Metraipolis › archive › dc_33cdef58bef1

EdgeVLN: Runtime-Aware Deployment Ready Quantized Vision Language Navigation Model — arXiv:2609.35570

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35570
License (as archived)
CC0-1.0
Archived
2026-10-07T23:55:13.580Z
SHA-256 of the archived text
5127f800efd3da87e6a39188cc75881e4c7f693924dd1cfa5fdea8712f405b9a
Stable URL of this page
https://metraipolis.com/documento/dc_33cdef58bef1

How to cite

EdgeVLN: Runtime-Aware Deployment Ready Quantized Vision Language Navigation Model — arXiv:2609.35570 — arXiv (en), https://arxiv.org/abs/2609.35570 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_33cdef58bef1 (Metraipolis archive dc_33cdef58bef1, sha256 5127f800efd3da87e6a39188cc75881e4c7f693924dd1cfa5fdea8712f405b9a)
BibTeX
@misc{dc33cdef58bef1,
  title = {EdgeVLN: Runtime-Aware Deployment Ready Quantized Vision Language Navigation Model — arXiv:2609.35570},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35570},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_33cdef58bef1; sha256 5127f800efd3da87e6a39188cc75881e4c7f693924dd1cfa5fdea8712f405b9a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35570. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language navigation (VLN) models perform well but target compute-rich platforms, limiting deployment on memory- and power-constrained robotic edge devices. Compression alone does not establish whether a VLN model fits the memory, latency, and energy budgets of an edge platform while preserving navigation behavior. We introduce EdgeVLN, a runtime-aware, deployment-ready quantized VLN model that closes this gap. EdgeVLN combines a quantized StreamVLN model with Latent Trajectory Termination Extractor (LATTE), a lightweight causal transformer that improves real-time stopping by predicting a Stop Action verifier rank. Both execute through our this http URL VLN driver, which reconstructs streaming context and prunes memory tokens on-board. We characterize a pretrained StreamVLN backbone across weight quantization from 8 to 2 bits and multiple inference runtimes to identify a feasible operating point. LATTE reuses backbone hidden states within the budget freed by quantization, requiring neither a second vision encoder nor an additional backbone forward pass. We evaluate six backbone precisions and seven candidate stop heads on BF16 and IQ4 NL across all 1,839 R2R VLN-CE val-unseen episodes. We measure success rate (SR) in simulation and latency, energy, and resident memory on an NVIDIA Jetson Orin NX 16 GB. LATTE achieves our highest SR, 58.02 percent on the deployed 4-bit model, exceeding the BF16 baseline with only 0.013 s additional latency per navigation step. Four-bit formats achieve nearly identical SR, but step energy varies 36.8 times by execution path. Only IQ4 NL under our VLN driver fits the board, using 11.35 GB resident memory while running 20.8 times faster and using 13.3 times less energy than storage-streamed BF16. INT2 collapses. Runtime selection, memory-token pruning, and quantization are essential for efficient edge deployment.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.35570 — 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 2026-10-07T23:29:27.731Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).