跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.15082 · v1

Prediction or Persistence? What Future-Prediction Scores Miss About World Models

首次提交 2026-11-11 13:56
版本更新 2026-11-11 13:56
雷达观测 2026-11-13 11:11
arXiv ↗PDF ↗

摘要

Pretrained video world models are compared and chosen as components for robot learning largely by future-prediction scores. On its own, such a score cannot be read as evidence of prediction, because it mixes encoder quality, the near future's resemblance to the recent past, and the predictor's own contribution. We separate the three with a matched control. On 27,000 future-retrieval trials from household egocentric video, a model must pick the true continuation of a clip among eleven candidates from the same video (chance 0.091), ranked zero-shot by its own training criterion. Each model is paired with a persistence control, an identical copy whose predictor is never invoked, so the paired difference isolates the predictor. Score and contribution dissociate within the native-video V-JEPA family alone: the predictor adds +0.291 top-1 for V-JEPA 2.1 but +0.03 to +0.04 for V-JEPA 2 on the identical trials. Across five models the contribution ranges from +0.291 to negative, and it is negative only for the two action-conditioned models, which we evaluate outside their training regime. Persistence alone recovers the encoder ordering expected from pretraining, a sign that the probe tracks real signal. We suggest reporting each future-prediction score together with its persistence baseline. The gap between them, not the score, is what makes the number interpretable: a necessary check costing one forward pass. Establishing downstream usefulness is a separate matter, for a planning or control evaluation.

作者备注:Accepted at the CoRL 2026 Workshop "Do Robots Need World Models?"

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

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

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。

v1 · 2026-11-11 13:56 · 最新观测版本

Prediction or Persistence? What Future-Prediction Scores Miss About World Models

Pretrained video world models are compared and chosen as components for robot learning largely by future-prediction scores. On its own, such a score cannot be read as evidence of prediction, because it mixes encoder quality, the near future's resemblance to the recent past, and the predictor's own contribution. We separate the three with a matched control. On 27,000 future-retrieval trials from household egocentric video, a model must pick the true continuation of a clip among eleven candidates from the same video (chance 0.091), ranked zero-shot by its own training criterion. Each model is paired with a persistence control, an identical copy whose predictor is never invoked, so the paired difference isolates the predictor. Score and contribution dissociate within the native-video V-JEPA family alone: the predictor adds +0.291 top-1 for V-JEPA 2.1 but +0.03 to +0.04 for V-JEPA 2 on the identical trials. Across five models the contribution ranges from +0.291 to negative, and it is negative only for the two action-conditioned models, which we evaluate outside their training regime. Persistence alone recovers the encoder ordering expected from pretraining, a sign that the probe tracks real signal. We suggest reporting each future-prediction score together with its persistence baseline. The gap between them, not the score, is what makes the number interpretable: a necessary check costing one forward pass. Establishing downstream usefulness is a separate matter, for a planning or control evaluation.

查看此版本解读 →查看此版本原文 ↗