SCALE: Real-Time Latent Communication for Cyber-Physical Agents
SCALE: Real-Time Latent Communication for Cyber-Physical Agents
This paper introduces SCALE, a method enabling cyber-physical agents to exchange intermediate model representations efficiently. By matching tokens across consecutive latent messages using similarity scores and profiling semantic gain versus communication cost, SCALE employs a real-time scheduler to optimize transmissions under bandwidth constraints and heterogeneous task deadlines. The approach is evaluated on three benchmarks and a physical spider robot case study, demonstrating improved task performance and on-time delivery in bandwidth-constrained networks. Code is available at https://anonymous.4open.science/r/SCALE-4E80/.
The authors propose SCALE, a real-time latent communication method for cyber-physical agents that matches tokens across messages to reduce redundancy and schedules transmissions based on bandwidth and deadlines; evaluation covers three benchmarks and a physical spider robot case study under diverse network conditions.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
benchmarks
3
Reported trials
View original evidence
these profiles, SCALE employs a real-time scheduler that jointly determines which pending messages to transmit and how many latent tokens to send, subject to available bandwidth and task deadlines. We evaluate SCALE across three cyber-physical benchmarks and a physical spider robot case study under diverse network conditions and heterogeneous task deadlines. Our results show that SCALE significantly improves both tas
Open source S4
- code: Available Artifact link Open source S5
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
these profiles, SCALE employs a real-time scheduler that jointly determines which pending messages to transmit and how many latent tokens to send, subject to available bandwidth and task deadlines. We evaluate SCALE across three cyber-physical benchmarks and a physical spider robot case study under diverse network conditions and heterogeneous task deadlines. Our results show that SCALE significantly improves both tas
Open source S4
k performance and on-time delivery in bandwidth-constrained networks. Our code is available at: https://anonymous.4open.science/r/SCALE-4E80/.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org