Add target2latent goal heads (mlp, mlp_z, shortcut, shortcut_z) under point-lewm and point-delta-jepa; README section

README.md

@@ -43,6 +43,15 @@ model = swm.wm.utils.load_pretrained(f"{local}/point-lewm/")
 [dino_wm](https://github.com/gaoyuezhou/dino_wm) code loads (`hydra.yaml`, `checkpoints/model_latest.pth`,
 `swm_h5_norm_stats.json` with the action normalisation); see the code repository for the evaluation wrapper.
 
+## target2latent goal heads
+
+`point-lewm/target2latent/` and `point-delta-jepa/target2latent/` hold the typed-3-D-goal heads that map an object
+pose (cube position, T pose + ball position, agent position, or wrist + fingertip position) to the goal latent of the
+encoder they sit under, so planning needs no recorded goal observation. Four heads each: `mlp`, `mlp_z` (also
+conditioned on the current latent), `shortcut` (flow-matching shortcut model, samples in 1 to 64 steps), `shortcut_z`.
+Each folder holds one `model.pt` (state dict + architecture config); load with `target2latent.models.load_goal_model`
+from the code repository.
+
 ## Citation
 
 For the most recent citation see the [code repository](https://github.com/fafraob/point-lewm).

point-delta-jepa/target2latent/mlp/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:debaec6ea1bee52c2e2060daa4ac758b74a2ed1c1bd7d1dc388a51f019b77eab
+size 69600045

point-delta-jepa/target2latent/mlp_z/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a53927f76bef589e6c1a313f01cc34eb0d1aedcc19d187ed913a9af7faa08e4
+size 70391075

point-delta-jepa/target2latent/shortcut/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f52c1cc03862a3f827493d31c14d349989430cc02e74fce034221979aab3be
+size 83502144

point-delta-jepa/target2latent/shortcut_z/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bd855ab2a8eee1fa9573792308662963055f79e7897f3fc2ed2f46bd1057ded
+size 83700278

point-lewm/target2latent/mlp/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4cdf039a6dd4d033c816e37e4a2167c58354076121281055687fd9d5badf6dc5
+size 69600045

point-lewm/target2latent/mlp_z/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91ea10a009c7dc6a1b53512a2fd23935575675ebfe3b8c010c897bbb3e3573b2
+size 70391075

point-lewm/target2latent/shortcut/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c45afa57ecfd3cbe5ce39cd098d23992d791bf615f359015a88035640b7e9cc
+size 83502144

point-lewm/target2latent/shortcut_z/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59d45a56f21ed628ecbcbad03d86fc407af014355b715f6bc910b546b84c25f9
+size 83700278