RAFT: Reasoning-Aligned Fine-Tuning Improves Robustness of VLA Post-Training
Overview
This research introduces RAFT, a two-stage post-training procedure for vision-language-action (VLA) policies. By first training low-rank adapters on task-relevant reasoning annotations and then freezing them while fine-tuning on action demonstrations, RAFT achieves higher robustness to variations in appearance, objects, and instructions. On a physical DROID robot, it reaches 75.0% out-of-distribution success versus 27.9% for plain fine-tuning. On a multi-stage bimanual YAM task, it attains 70.0% mean success compared to 36.0% for the strongest baseline. The method also demonstrates improved performance under visual, object, and instruction shifts in simulated pick-and-place and open-drawer tasks.
Generated from attributed reports · Updated 2 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · linear-probe accuracy for object identity: 75 other · Basis not reported
“otations, then freezes them while fine-tuning the policy on action demonstrations. On a physical DROID robot, RAFT achieves $75.0\%$ out-of-distribution success, compared with $27.9\%$ for plain fine-tuning; on a multi-stage bimanual YAM task, it reaches $70.0\%$ mean success, compared with $36.0\%$ for the strongest baseline. RAFT also improves success under visual, object, and instruction shifts in simulated pick-a”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalRAFT: Reasoning-Aligned Fine-Tuning Improves Robustness of VLA Post-Training
This research introduces RAFT, a two-stage post-training procedure for vision-language-action (VLA) policies. By first training low-rank adapters on task-relevant reasoning annotations and then freezing them while fine-tuning on action demonstrations, RAFT achieves higher robustness to variations in appearance, objects, and instructions. On a physical DROID robot, it reaches 75.0% out-of-distribution success versus 27.9% for plain fine-tuning. On a multi-stage bimanual YAM task, it attains 70.0% mean success compared to 36.0% for the strongest baseline. The method also demonstrates improved performance under visual, object, and instruction shifts in simulated pick-and-place and open-drawer tasks.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).