跳到正文
研踪PaperTrail
← 研踪 PaperTrail
正在查看已观测的 v1。查看最新 v2 →

arXiv:2611.19839 · v1

Passable for Whom? Few-Shot Personal Calibration of Sidewalk Accessibility Perception

首次提交 2026-11-15 14:42
版本更新 2026-11-15 14:42
雷达观测 2026-11-17 12:06
arXiv ↗PDF ↗

摘要

Whether a sidewalk is passable depends on who is trying to pass: people who use the same mobility aid disagree about the same image, so one label per scene, or per aid, is wrong for many of them. We introduce few-shot personal accessibility perception: from K answers of a person never seen before, predict their judgment of sidewalk scenes never seen before. PassCal separates a shared description of the scene from the person's interpretation of it. The description comes from the internal state of a frozen vision-language model (VLM) and from barrier type, severity and context properties predicted by a detector trained on 14.9k public audits; the interpretation, a calibration of the person's response style and a weighting of the properties, is fit from their answers. On 8,883 judgments in which each of 135 verified mobility-aid users rated the same 50 sidewalk scenes, with people and scenes both held out, eight answers lower the Brier score from 0.550 to 0.479. The same VLM prompted with those answers reaches 0.562, and the same calibration without the scene 0.521. Conformal risk control turns predictions into passable, blocked or "depends on you" statements with a bounded expected rate of wrong statements over people; at 5%, eight answers raise the share of decisive scenes from 27% to 42%. Audit-supervised properties improve all sixteen vision and VLM encoders we test and outperform attributes from a sidewalk navigation VLM.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

仅对比实际观测的元数据,不代表 PDF 正文或实验变化。版本可能不连续;带删除线为移除,下划线为新增。

标题变化

Passable for Whom? Few-Shot Learning for Personal Calibration of Sidewalk Accessibility Perception for Human-Centered Autonomous Driving

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-18 04:07 · 最新观测版本

Passable for Whom? Few-Shot Learning for Personal Calibration of Sidewalk Accessibility Perception for Human-Centered Autonomous Driving

Whether a sidewalk is passable depends on who is trying to pass: people who use the same mobility aid disagree about the same image, so one label per scene, or per aid, is wrong for many of them. We introduce few-shot personal accessibility perception: from K answers of a person never seen before, predict their judgment of sidewalk scenes never seen before. PassCal separates a shared description of the scene from the person's interpretation of it. The description comes from the internal state of a frozen vision-language model (VLM) and from barrier type, severity and context properties predicted by a detector trained on 14.9k public audits; the interpretation, a calibration of the person's response style and a weighting of the properties, is fit from their answers. On 8,883 judgments in which each of 135 verified mobility-aid users rated the same 50 sidewalk scenes, with people and scenes both held out, eight answers lower the Brier score from 0.550 to 0.479. The same VLM prompted with those answers reaches 0.562, and the same calibration without the scene 0.521. Conformal risk control turns predictions into passable, blocked or "depends on you" statements with a bounded expected rate of wrong statements over people; at 5%, eight answers raise the share of decisive scenes from 27% to 42%. Audit-supervised properties improve all sixteen vision and VLM encoders we test and outperform attributes from a sidewalk navigation VLM.

查看此版本解读 →查看此版本原文 ↗
v1 · 2026-11-15 14:42

Passable for Whom? Few-Shot Personal Calibration of Sidewalk Accessibility Perception

Whether a sidewalk is passable depends on who is trying to pass: people who use the same mobility aid disagree about the same image, so one label per scene, or per aid, is wrong for many of them. We introduce few-shot personal accessibility perception: from K answers of a person never seen before, predict their judgment of sidewalk scenes never seen before. PassCal separates a shared description of the scene from the person's interpretation of it. The description comes from the internal state of a frozen vision-language model (VLM) and from barrier type, severity and context properties predicted by a detector trained on 14.9k public audits; the interpretation, a calibration of the person's response style and a weighting of the properties, is fit from their answers. On 8,883 judgments in which each of 135 verified mobility-aid users rated the same 50 sidewalk scenes, with people and scenes both held out, eight answers lower the Brier score from 0.550 to 0.479. The same VLM prompted with those answers reaches 0.562, and the same calibration without the scene 0.521. Conformal risk control turns predictions into passable, blocked or "depends on you" statements with a bounded expected rate of wrong statements over people; at 5%, eight answers raise the share of decisive scenes from 27% to 42%. Audit-supervised properties improve all sixteen vision and VLM encoders we test and outperform attributes from a sidewalk navigation VLM.

查看此版本解读 →查看此版本原文 ↗