy
2
LLM Response Scenarios unambiguous lie (test)
91Accuracy (Test)
2
LLM Response Scenarios unambiguous truthful reply (test)
97Truthfulness Accuracy
2
LLM Response Scenarios ambiguous lie LLaMA3-8B-Instruct (test)
—Accuracy
0
LLM Response Scenarios ambiguous truthful reply LLaMA3-8B-Instruct (test)
—Accuracy
0
LLM Response Scenarios unambiguous lie LLaMA3-8B-Instruct (test)
—Accuracy
0
LLM Response Scenarios unambiguous truthful reply LLaMA3-8B-Instruct (test)
—Accuracy
0