Instruction drift is a sampling error in VLA models that CAPS mitigates via power distributions and SNR-based metacognitive MCMC switching, yielding better long-horizon results without retraining.
The maximum number of steps satisfyingP(success)≈(1−ϵ) T ≥ηis: Tef f(Base)≈ −lnη ϵ (24)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Drift is a Sampling Error: SNR-Aware Power Distributions for Long-Horizon Robotic Planning
Instruction drift is a sampling error in VLA models that CAPS mitigates via power distributions and SNR-based metacognitive MCMC switching, yielding better long-horizon results without retraining.