Procedural Mistake Detection via Action Effect Modeling

Wenliang Guo, Yujiang Pu, Yu Kong · 2025

Read Procedural Mistake Detection via Action Effect Modeling by Wenliang Guo, Yujiang Pu, Yu Kong (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-03474.txt.

Description

Mistakes in a task often show in the result, not the motion. Wenliang Guo, Yujiang Pu, and Yu Kong present a mistake detection method in this 2025 work. Procedural tasks, like assembly or cooking, need systems that support learning and execution.

Most past methods analyze how an action is performed. The authors instead model the action effect, what the action produces. Wrong object states or layouts can reveal an error that motion alone hides. Effect-based checks could make tutoring and safety systems more helpful. Effects are often easier to verify than motions, since outcomes can be checked by sensors.

Catalog record

CREATORWenliang Guo, Yujiang Pu, Yu Kong
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE38 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-03474

Browse the full library · About the Tower of Babel