Note MVC IPT / Visual CoT training (17,079)

Note MVC Text CoT training (16,808)

Note PET IPT / Visual CoT training (20,531)

Note PET Text CoT training (20,531)

Note MVC MessyTable real-world OOD

Note MVC ScanNet OOD

Note PET AI2-THOR - human-verified, in-domain

Note SAT (perspective subset) OOD

Note PT (Path Tracing) train - IPT (11,204)

Note PT train - answer-only (serves no_thought + mixed) (11,204)

Note PT train - Text CoT (11,204)

Note MVC answer-only - serves label-only + Mixed (17,079)

Note PET answer-only - serves label-only + Mixed (20,531)

Note MVC AI2-THOR - human-verified, in-domain

Note PET Habitat - human-verified, different env

Note PT AI2-THOR - human-verified (td_ego_dir/td_path/td_path_arrow)

Note PT Real indoor - human-verified (td_path/td_path_arrow)

weikaih/imaginative-perception-token-mvc-mixed

Image-Text-to-Text • 15B • Updated Jun 8 • 11

Note IPT MVC Mixed model (answer-only + imaginative inference)