REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs
Updated 52 min ago · first seen 11 Sept 2026
paper_01M294AHKW5WDXFV7XAS5NW72J
- Published
- 2 Sept 2026
- T1 · 52 min ago
- arXiv
- 2609.01215
- T1 · 52 min ago
Abstract
Most current vision-language-action (VLA) models—such as OpenVLA, π0, RT-2, and RDT-1B—are “monolithic.” This means they generate raw motor commands or very short sequences of actions, without organizing behaviors into reusable, well-defined abstractions. As a result, these models perform poorly on long-horizon (multi-step) tasks, and it’s difficult to interpret what they have learned. Existing approaches for discovering skills often avoid the core problem of deciding when two action sequences are “behaviorally equivalent.” For example, AtomicVLA and AtomSkill group action sequences by…
Authors 2
Riyaaz Shaik, Chandru Venkataraman
Specification
- Paper
Source:Apple Machine Learning ResearchT1observed 52 min agohigh
- arXiv id
- 2609.01215
Source:Apple Machine Learning ResearchT1observed 52 min agohigh
Source:Apple Machine Learning ResearchT1observed 52 min agohigh
- Published
- 2 Sept 2026
Source:Apple Machine Learning ResearchT1observed 52 min agohigh
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
6
Source tiers
T16
Freshest observation
52 min ago
Conflicts
None
No models linked to this paper yet.
- Published by
- Apple
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history
Paperpaper_url1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| https://machinelearning.apple.com/research/refactor-vla-motor-programs | → current | current | Apple Machine Learning ResearchT1 | high | deterministic |
Abstractabstract1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| Most current vision-language-action (VLA) models—such as OpenVLA, π0, RT-2, and RDT-1B—are “monolithic.” This means they generate raw motor commands or very short sequences of actions, without organizing behaviors into reusable, well-defined abstractions. As a result, these models perform poorly on long-horizon (multi-step) tasks, and it’s difficult to interpret what they have learned. Existing approaches for discovering skills often avoid the core problem of deciding when two action sequences are “behaviorally equivalent.” For example, AtomicVLA and AtomSkill group action sequences by… | → current | current | Apple Machine Learning ResearchT1 | high | deterministic |
arXiv idarxiv_id1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| 2609.01215 | → current | current | Apple Machine Learning ResearchT1 | high | deterministic |
PDFpdf_url1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| https://arxiv.org/pdf/2609.01215 | → current | current | Apple Machine Learning ResearchT1 | high | deterministic |
Publishedpublished_at1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| 2 Sept 2026 | → current | current | Apple Machine Learning ResearchT1 | high | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
New paper: REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs (Apple)
apple_ml
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Apple Machine Learning Research | machinelearning.apple.com/rss.xml | feed | T1· Official | 49 min ago | 1 |
| Apple Machine Learning Research | machinelearning.apple.com/research/refactor-vla-motor-programs | paper_page | T1· Official | 52 min ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.