TheMachine Press

A daily newspaper for the age of artificial intelligence.

Morning editionPermanent story

robotics

The Driving Model Spent Yesterday’s Saved Milliseconds

SlackDrive used realized latency to choose each next control step's compute budget.

Published Updated Story ID: mp-2026-09-24-027
Read the complete editionStory JSON

Summary

SlackDrive used realized latency to choose each next control step's compute budget.

SlackDrive profiles a small set of inference budgets, watches recent execution time and selects the highest-utility budget predicted to stay within the next control deadline. On NAVSIM v2 with DriveDreamer-Policy, the authors report 21.7% better latency-constrained EPDMS than the strongest baseline under heavy contention, while full-budget and static token-pruning methods missed the allowed envelope. These are simulator and benchmark results, not evidence of safe road operation.

Why it matters

SlackDrive used realized latency to choose each next control step's compute budget.

Limits and context

  • These are simulator and benchmark results, not evidence of safe road operation.

Key claims

  1. SlackDrive used realized latency to choose each next control step's compute budget.

    Qualification: These are simulator and benchmark results, not evidence of safe road operation.

    Evidence: source-2026-09-24-016

Sources

  1. arXiv preprint 2609.28064arXiv · primary research

Corrections

No corrections have been recorded for this story.