# FRESHLATENT: Channel-Aware Latent Adaptation for Resource-Constrained Embodied VLM Perception

arXiv: https://arxiv.org/abs/2609.30629 · HTML: https://mlsanity.com/paper/2609.30629/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 24 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 18 of 18)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Semantic Understanding of Scenes through the ADE20K Dataset](https://mlsanity.com/paper/1608.05442/) (2016)
- [LISA: Reasoning Segmentation via Large Language Model](https://mlsanity.com/paper/2308.00692/) (2023)
- [Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges](https://mlsanity.com/paper/2103.04505/) (2021)
- [Joint Source-Channel Coding: Fundamentals and Recent Progress in Practical Designs](https://mlsanity.com/paper/2409.17557/) (2024)
- [BottleFit: Learning Compressed Representations in Deep Neural Networks for Effective and Efficient Split Computing](https://mlsanity.com/paper/2201.02693/) (2022)
- [Dynamic Split Computing for Efficient Deep Edge Intelligence](https://mlsanity.com/paper/2205.11269/) (2022)
- [Air-Ground Collaboration for Language-Specified Missions in Unknown Environments](https://mlsanity.com/paper/2505.09108/) (2025)
- [Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects](https://mlsanity.com/paper/2604.05546/) (2026)
- [UAV-VLRR: Vision-Language Informed NMPC for Rapid Response in UAV Search and Rescue](https://mlsanity.com/paper/2503.02465/) (2025)
- [Deep Joint Source-Channel Coding for Multi-Task Network](https://mlsanity.com/paper/2109.05779/) (2021)
- [ACCESS-AV: Adaptive Communication-Computation Codesign for Sustainable Autonomous Vehicle Localization in Smart Factories](https://mlsanity.com/paper/2507.20399/) (2025)
- [AVERY: Intent-Driven Adaptive VLM Split Computing via Embodied Self-Awareness for Efficient Disaster Response Systems](https://mlsanity.com/paper/2511.18151/) (2025)
- [PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation](https://mlsanity.com/paper/2604.15670/) (2026)
- [MemoGuard: An Adaptive Runtime for Guarding Against Memory Traps in Communication-Limited Robot Navigation](https://mlsanity.com/paper/2607.15589/) (2026)
- [Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models](https://mlsanity.com/paper/2604.26508/) (2026)
- [FloodReasonBench: Benchmarking VLM Reasoning Segmentation for Embodied Flood Response at the Edge](https://mlsanity.com/paper/2608.15410/) (2026)
- [Compressing AI Traffic: Standardized Neural Network Coding of Visual-Token Representations in Split Vision-Language Inference](https://mlsanity.com/paper/2609.01200/) (2026)

## Abstract (arXiv)

Mission-critical UAVs increasingly rely on split vision-language model (VLM) perception under tight onboard-resource and wireless-communication constraints. However, corruption of transmitted intermediate features creates a deployment mismatch for clean-trained split interfaces, while stronger channel-aware codecs can impose substantial onboard cost. We present FreshLatent, a lightweight channel-aware latent adapter that trains a power-normalized encoder-decoder through wireless corruption while keeping the surrounding VLM frozen. We formulate deployment around a mission-conditioned perception requirement and embedded interface cost, linking channel quality and communication budget to the operating conditions under which perception remains usable. At 0 dB and the tightest communication budget, FreshLatent improves gIoU and cIoU over clean split compression by 20.79 and 20.87 points, respectively. At the most adverse evaluated SNR (0 dB), across all three communication budgets, FreshLatent recovers 63.5-69.1% of the gIoU improvement achieved by a much heavier, range-trained feature-JSCC codec. On an NVIDIA Jetson AGX Xavier in 10-W mode, FreshLatent uses 37-40x fewer encoder parameters, 7.7-9.9x lower edge-interface latency, and 8.8-10.0x lower edge-interface energy than the heavier codec. Together, these results show that lightweight channel-aware adaptation can recover a substantial fraction of the robustness of a much larger communication interface while broadening quality-valid operation under constrained wireless conditions.