A unified threat model for LLM-enabled robots reveals three cross-boundary attack chains from user input to unsafe physical actuation due to missing validations and unmediated crossings.
Code as policies: Language model programs for embod- ied control
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A neuro-symbolic 2.5-D decomposition pipeline separates LLM horizontal planning from deterministic vertical execution, achieving 94.6% structural accuracy on the Build What I Mean benchmark.
citing papers explorer
-
2.5-D Decomposition for LLM-Based Spatial Construction
A neuro-symbolic 2.5-D decomposition pipeline separates LLM horizontal planning from deterministic vertical execution, achieving 94.6% structural accuracy on the Build What I Mean benchmark.