SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception
Abstract
Vehicle-to-Everything (V2X) cooperative perception improves 3-D detection by sharing intermediate features, but dense remote features may repeat context that the ego agent can infer locally. Most communication-efficient designs optimize masks or codes empirically, leaving a more basic question open: which remote evidence is indispensable given the receiver's own observation? We introduce a closure-fidelity perspective on ego conditioned remote perception. Under a finite deductive abstraction and explicit conditions, its rate--distortion function decomposes over an irredundant core, and the exact zero-distortion rate becomes PA H(πA). This analysis suggests a concrete design principle: transmit compact evidence and recover derivable context with bounded receiver-side inference. Guided by this principle, SemRD-V2X is an operational neural proxy that combines exact-budget BEV support selection, pointwise channel compression, and masked shared-weight reconstruction before standard fusion. Experiments on simulated V2XSet and real-world DAIR-V2X validate the resulting design. In a controlled five-run V2XSet comparison against a locally reproduced V2X-ViT-v1 baseline on one Tesla V100, SemRD-V2X reduces the analytical feature payload by 26.6× while improving [email protected]/[email protected] by 4.13/8.57 points, with 3.81\% additional mean compute latency. These results position closure fidelity as both an analytical lens and an actionable design principle for communication-efficient cooperative perception.
Related
- Cooperative Multi-Task Semantic Communication for Joint Classification and Regression Tasks
- Principal profiles, characteristic communication and critical asymptotics for cooperative nonlocal systems on the circle
- Amortized Localized Posterior Inference for Model Calibration with Application to an Ocean Turbulence Closure
- Distributed LLM Inference via Communication-Efficient Speculative Decoding
- Phylodynamic inference with the bounded coalescent: a point process perspective
BibTeX
@misc{xu2026semrdv2x,
title = {{SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception}},
author = {Hu Xu and Chun Li and Siyuan Qiu and Zeyan Li and Jianfeng Xu},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.34353},
eprint = {2609.34353},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.34353},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception
AU - Hu Xu
AU - Chun Li
AU - Siyuan Qiu
AU - Zeyan Li
AU - Jianfeng Xu
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.34353
UR - https://arxiv.org/abs/2609.34353
AB - Vehicle-to-Everything (V2X) cooperative perception improves 3-D detection by sharing intermediate features, but dense remote features may repeat context that the ego agent can infer locally. Most communication-efficient designs optimize masks or codes empirically, leaving a more basic question open: which remote evidence is indispensable given the receiver's own observation? We introduce a closure-fidelity perspective on ego conditioned remote perception. Under a finite deductive abstraction and explicit conditions, its rate--distortion function decomposes over an irredundant core, and the exact zero-distortion rate becomes P_A H(π_A). This analysis suggests a concrete design principle: transmit compact evidence and recover derivable context with bounded receiver-side inference. Guided by this principle, SemRD-V2X is an operational neural proxy that combines exact-budget BEV support selection, pointwise channel compression, and masked shared-weight reconstruction before standard fusion. Experiments on simulated V2XSet and real-world DAIR-V2X validate the resulting design. In a controlled five-run V2XSet comparison against a locally reproduced V2X-ViT-v1 baseline on one Tesla V100, SemRD-V2X reduces the analytical feature payload by 26.6× while improving [email protected]/[email protected] by 4.13/8.57 points, with 3.81\% additional mean compute latency. These results position closure fidelity as both an analytical lens and an actionable design principle for communication-efficient cooperative perception.
ER -