{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JDXVIWSW5JZDZSE2WDFFASEPV2","short_pith_number":"pith:JDXVIWSW","schema_version":"1.0","canonical_sha256":"48ef545a56ea723cc89ab0ca50488fae86d69874759a8dc19a9043ae1164f973","source":{"kind":"arxiv","id":"2402.04012","version":2},"attestation_state":"computed","paper":{"title":"Quantized Approximately Orthogonal Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.SP","math.ST","stat.TH"],"primary_cat":"cs.NE","authors_text":"Armand Foucault (IMT), Fran\\c{c}ois Malgouyres (IMT), Franck Mamalet (UT)","submitted_at":"2024-02-05T09:59:57Z","abstract_excerpt":"In recent years, Orthogonal Recurrent Neural Networks (ORNNs) have gained popularity due to their ability to manage tasks involving long-term dependencies, such as the copy-task, and their linear complexity. However, existing ORNNs utilize full precision weights and activations, which prevents their deployment on compact devices.In this paper, we explore the quantization of the weight matrices in ORNNs, leading to Quantized approximately Orthogonal RNNs (QORNNs). The construction of such networks remained an open problem, acknowledged for its inherent instability. We propose and investigate tw"},"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":"2402.04012","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2024-02-05T09:59:57Z","cross_cats_sorted":["cs.AI","cs.LG","eess.SP","math.ST","stat.TH"],"title_canon_sha256":"f6b7beb59f04ad24859a51def13b00dd45b73844a9466be554540a07489dc337","abstract_canon_sha256":"3fe9010d11a866b9ed81c3c385235bb48503043bb49cad81fefc939e2ac0ff64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:21.743179Z","signature_b64":"/hOnjvm7BO226WclGzXuMVQUKLNbAedNZQgpncxmH+XiTTNJfzZX1YxsvkhMAgrs0dF63vZeipFIvLG8aQBOBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48ef545a56ea723cc89ab0ca50488fae86d69874759a8dc19a9043ae1164f973","last_reissued_at":"2026-07-05T08:29:21.742714Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:21.742714Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantized Approximately Orthogonal Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.SP","math.ST","stat.TH"],"primary_cat":"cs.NE","authors_text":"Armand Foucault (IMT), Fran\\c{c}ois Malgouyres (IMT), Franck Mamalet (UT)","submitted_at":"2024-02-05T09:59:57Z","abstract_excerpt":"In recent years, Orthogonal Recurrent Neural Networks (ORNNs) have gained popularity due to their ability to manage tasks involving long-term dependencies, such as the copy-task, and their linear complexity. However, existing ORNNs utilize full precision weights and activations, which prevents their deployment on compact devices.In this paper, we explore the quantization of the weight matrices in ORNNs, leading to Quantized approximately Orthogonal RNNs (QORNNs). The construction of such networks remained an open problem, acknowledged for its inherent instability. We propose and investigate tw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.04012","kind":"arxiv","version":2},"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/2402.04012/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":"2402.04012","created_at":"2026-07-05T08:29:21.742771+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.04012v2","created_at":"2026-07-05T08:29:21.742771+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.04012","created_at":"2026-07-05T08:29:21.742771+00:00"},{"alias_kind":"pith_short_12","alias_value":"JDXVIWSW5JZD","created_at":"2026-07-05T08:29:21.742771+00:00"},{"alias_kind":"pith_short_16","alias_value":"JDXVIWSW5JZDZSE2","created_at":"2026-07-05T08:29:21.742771+00:00"},{"alias_kind":"pith_short_8","alias_value":"JDXVIWSW","created_at":"2026-07-05T08:29:21.742771+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00047","citing_title":"HadamRNN: Binary and Sparse Ternary Orthogonal RNNs","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2","json":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2.json","graph_json":"https://pith.science/api/pith-number/JDXVIWSW5JZDZSE2WDFFASEPV2/graph.json","events_json":"https://pith.science/api/pith-number/JDXVIWSW5JZDZSE2WDFFASEPV2/events.json","paper":"https://pith.science/paper/JDXVIWSW"},"agent_actions":{"view_html":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2","download_json":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2.json","view_paper":"https://pith.science/paper/JDXVIWSW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.04012&json=true","fetch_graph":"https://pith.science/api/pith-number/JDXVIWSW5JZDZSE2WDFFASEPV2/graph.json","fetch_events":"https://pith.science/api/pith-number/JDXVIWSW5JZDZSE2WDFFASEPV2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2/action/storage_attestation","attest_author":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2/action/author_attestation","sign_citation":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2/action/citation_signature","submit_replication":"https://pith.science/pith/JDXVIWSW5JZDZSE2WDFFASEPV2/action/replication_record"}},"created_at":"2026-07-05T08:29:21.742771+00:00","updated_at":"2026-07-05T08:29:21.742771+00:00"}