SHERLOCK
Introduced by Hessel et al. in The Abduction of Sherlock Holmes: A Dataset for Visual Abductive Reasoning
SHERLOCK is a corpus of 363K commonsense inferences grounded in 103K images. Annotators highlight localized clues (color bubbles) and draw plausible abductive inferences about them (speech bubbles). It can be used for testing machine capacity for abductive reasoning beyond literal image contents.
Benchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
SHERLOCK
|
Dual-Contrast RGPs
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
Dataset Loaders
No data loaders found. You can
submit your data loader here.
No data loaders found. You can
submit your data loader here.
