R-HTN: Rebellious Online HTN Planning for Safety and Game AI

Paola Rizzo · 2026

Read R-HTN: Rebellious Online HTN Planning for Safety and Game AI by Paola Rizzo (2026) 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-2602-00951.txt.

Description

Hector Munoz-Avila and colleagues give task planners a right to refuse unsafe or out-of-role orders. Their R-HTN agents follow built-in rules, called directives, on user-set tasks. One type halts when it finds it would break a rule. The other type rewrites its own plan to seek a new route to the same goal. Tests in two task areas found these agents never broke a rule, yet still aimed for the set goal when a route could be found.

Catalog record

CREATORPaola Rizzo
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE44625
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602.00951

Browse the full library · About the Tower of Babel