HoneyLane/vlmphysics-human-eval
0
Pairwise video preference comparison for the VLM Physics paper rebuttal. Pick a model, watch each (A, B) video pair, rate preference (1-5), export ratings JSON when done.
Pairwise video preference comparison for the VLM Physics paper rebuttal. Pick a model, watch each (A, B) video pair, rate preference (1-5), export ratings JSON when done.