mirror of
https://github.com/huggingface/lerobot.git
synced 2026-07-29 12:39:41 +00:00
feat(data): add recipe-driven language supervision
This commit is contained in:
@@ -0,0 +1,13 @@
|
||||
# Paper-style joint sequence (pi0.5 §IV-B): one sample supervises the subtask
|
||||
# text with CE and, because the assistant turn is part of the prefix, conditions
|
||||
# the FAST and flow action losses on the same annotated subtask in one forward.
|
||||
# The supervised span is attended causally; the action losses see task + subtask.
|
||||
#
|
||||
# Pair with `--policy.joint_subtask_conditioning=true` at inference so the flow
|
||||
# prefix reproduces this layout (task turn with state + causal generated subtask).
|
||||
# Samples without a `subtask` annotation fall back to a plain task-prompt
|
||||
# low-level sample via `if_present`.
|
||||
|
||||
messages:
|
||||
- {role: user, content: "${task}", stream: low_level}
|
||||
- {role: assistant, content: "${subtask}", stream: low_level, target: true, if_present: subtask}
|
||||
Reference in New Issue
Block a user