{"paper":{"title":"DSPR: Dual-Stream Physics-Residual Networks for Trustworthy Industrial Time Series Forecasting","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"DSPR decouples stable temporal patterns from regime-dependent residual dynamics to forecast industrial time series with high accuracy and physical consistency.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Guoqing Wang, Pengwei Yang, Tianyu Li, Yeran Zhang","submitted_at":"2026-04-08T06:21:10Z","abstract_excerpt":"Accurate forecasting of industrial time series requires balancing predictive accuracy with physical plausibility under non-stationary operating conditions. Existing data-driven models often achieve strong statistical performance but struggle to respect regime-dependent interaction structures and transport delays inherent in real-world systems. To address this challenge, we propose DSPR (Dual-Stream Physics-Residual Networks), a forecasting framework that explicitly decouples stable temporal patterns from regime-dependent residual dynamics. The first stream models the statistical temporal evolu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on four industrial benchmarks spanning heterogeneous regimes demonstrate that DSPR consistently improves forecasting accuracy and robustness under regime shifts while maintaining strong physical plausibility. It achieves state-of-the-art predictive performance, with Mean Conservation Accuracy exceeding 99% and Total Variation Ratio reaching up to 97.2%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the physical priors embedded in the Physics-Guided Dynamic Graph accurately reflect true regime-dependent interaction structures without introducing bias or suppressing valid correlations, and that the Adaptive Window module reliably estimates flow-dependent transport delays from data alone.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DSPR decouples statistical temporal evolution from physics-informed residual dynamics via an adaptive window for transport delays and a physics-guided dynamic graph to achieve accurate, physically plausible forecasts on industrial benchmarks with over 99% mean conservation accuracy.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"DSPR decouples stable temporal patterns from regime-dependent residual dynamics to forecast industrial time series with high accuracy and physical consistency.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"80c653d6f1a0bafd49fdcf261c90d6af1ada24a7a4579f0a77a39dcf39c7c3b0"},"source":{"id":"2604.07393","kind":"arxiv","version":2},"verdict":{"id":"7df57715-5f0f-4ae7-9b2d-d922eec94555","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T18:09:01.243903Z","strongest_claim":"Experiments on four industrial benchmarks spanning heterogeneous regimes demonstrate that DSPR consistently improves forecasting accuracy and robustness under regime shifts while maintaining strong physical plausibility. It achieves state-of-the-art predictive performance, with Mean Conservation Accuracy exceeding 99% and Total Variation Ratio reaching up to 97.2%.","one_line_summary":"DSPR decouples statistical temporal evolution from physics-informed residual dynamics via an adaptive window for transport delays and a physics-guided dynamic graph to achieve accurate, physically plausible forecasts on industrial benchmarks with over 99% mean conservation accuracy.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the physical priors embedded in the Physics-Guided Dynamic Graph accurately reflect true regime-dependent interaction structures without introducing bias or suppressing valid correlations, and that the Adaptive Window module reliably estimates flow-dependent transport delays from data alone.","pith_extraction_headline":"DSPR decouples stable temporal patterns from regime-dependent residual dynamics to forecast industrial time series with high accuracy and physical consistency."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.07393/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":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}