Hugging Face's logo
JLuk's picture

JLuk

jlukabc
Ā·

AI & ML interests

None yet

Recent Activity

reacted to IliaLarchenko's post with šŸ”„ about 1 month ago
šŸ† BEHAVIOR Challenge 1st Place ā€“ Solution Summary My team recently won 1st place in the BEHAVIOR Challenge at NeurIPS. The competition focused on training a single policy to complete 50 long-horizon household tasks in simulation. We built an end-to-end policy based on Pi0.5 with a bunch of custom modifications. Everything is open-sourced, and it should be useful for anyone exploring VLAs or adapting them to specific tasks. Key Architecture Changes: - Replaced language model with 50 trainable task embeddings (no text at all) - Correlated noise for Flow Matching: Ļµ āˆ¼ N(0, 0.5I + 0.5Ī£) using dataset action covariance - Learnable mixed-layer attention: each action expert layer attends to a trainable mix of all VLM layers - System 2 stage tracking: model predicts task stage, we smooth it with voting and feed it back as context Training: - Multi-sample Flow Matching: 15 FM samples per VLM pass to reduce gradient variance - Delta action space + per-timestamp normalization - FAST auxiliary loss and stage prediction loss - Trained on 224Ɨ224 RGB + proprioception only - We use 4 fine-tuned checkpoints, all derived from a multi-task model trained on all 50 tasks Inference Optimizations: - Soft inpainting: predict 30 actions, execute 26, use 4 as an input for the next chunk - Correlation-aware guidance of inpainting to keep action chunks smooth - 1.3Ɨ speedup via cubic spline compression - General correction rule: reopen gripper after failed grasps šŸ”— Code and Models: - Code: https://github.com/IliaLarchenko/behavior-1k-solution - Weights: https://huggingface.co/IliaLarchenko/behavior_submission - Paper: https://huggingface.co/papers/2512.06951
updated a model about 1 month ago
jlukabc/pi0_place_item_on_box_clean_no_bad4_63ep_10k
published a model about 1 month ago
jlukabc/pi0_place_item_on_box_clean_no_bad4_63ep_10k
View all activity

Organizations

None yet