AI Science Tracker
EN

Locating and Repairing Domain Shift in VLM Trajectory Planning

Affiliation not stated in the reviewed material · 7 Nov 2026 · computer science

Discovery

Locating and Repairing Domain Shift in VLM Trajectory Planning

Source check: Source checked. Not independent scientific confirmation.

Finding / tested result: Activation patching showed that image and text tokens have distinct domain-shift patterns and that layer-wise patterns were stable across two driving datasets. The proposed MoLaRx selective-repair method had the lowest cross-city trajectory-error gap in six tested settings across three VLM planners; conclusions are limited to those models and benchmarks.

Technology: Vision-language models with layer-local image and text sensitivity

Publication: Preprint · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
Activation patching estimates modality-by-layer domain shift in three driving VLMs; MoLaRx applies matched selective repair to the affected layers.
Limitation / still unknown
The study evaluates the reported models and datasets only; broader generalization and real-world deployment outcomes were not established.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/html/2611.10720v1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: linked separately.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse computer science discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.