INAR-VL: Input-Aware Routing for Edge-Cloud Vision-Language Inference

• URL: http://arxiv.org/abs/2605.18853v1
• Date: Wed, 13 May 2026 12:44:10 GMT
• Title: INAR-VL: Input-Aware Routing for Edge-Cloud Vision-Language Inference
• Abstract summary: We present INAR-VL, a lightweight edge-cloud routing system for multimodal inference in a two-tier deployment.<n>We show that INAR-VL executes 36% of requests on the edge, reduces latency by 24%, lowers energy by 26%, and preserves 97% of cloud-level accuracy.
• Score: 0.8633013637160062
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Edge deployment of Vision-Language Models (VLMs) faces a tradeoff between latency and accuracy: cloud execution provides high-quality predictions but incurs communication delay and energy cost, while edge-only execution is faster but less accurate due to limited model capacity. This trade-off is further complicated by heterogeneity in image quality and reasoning complexity, making static placement suboptimal. We present INAR-VL, a lightweight edge-cloud routing system for multimodal inference in a two-tier deployment. INAR-VL maintains complementary VLMs across edge and cloud and uses lightweight image and text complexity signals to guide routing and model selection, executing simple queries locally while offloading complex ones when beneficial. Evaluation on visual question answering shows that INAR-VL executes 36% of requests on the edge, reduces latency by 24%, lowers energy by 26%, and preserves 97% of cloud-level accuracy.

Related papers

• Latency-Tolerant Cloud-Edge Collaborative Vision-Language-Action Models via Emergent Representational Specialization [6.831202558046453]
  We introduce CloudEdgeVLA, a cloud-edge policy that treats temporal misalignment as a representation-learning problem.<n>A lightweight edge head combines the latest available cloud feature with current local vision.<n>CloudEdgeVLA retains 63.8--78.0% success with a 40-step uniform-delay window.
  arXiv  Detail & Related papers  (2026-08-01T10:11:59Z)
• ETA-VLA: Efficient Token Adaptation via Temporal Fusion and Intra-LLM Sparsification for Vision-Language-Action Models [14.841746872486333]
  We propose ETA-VLA, an Efficient Token Adaptation framework for VLA models.<n>We show that ETA-VLA achieves driving performance comparable to state-of-the-art baselines.<n>Our method prunes 85% of visual tokens and reduces inference FLOPs by 61%, but still retains 94% of the original accuracy on the NAVSIM v2 benchmark.
  arXiv  Detail & Related papers  (2026-03-26T06:53:42Z)
• AgentVLN: Towards Agentic Vision-and-Language Navigation [78.739525400071]
  Vision-and-Language Navigation (VLN) requires an embodied agent to ground complex natural-language instructions into long-horizon navigation in unseen environments.<n>In this paper, we propose AgentVLN, a novel and efficient embodied navigation framework that can be deployed on edge computing platforms.
  arXiv  Detail & Related papers  (2026-03-18T12:43:47Z)
• Bandwidth-adaptive Cloud-Assisted 360-Degree 3D Perception for Autonomous Vehicles [0.7557499794873328]
  Key challenge for autonomous driving is maintaining real-time situational awareness regarding surrounding obstacles under strict latency constraints.<n>We propose leveraging Vehicle-to-Everything (V2X) communication to partially offload processing to the cloud.<n>Our approach utilizes transformer-based models to fuse multi-camera sensor data into a comprehensive Bird's-Eye View (BEV) representation.
  arXiv  Detail & Related papers  (2026-02-27T10:12:02Z)
• Vision-Language Models on the Edge for Real-Time Robotic Perception [0.22940141855172028]
  Edge intelligence within 6G, particularly Open RAN and Multi-access Edge Computing, offers a pathway to address these challenges.<n>This work investigates the deployment of Vision-Language Models on ORAN/MEC infrastructure using the Unitree G1 humanoid robot as an embodied testbed.<n>Our results show that edge deployment preserves near-cloud accuracy while reducing end-to-end latency by 5%.
  arXiv  Detail & Related papers  (2026-01-21T12:09:48Z)
• InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models [49.08289742711585]
  We propose InfiniteVL, a linear-complexity VLM architecture that synergizes sliding window attention (SWA) with Gated DeltaNet.<n>We show that InfiniteVL achieves over 3.6times inference speedup while maintaining constant latency and memory footprint.<n>In streaming video understanding scenarios, it sustains a stable 24 FPS real-time prefill speed while preserving long-term memory cache.
  arXiv  Detail & Related papers  (2025-12-09T17:18:32Z)
• AVERY: Adaptive VLM Split Computing through Embodied Self-Awareness for Efficient Disaster Response Systems [6.294240680169978]
  Unmanned Aerial Vehicles (UAVs) in disaster response require complex, queryable intelligence that on-board CNNs cannot provide.<n>We present AVERY, a framework that enables VLM deployment through adaptive split computing.
  arXiv  Detail & Related papers  (2025-11-22T18:42:04Z)
• Efficient Onboard Vision-Language Inference in UAV-Enabled Low-Altitude Economy Networks via LLM-Enhanced Optimization [61.55616421408666]
  Low-Altitude Economy Networks (LAENets) have enabled a variety of applications, including aerial surveillance, environmental sensing, and semantic data collection.<n> onboard vision (VLMs) offer inference for real-time inference but limited onboard dynamic network conditions.<n>We propose a UAV-enabled LAENet system that improves communication efficiency under dynamic LAENet conditions.
  arXiv  Detail & Related papers  (2025-10-11T05:11:21Z)
• Open-Vocabulary Spatio-Temporal Scene Graph for Robot Perception and Teleoperation Planning [55.90805559207812]
  In dynamic remote scenes, transmission latency during bidirectional communication creates gaps between remote perceived states and operator intent.<n>We present a representation that enriches open-vocabulary perception with temporal dynamics and lightweight latency annotations.<n>We show that our method achieves 74 percent node accuracy on the Replica benchmark, outperforming Concept.Graph.
  arXiv  Detail & Related papers  (2025-09-27T04:31:24Z)
• Aligned Vector Quantization for Edge-Cloud Collabrative Vision-Language Models [19.938589623698338]
  We introduce an edge-cloud collaborative VQA system called LLaVA-AlignedVQ. LLaVA-AlignedVQ features a novel Aligned Vector Quantization algorithm (AlignedVQ) that efficiently compress intermediate features. Experiments demonstrate that LLaVA-AlignedVQ achieves approximately 1365x compression rate of intermediate features.
  arXiv  Detail & Related papers  (2024-11-08T20:48:37Z)
• An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models [65.37846460916042]
  We find out that the attention computation over visual tokens is of extreme inefficiency in the deep layers of popular LVLMs. We introduce FastV, a versatile plug-and-play method designed to optimize computational efficiency.
  arXiv  Detail & Related papers  (2024-03-11T14:35:32Z)
• Perceiver-VL: Efficient Vision-and-Language Modeling with Iterative Latent Attention [100.81495948184649]
  We present Perceiver-VL, a vision-and-language framework that efficiently handles high-dimensional multimodal inputs such as long videos and text. Our framework scales with linear complexity, in contrast to the quadratic complexity of self-attention used in many state-of-the-art transformer-based models.
  arXiv  Detail & Related papers  (2022-11-21T18:22:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.