독서리서치레이더투자 프레임워크
로그인 / 회원가입
← 읽기로 돌아가기
뉴스간단한 요점1 분 소요DisconfirmAI 편집부

Single-pass latent steering halves VLM driving inference latency

Authors self-report LCS single-pass steering cuts latency about 50% vs two-pass CFG, with better instruction following and driving scores on Bench2Drive and nuScenes.

검증됨 2026-09-18 13:30 GMT+8

Inference latency for vision-language model end-to-end driving can be cut by about 50% versus standard two-pass CFG guidance, while instruction following and driving scores improve rather than degrade — a result self-reported by Meibo Hu and three co-authors.

Previously such models followed instructions weakly and typically relied on standard two-pass CFG guidance to compensate, at the cost of running inference twice. The authors propose Latent-Centroid Steering (LCS), which replaces this with single-pass latent-space steering that projects toward a precomputed instruction centroid.

By the authors' own measurement, on the Bench2Drive closed-loop and nuScenes open-loop benchmarks, instruction following and driving scores both beat the two-pass CFG baseline, with inference latency reduced by about 50%. The code repository is public.

The result comes from an arXiv preprint (submitted July 31, updated to v3 on September 16, page marked IROS 2026) and has not yet been reproduced by others.

출처

  • arxiv.org↗
  • github.com↗
리서치 계속하기 →
로그인 / 회원가입
독서리서치레이더투자 프레임워크