SIEVE: Structure-Aware Data Selection for Imitation Learning with VLA Models
:::info Stub — Full Engineering Breakdown Coming This paper was featured on Hugging Face Daily Papers on 2026-07-07 with 5 upvotes. A full breakdown with production viability rating, implementation notes, and honest limitations is being written. Subscribe to AI Letters → :::
| Authors | Changti Wu et al. |
| Year | 2026 |
| HF Upvotes | 5 |
| arXiv | 2607.06442 |
| Download | |
| HF Page | View on Hugging Face |
Abstract
Vision-Language-Action (VLA) models are typically trained by imitation learning on large-scale robot demonstration datasets, but more data does not necessarily yield better policies due to redundancy, noise, and uneven coverage. Existing data selection methods often assess demonstrations at either the trajectory or state-action level, missing the reusable structures that compose long-horizon behaviors. In this paper, we propose SIEVE, a structure-aware data selection method for VLA imitation learning. SIEVE views demonstrations as compositions of reusable primitives and transition interfaces. It first discovers visuo-motor primitives from segmented trajectories, then allocates selection budgets to composition patterns by maximizing reuse-aware structural exposure under diminishing returns. Finally, it selects medoid trajectories within each composition-pattern bucket to retain central, stable, and imitation-friendly demonstrations. Experiments across multiple datasets, benchmarks, and VLA models show that SIEVE consistently outperforms competitive data selection baselines. Notably, SIEVE can surpass full-data training while using only 50% of demonstrations and 50% of training steps, suggesting that reusable structure, captured through primitives and transitions, is an important signal for efficient VLA imitation learning.
Engineering Breakdown
The Problem
Vision-Language-Action (VLA) models are typically trained by imitation learning on large-scale robot demonstration datasets, but more data does not necessarily yield better policies due to redundancy, noise, and uneven coverage.
The Approach
In this paper, we propose SIEVE, a structure-aware data selection method for VLA imitation learning.
Key Results
Notably, SIEVE can surpass full-data training while using only 50% of demonstrations and 50% of training steps, suggesting that reusable structure, captured through primitives and transitions, is an important signal for efficient VLA imitation learning.
Research Areas
This paper contributes to the following areas of AI/ML engineering:
- Machine learning
- Deep learning
- Neural networks
- Model optimization
- AI systems
- Structureaware
:::tip Subscribe Get weekly breakdowns of papers like this in AI Letters - the newsletter for engineers building production AI systems. :::