{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:D7QV3S32CHKS5RVNAHWUQDXLPK","short_pith_number":"pith:D7QV3S32","schema_version":"1.0","canonical_sha256":"1fe15dcb7a11d52ec6ad01ed480eeb7abf7bb00b5a388c37023f06f56d928b3e","source":{"kind":"arxiv","id":"2605.26035","version":1},"attestation_state":"computed","paper":{"title":"Length Generalization with Log-Depth Recurrent Units","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alessandra Russo, Charles Pert, Dalal Alrajeh","submitted_at":"2026-05-25T17:02:29Z","abstract_excerpt":"Length generalization remains a persistent challenge for neural networks: recurrent models tend to suffer from positional biases, while transformers are constrained by fixed computational depth. Regular languages provide a frequently used testbed for evaluating length generalization, as label prediction can be checked for any sequence length. We propose MLP-LDRU, a type of Log-Depth Recurrent Unit, which captures a class of associativity-biased operators designed to approximate recurrence through parallel reduction. We evaluate MLP-LDRU on 21 regular-language tasks, consisting of standard benc"},"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":"2605.26035","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-25T17:02:29Z","cross_cats_sorted":[],"title_canon_sha256":"5d8b8737cdfb43e283e52e3a0dbfffc9cad8213a745d0cdc62cc35e0793ba6b0","abstract_canon_sha256":"29709a7443ac86802046b295149edaa0494e7e6c1c03cba5c73ecebd74beabc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:05:24.335363Z","signature_b64":"DQkO6f8dQfJUOBDPd16W+/6jAwHTVuROM54d09m2SgsQDtytkgI08aAfhWWNtpWc9SSxyWMlcZa2LQELI2drDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1fe15dcb7a11d52ec6ad01ed480eeb7abf7bb00b5a388c37023f06f56d928b3e","last_reissued_at":"2026-05-26T02:05:24.334772Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:05:24.334772Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Length Generalization with Log-Depth Recurrent Units","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alessandra Russo, Charles Pert, Dalal Alrajeh","submitted_at":"2026-05-25T17:02:29Z","abstract_excerpt":"Length generalization remains a persistent challenge for neural networks: recurrent models tend to suffer from positional biases, while transformers are constrained by fixed computational depth. Regular languages provide a frequently used testbed for evaluating length generalization, as label prediction can be checked for any sequence length. We propose MLP-LDRU, a type of Log-Depth Recurrent Unit, which captures a class of associativity-biased operators designed to approximate recurrence through parallel reduction. We evaluate MLP-LDRU on 21 regular-language tasks, consisting of standard benc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26035","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/2605.26035/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":"2605.26035","created_at":"2026-05-26T02:05:24.334886+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.26035v1","created_at":"2026-05-26T02:05:24.334886+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26035","created_at":"2026-05-26T02:05:24.334886+00:00"},{"alias_kind":"pith_short_12","alias_value":"D7QV3S32CHKS","created_at":"2026-05-26T02:05:24.334886+00:00"},{"alias_kind":"pith_short_16","alias_value":"D7QV3S32CHKS5RVN","created_at":"2026-05-26T02:05:24.334886+00:00"},{"alias_kind":"pith_short_8","alias_value":"D7QV3S32","created_at":"2026-05-26T02:05:24.334886+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK","json":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK.json","graph_json":"https://pith.science/api/pith-number/D7QV3S32CHKS5RVNAHWUQDXLPK/graph.json","events_json":"https://pith.science/api/pith-number/D7QV3S32CHKS5RVNAHWUQDXLPK/events.json","paper":"https://pith.science/paper/D7QV3S32"},"agent_actions":{"view_html":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK","download_json":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK.json","view_paper":"https://pith.science/paper/D7QV3S32","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.26035&json=true","fetch_graph":"https://pith.science/api/pith-number/D7QV3S32CHKS5RVNAHWUQDXLPK/graph.json","fetch_events":"https://pith.science/api/pith-number/D7QV3S32CHKS5RVNAHWUQDXLPK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK/action/storage_attestation","attest_author":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK/action/author_attestation","sign_citation":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK/action/citation_signature","submit_replication":"https://pith.science/pith/D7QV3S32CHKS5RVNAHWUQDXLPK/action/replication_record"}},"created_at":"2026-05-26T02:05:24.334886+00:00","updated_at":"2026-05-26T02:05:24.334886+00:00"}