{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EAXWJMPE6FVBSH5GI35CHZCSFT","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2b1e0257f26ddc4287a3634db34e00062610d13db29460635fa776ee116c13f4","cross_cats_sorted":["cs.HC"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-02-06T09:26:29Z","title_canon_sha256":"3bb728d23c539060c54d30b33c7a7d14dbfc490dbee118104af491b3b11ac7c5"},"schema_version":"1.0","source":{"id":"2602.06523","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.06523","created_at":"2026-05-20T00:03:05Z"},{"alias_kind":"arxiv_version","alias_value":"2602.06523v3","created_at":"2026-05-20T00:03:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.06523","created_at":"2026-05-20T00:03:05Z"},{"alias_kind":"pith_short_12","alias_value":"EAXWJMPE6FVB","created_at":"2026-05-20T00:03:05Z"},{"alias_kind":"pith_short_16","alias_value":"EAXWJMPE6FVBSH5G","created_at":"2026-05-20T00:03:05Z"},{"alias_kind":"pith_short_8","alias_value":"EAXWJMPE","created_at":"2026-05-20T00:03:05Z"}],"graph_snapshots":[{"event_id":"sha256:034d0952bc8158e2e542ef03ab4ae50ed68774ed8ec395114fa048238064957e","target":"graph","created_at":"2026-05-20T00:03:05Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"MicroBi-ConvLSTM is the only architecture achieving full 8/8 dataset coverage on both platforms, with 72.8 ms average latency on Pico 2 and 97.9% PyTorch parity on ESP32, while all three baselines show partial or complete deployment failure."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the reported parameter count and accuracy hold after accounting for operating-system overhead on the target microcontrollers and that the chosen eight benchmarks are representative enough to generalize to real-world deployment without additional tuning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"MicroBi-ConvLSTM achieves competitive accuracy on eight HAR benchmarks with only 11.4K parameters and successful full deployment on two microcontroller platforms where prior models fail."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"MicroBi-ConvLSTM deploys successfully on all eight HAR benchmarks across both Pico 2 and ESP32 microcontrollers with an average of 11.4K parameters."}],"snapshot_sha256":"335a0ee9f3d46c1303cec2519b9f8ec41793832cb4b41a48862b8f2301c76ca0"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"4303436e0504181b7c7903064b1540dbae207c689b6820c4139e9caba2098d32"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2602.06523/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human Activity Recognition (HAR) on resource constrained wearables requires models that balance accuracy against strict memory and computational budgets. State of the art lightweight architectures such as TinierHAR (34K parameters) and TinyHAR (55K parameters) achieve strong accuracy, but exceed memory budgets of microcontrollers with limited SRAM once operating system overhead is considered.\n  We present MicroBi-ConvLSTM, an ultra-lightweight convolutional recurrent architecture achieving 11.4K parameters on average through two stage convolutional feature extraction with 4x temporal pooling, ","authors_text":"Mridankan Mandal","cross_cats":["cs.HC"],"headline":"MicroBi-ConvLSTM deploys successfully on all eight HAR benchmarks across both Pico 2 and ESP32 microcontrollers with an average of 11.4K parameters.","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-02-06T09:26:29Z","title":"MicroBi-ConvLSTM: An Ultra-Lightweight Efficient Model for Human Activity Recognition on Resource Constrained Devices"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.06523","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-16T07:14:06.687629Z","id":"103f1875-ed64-4d40-9f8b-a42eeaeb863d","model_set":{"reader":"grok-4.3"},"one_line_summary":"MicroBi-ConvLSTM achieves competitive accuracy on eight HAR benchmarks with only 11.4K parameters and successful full deployment on two microcontroller platforms where prior models fail.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"MicroBi-ConvLSTM deploys successfully on all eight HAR benchmarks across both Pico 2 and ESP32 microcontrollers with an average of 11.4K parameters.","strongest_claim":"MicroBi-ConvLSTM is the only architecture achieving full 8/8 dataset coverage on both platforms, with 72.8 ms average latency on Pico 2 and 97.9% PyTorch parity on ESP32, while all three baselines show partial or complete deployment failure.","weakest_assumption":"That the reported parameter count and accuracy hold after accounting for operating-system overhead on the target microcontrollers and that the chosen eight benchmarks are representative enough to generalize to real-world deployment without additional tuning."}},"verdict_id":"103f1875-ed64-4d40-9f8b-a42eeaeb863d"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0a613b681a9329fa1187f49b1e07445347eb1debd22ecec01b51fbc285fe3ee5","target":"record","created_at":"2026-05-20T00:03:05Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2b1e0257f26ddc4287a3634db34e00062610d13db29460635fa776ee116c13f4","cross_cats_sorted":["cs.HC"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-02-06T09:26:29Z","title_canon_sha256":"3bb728d23c539060c54d30b33c7a7d14dbfc490dbee118104af491b3b11ac7c5"},"schema_version":"1.0","source":{"id":"2602.06523","kind":"arxiv","version":3}},"canonical_sha256":"202f64b1e4f16a191fa646fa23e4522cf50b35d34834f53598470ca80f38069c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"202f64b1e4f16a191fa646fa23e4522cf50b35d34834f53598470ca80f38069c","first_computed_at":"2026-05-20T00:03:05.027582Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:05.027582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qIf26TmFgIU283TLIky1qCfDUXrq78EEfehbFiqhsNQlno1McKXqvpjHs5uX3k7VVQ6JCZNMuSxqxqSWiPtTCw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:05.028441Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.06523","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0a613b681a9329fa1187f49b1e07445347eb1debd22ecec01b51fbc285fe3ee5","sha256:034d0952bc8158e2e542ef03ab4ae50ed68774ed8ec395114fa048238064957e"],"state_sha256":"a058bf0c388a063bbb434cdc39c42c810ad4b23cdb899186b3d65fd18f52deae"}