FaultBench-Industrial Research documentation
Research documentation ยท Experimental record
Sensor noise / Permanent

LLM-agent replanning

Permanent sensor noise evaluation

Fault: Sensor noise
Duration: Permanent
Recovery mechanism: LLM-agent replanning

Experiment definition

The experiment evaluates LLM-agent replanning under a permanent sensor noise injection. Fault onset is randomized between environment steps 40 and 60.

Expected evaluation record

MetricRecorded value
Degradation-area25.74 ± 10.36 (corrected; mean ± 95% CI, n=24)
Safety-violation excess132.8 ± 55.8 (weighted, control-subtracted; mean ± 95% CI, n=24)
Fraction of time healthy0.491 ± 0.055 (mean ± 95% CI, n=24)
Relapse count0.21 ± 0.18 (mean ± 95% CI, n=24)
False-recovery rate0.016 ± 0.016 (mean ± 95% CI, n=24)
Intervention recall0.422 ± 0.076 (mean ± 95% CI, n=24)

Values are means with 95% confidence intervals across every target × duration-fixed combination shown on this page (see graphs below), 8 seeds each. See the Results page for the cross-mechanism comparison.

Experimental trajectory graphs

One representative seed (seed 0 of 8) per target combination from the full sweep. See the Results page for the multi-seed aggregate comparison across mechanisms. The graphs for every seed are archived at doi:10.5281/zenodo.23304785 (files available from 10 November 2026).

Sorting Agent Target

Sensor noise permanent fault, LLM-agent replanning, Sorting Agent Target
LLM-agent replanning - Sorting Agent Target - evaluation trajectory under permanent sensor noise (fault onset randomized between steps 40-60), seed 0 of the 8-seed sweep.

Pressing Agent Target

Sensor noise permanent fault, LLM-agent replanning, Pressing Agent Target
LLM-agent replanning - Pressing Agent Target - evaluation trajectory under permanent sensor noise (fault onset randomized between steps 40-60), seed 0 of the 8-seed sweep.

Both Agents Target

Sensor noise permanent fault, LLM-agent replanning, Both Agents Target
LLM-agent replanning - Both Agents Target - evaluation trajectory under permanent sensor noise (fault onset randomized between steps 40-60), seed 0 of the 8-seed sweep.

Notes

This page is intentionally separated from the paper-facing results so that full-resolution plots, additional seeds, and diagnostic figures can be documented without overloading the publication.