aiqbench analysis
The paper shows that conformal prediction intervals with "90% coverage" can hide severe per-cell undercoverage on battery data, with the worst cell covered only 0.1-2% of the time despite healthy pooled numbers. Its hierarchically weighted engine recovers about half that gap using only frozen side information, but it fails under protocol drift and on a cross-chemistry dataset. For ML readers, the main lesson is that calibration guarantees are often marginal, so per-subgroup reporting and worst-case statistics should be standard before interval claims are deployed.
From fleet trajectories to single-cell promises: hierarchically weighted conformal prediction for battery SOH monitoring
arXiv:2610.41696v1 Announce Type: new Abstract: Interval predictions from conformal prediction are usually read as "90% coverage", but on nested battery data that number promises the fleet average, not the individual cell. We measure the gap on 124 public LFP cells under a preregistered three-tier reporting protocol (fleet, batch, cell): pooled coverage looks healthy (0.875-0.906 across splits) while the worst cell is covered 0.1-2% of the time, with exact binomial tests against the nominal level. We then repair part of the gap with a hierarchically weighted conformal engine whose weights consume only side information frozen at prediction time - signature features, batch labels, cycle indices - and never touch residuals. The preregistered repair criterion passes: 10th-percentile cell coverage moves 0.750 to 0.830 (53% of the gap), the worst cell 0.642 to 0.694, with zero deferrals and a flat sensitivity spectrum. What the engine does not repair is localized rather than left open: under protocol drift the frozen weights harm monotonically on the SOH scale (the absolute scale is reported in full), a structural dead end we report instead of tuning away; and repair capacity is bounded by calibration-pool battery diversity, measured on both datasets. On a cross-chemistry external set the calibrated intervals collapse (0.286), while the within-domain skeleton replicates (pooled 0.933, worst cell 0.686). The deliverable is a reporting protocol - three tiers of promise anchored to measured regimes, worst-cell statistics, deferral share - that makes per-cell interval claims auditable before deployment.
Abstract:Interval predictions from conformal prediction are usually read as "90% coverage", but on nested battery data that number promises the fleet average, not the individual cell. We measure the gap on 124 public LFP cells under a preregistered three-tier reporting protocol (fleet, batch, cell): pooled coverage looks healthy (0.875-0.906 across splits) while the worst cell is covered 0.1-2% of the time, with exact binomial tests against the nominal level. We then repair part of the gap with a hierarchically weighted conformal engine whose weights consume only side information frozen at prediction time - signature features, batch labels, cycle indices - and never touch residuals. The preregistered repair criterion passes: 10th-percentile cell coverage moves 0.750 to 0.830 (53% of the gap), the worst cell 0.642 to 0.694, with zero deferrals and a flat sensitivity spectrum. What the engine does not repair is localized rather than left open: under protocol drift the frozen weights harm monotonically on the SOH scale (the absolute scale is reported in full), a structural dead end we report instead of tuning away; and repair capacity is bounded by calibration-pool battery diversity, measured on both datasets. On a cross-chemistry external set the calibrated intervals collapse (0.286), while the within-domain skeleton replicates (pooled 0.933, worst cell 0.686). The deliverable is a reporting protocol - three tiers of promise anchored to measured regimes, worst-cell statistics, deferral share - that makes per-cell interval claims auditable before deployment.
Machine Learning (cs.LG); Applications (stat.AP)
arXiv:2610.41696 [cs.LG]
(or arXiv:2610.41696v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.41696 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 332 words
Related coverage
TrajGenAgent: A Hierarchical LLM Agent for Human Mobility Trajectory Generation
arXiv · Jun 12, 2026
Hierarchical Skill Evolution for Coding Agent
arXiv · Oct 27, 2026
Offline Apprenticeship Learning for Human-Centric Environments
arXiv · Oct 22, 2026
TrajEdit: Real-Time Flow-Matching Trajectory Adaptation via FlowEdit and CBF Safety Filter
arXiv · Oct 20, 2026