You can deploy this model and start earning money today!

shrg7/
MolmoBot-Pi0-DROID-absjoint-RETAIN-a0.5

Views

No views yet

MolmoBot-Pi0-DROID-absjoint — RETAIN (alpha=0.5)

RETAIN baseline (arXiv:2512.08333, Robust Finetuning of VLA Policies via Parameter Merging) for the absolute-joint DROID pi0 policy.

The language-model weights of the finetuned specialist are interpolated back toward the pretrained generalist MolmoBot-Pi0-DROID, while the vision encoder and action expert are kept fully finetuned:

theta_llm = (1 - alpha) * theta_base + alpha * theta_finetuned    (alpha = 0.5)
vision_tower + gemma_expert + action heads : fully finetuned

Base (generalist): MolmoBot-Pi0-DROID
Finetuned (specialist): MolmoBot-Pi0-DROID-absjoint (full-FT, step 5000)
Merged tensors: 164 language-model / 613 kept
alpha = 0.5 (1.0 = pure finetuned, 0.0 = LLM reset to base)

Flat pi0 checkpoint: model.safetensors + metadata.pt + assets/droid_equad/norm_stats.json. Actions are absolute future joint positions q[g+i+1].

shrg7/MolmoBot-Pi0-DROID-absjoint-RETAIN-a0.5

MolmoBot-Pi0-DROID-absjoint — RETAIN (alpha=0.5)

shrg7/
MolmoBot-Pi0-DROID-absjoint-RETAIN-a0.5