fine-tuning-openvla-oft
FeaturedFine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.
Install
Quality Score: 94/100
Skill Content
Details
- Author
- Orchestra-Research
- Repository
- Orchestra-Research/AI-Research-SKILLs
- Created
- 1 years ago
- Last Updated
- 1 months ago
- Language
- TeX
- License
- MIT
Integrates with
Similar Skills
Semantically similar based on skill content — not just same category
ml-multimodal-robotics
Use when working with multimodal ML models or vision-language-action (VLA) robotics policies: zero-shot image-text matching and retrieval (CLIP), image captioning and VQA (BLIP-2), conversational visual chat (LLaVA), promptable segmentation (Segment Anything / SAM), text-to-image diffusion and LoRA/ControlNet (Stable Diffusion / Diffusers), speech transcription and translation (Whisper), music and sound-effect generation (AudioCraft / MusicGen / AudioGen), or robot policy fine-tuning, serving, and simulation eval (OpenPI pi0/pi0.5, OpenVLA-OFT, NVIDIA Cosmos Policy on LIBERO, ALOHA, DROID, RoboCasa).
ml-finetuning-rl
Use when fine-tuning or post-training LLMs: SFT, LoRA/QLoRA/DoRA/PEFT, Unsloth, Axolotl, LLaMA-Factory, TRL (SFTTrainer, DPOTrainer, RewardTrainer, GRPOTrainer, PPO), SimPO, or distributed RLHF/RLVR with OpenRLHF, veRL, Slime, Miles, or TorchForge. Triggers on 'LoRA', 'QLoRA', 'PEFT', 'Unsloth', 'Axolotl', 'LLaMA-Factory', 'TRL', 'DPO', 'SimPO', 'GRPO', 'PPO', 'RLHF', 'OpenRLHF', 'veRL', 'Slime', 'Miles', 'TorchForge'.
openrlhf-training
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.