doubleblindsubmission/Monet
0
Leveraging Latent Visual Reasoning in Silence
Official repo for the paper.
Prep env
see
prepare_everything.shTraining
see
examples/monet-rl-thymerl-attn1-latent0-includelatentsample1-kl0.01-temp0.5-rollout8-numgpu2.shEval
see VLMEvalKit
