Metraipolis › archive › dc_d4bbfbf5fe72

Sensing Which Modality Matters: Evidence-Gated Regularization for Robust VLA Policies — arXiv:2609.03142

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.03142
License (as archived): CC0-1.0
Archived: 2026-10-17T02:08:33.283Z
SHA-256 of the archived text: 2067b2df03b88e924933582fa3a588d6d30aa07cd4ad8c57b9e2b43e65dd7197
Stable URL of this page: https://metraipolis.com/documento/dc_d4bbfbf5fe72

How to cite

Sensing Which Modality Matters: Evidence-Gated Regularization for Robust VLA Policies — arXiv:2609.03142 — arXiv (en), https://arxiv.org/abs/2609.03142 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d4bbfbf5fe72 (Metraipolis archive dc_d4bbfbf5fe72, sha256 2067b2df03b88e924933582fa3a588d6d30aa07cd4ad8c57b9e2b43e65dd7197)

BibTeX

@misc{dcd4bbfbf5fe72,
  title = {Sensing Which Modality Matters: Evidence-Gated Regularization for Robust VLA Policies — arXiv:2609.03142},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03142},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d4bbfbf5fe72; sha256 2067b2df03b88e924933582fa3a588d6d30aa07cd4ad8c57b9e2b43e65dd7197},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03142. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-Language-Action (VLA) policies fuse multimodal sensory inputs, but training on limited and homogeneous robot demonstrations encourages spurious inter-sensor correlations rather than task-relevant signal, a failure we term modality entanglement. Under real-world occlusions and distractors, this manifests as nuisance sensitivity to corruption of uninformative sensors and single-modality insufficiency when only one informative sensor remains intact. We propose Evidence-Gated Regularization (EGR), a modality-agnostic training objective that introduces zero inference-time overhead. EGR derives a per-frame and per-sensor task-relevance signal to gate two state-conditional consistency objectives: invariance on low-evidence sensors, and single-sensor sufficiency on high-evidence ones. We introduce a benchmark based on BEHAVIOR-1K, comprising a fast inference-only diagnostic suite and 47 rollout-based skills targeting modality entanglement. We validate EGR on this benchmark and on two real-robot setups with fundamentally different embodiments: a bi-manual setup with two Kinova arms and three RGB cameras, and a single-arm MELFA ASSISTA setup combining vision and GelSight tactile sensors. EGR improves simulation success rates (SR) from 12.5% to 16.4% under full modalities (+31%), from 9.4% to 16.5% under uninformative-sensor corruption (+75%), and from 2.8% to 6.1% under single-sensor fallback (+120%). Under physical-object distractors, EGR boosts SR from 30% to 85% on the bi-manual setup (+183%) and from 55% to 70% on the tactile setup (+27%).

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.LG).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.03142 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T01:47:25.602Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).