{"paper":{"title":"Scalable Explainability-as-a-Service (XaaS) for Edge AI Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"XaaS decouples explanation generation from inference so edge devices can cache and reuse explanations, cutting latency by 38 percent while keeping quality high.","cross_cats":["cs.AI","cs.DC","cs.SE"],"primary_cat":"cs.LG","authors_text":"Joyjit Roy, Samaresh Kumar Singh","submitted_at":"2026-02-04T01:28:57Z","abstract_excerpt":"Though Explainable AI (XAI) has made significant advancements, its inclusion in edge and IoT systems is typically ad-hoc and inefficient. Most current methods are \"coupled\" in such a way that they generate explanations simultaneously with model inferences. As a result, these approaches incur redundant computation, high latency and poor scalability when deployed across heterogeneous sets of edge devices. In this work we propose Explainability-as-a-Service (XaaS), a distributed architecture for treating explainability as a first-class system service (as opposed to a model-specific feature). The "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"XaaS reduces latency by 38% while maintaining high explanation quality across three real-world edge-AI deployments.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That semantic similarity reliably identifies reusable explanations without significant loss of fidelity across heterogeneous devices and tasks.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"XaaS decouples explanation generation from model inference via a distributed cache, verification protocol, and adaptive engine, achieving 38% lower latency in three edge-AI use cases.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"XaaS decouples explanation generation from inference so edge devices can cache and reuse explanations, cutting latency by 38 percent while keeping quality high.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"6356f678a959fb30bc0549467400879751f9c73f81d371aacbc2f7e860512870"},"source":{"id":"2602.04120","kind":"arxiv","version":3},"verdict":{"id":"14b1dd79-8960-4099-a558-9c72d0dd6df5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T07:30:37.551742Z","strongest_claim":"XaaS reduces latency by 38% while maintaining high explanation quality across three real-world edge-AI deployments.","one_line_summary":"XaaS decouples explanation generation from model inference via a distributed cache, verification protocol, and adaptive engine, achieving 38% lower latency in three edge-AI use cases.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That semantic similarity reliably identifies reusable explanations without significant loss of fidelity across heterogeneous devices and tasks.","pith_extraction_headline":"XaaS decouples explanation generation from inference so edge devices can cache and reuse explanations, cutting latency by 38 percent while keeping quality high."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.04120/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":1,"snapshot_sha256":"fd940516ebe4c269a91967a5bc6116a9989d5e0da5d9207850c0b95e3f1b5976"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}