{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7KMYFIYAYJPU2X5HCHMLEKTS5P","short_pith_number":"pith:7KMYFIYA","schema_version":"1.0","canonical_sha256":"fa9982a300c25f4d5fa711d8b22a72ebe9e187d70418b7ed2163e4a3726c3e04","source":{"kind":"arxiv","id":"2505.08599","version":1},"attestation_state":"computed","paper":{"title":"MINIMALIST: switched-capacitor circuits for efficient in-memory computation of gated recurrent units","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.SP"],"primary_cat":"cs.AR","authors_text":"Filippo Moro, Laura Kriener, Melika Payvand, Sebastian Billaudelle, Tristan Torchet","submitted_at":"2025-05-13T14:13:41Z","abstract_excerpt":"Recurrent neural networks (RNNs) have been a long-standing candidate for processing of temporal sequence data, especially in memory-constrained systems that one may find in embedded edge computing environments. Recent advances in training paradigms have now inspired new generations of efficient RNNs. We introduce a streamlined and hardware-compatible architecture based on minimal gated recurrent units (GRUs), and an accompanying efficient mixed-signal hardware implementation of the model. The proposed design leverages switched-capacitor circuits not only for in-memory computation (IMC), but al"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.08599","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-05-13T14:13:41Z","cross_cats_sorted":["cs.AI","cs.LG","eess.SP"],"title_canon_sha256":"1e74a5f70b7d1598880876285cc6d13dbcb98248b18964a2aa3e08d58a2226de","abstract_canon_sha256":"6078da1a1f7d8bbbb2b5fc51195769e0fdd61ebe01177c1314a74f36b286944a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:33.307127Z","signature_b64":"7Bvfq+unS+7/H2YfJaJvPEL7l9h1jmkhLEMmqBR05hkVVU5vXSouMRQOqgvSsdhl4AWfpyBdj8HEdvHX2Lx7Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa9982a300c25f4d5fa711d8b22a72ebe9e187d70418b7ed2163e4a3726c3e04","last_reissued_at":"2026-07-05T11:02:33.306604Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:33.306604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MINIMALIST: switched-capacitor circuits for efficient in-memory computation of gated recurrent units","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.SP"],"primary_cat":"cs.AR","authors_text":"Filippo Moro, Laura Kriener, Melika Payvand, Sebastian Billaudelle, Tristan Torchet","submitted_at":"2025-05-13T14:13:41Z","abstract_excerpt":"Recurrent neural networks (RNNs) have been a long-standing candidate for processing of temporal sequence data, especially in memory-constrained systems that one may find in embedded edge computing environments. Recent advances in training paradigms have now inspired new generations of efficient RNNs. We introduce a streamlined and hardware-compatible architecture based on minimal gated recurrent units (GRUs), and an accompanying efficient mixed-signal hardware implementation of the model. The proposed design leverages switched-capacitor circuits not only for in-memory computation (IMC), but al"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08599","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.08599/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.08599","created_at":"2026-07-05T11:02:33.306677+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08599v1","created_at":"2026-07-05T11:02:33.306677+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08599","created_at":"2026-07-05T11:02:33.306677+00:00"},{"alias_kind":"pith_short_12","alias_value":"7KMYFIYAYJPU","created_at":"2026-07-05T11:02:33.306677+00:00"},{"alias_kind":"pith_short_16","alias_value":"7KMYFIYAYJPU2X5H","created_at":"2026-07-05T11:02:33.306677+00:00"},{"alias_kind":"pith_short_8","alias_value":"7KMYFIYA","created_at":"2026-07-05T11:02:33.306677+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.01829","citing_title":"mGRADE: Minimal Recurrent Gating Meets Delay Convolutions for Lightweight Sequence Modeling","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11855","citing_title":"Improving the Performance and Learning Stability of Parallelizable RNNs Designed for Ultra-Low Power Applications","ref_index":61,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P","json":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P.json","graph_json":"https://pith.science/api/pith-number/7KMYFIYAYJPU2X5HCHMLEKTS5P/graph.json","events_json":"https://pith.science/api/pith-number/7KMYFIYAYJPU2X5HCHMLEKTS5P/events.json","paper":"https://pith.science/paper/7KMYFIYA"},"agent_actions":{"view_html":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P","download_json":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P.json","view_paper":"https://pith.science/paper/7KMYFIYA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08599&json=true","fetch_graph":"https://pith.science/api/pith-number/7KMYFIYAYJPU2X5HCHMLEKTS5P/graph.json","fetch_events":"https://pith.science/api/pith-number/7KMYFIYAYJPU2X5HCHMLEKTS5P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P/action/storage_attestation","attest_author":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P/action/author_attestation","sign_citation":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P/action/citation_signature","submit_replication":"https://pith.science/pith/7KMYFIYAYJPU2X5HCHMLEKTS5P/action/replication_record"}},"created_at":"2026-07-05T11:02:33.306677+00:00","updated_at":"2026-07-05T11:02:33.306677+00:00"}