--- tags: - libero - latent-dynamics - action-free - robotics - shuffled-future - control --- # Latent dynamics (IDM/FDM) · libero_90 · seed 1000 · shuffled-future control Action-free visual dynamics pretrain: frozen SmolVLA vision features → IDM → FDM. Trained on RGB from `nvidia/libero_90` only (no action / state / reward labels). **Control run:** trained with `shuffle_future=True` — the future RGB frame is sampled from a different episode, so temporal dynamics are broken. Same architecture / steps / data volume as the real IDM; use as a negative control for video-latent fine-tuning (`VIDEO_CKPT=.../idm_fdm.pt` from this repo). Files: - `idm_fdm.pt` — model weights + config dict - `idm_fdm.json` — training hyperparameters Used as frozen teacher for video-latent few-shot SmolVLA fine-tuning (`--video.checkpoint`).