{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6PYY7K2XSOIWFCTVAW7YP2RHHG","short_pith_number":"pith:6PYY7K2X","schema_version":"1.0","canonical_sha256":"f3f18fab579391628a7505bf87ea2739adbf2b0076b89b47ae9e8ff40d6f9232","source":{"kind":"arxiv","id":"2602.18333","version":2},"attestation_state":"computed","paper":{"title":"On the \"Induction Bias\" in Sequence Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Micha\\\"el Defferrard, M.Reza Ebrahimi, Roland Memisevic, Sunny Panchal","submitted_at":"2026-02-20T16:39:07Z","abstract_excerpt":"Despite the remarkable practical success of transformer-based language models, recent work has raised concerns about their ability to perform state tracking. In particular, a growing body of literature has shown this limitation primarily through failures in out-of-distribution (OOD) generalization, such as length extrapolation. In this work, we shift attention to the in-distribution implications of these limitations. We conduct a large-scale experimental study of the data efficiency of transformers and recurrent neural networks (RNNs) across multiple supervision regimes. We find that the amoun"},"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":"2602.18333","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-20T16:39:07Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"02dd2a2031345e0b3cf21daa44aa38933eefc7e3edd5073b4f0ecbf357a540cc","abstract_canon_sha256":"b9e6d0039fac9b603e76b2bc6468cbf458cfead1f6622ecae7c20843585a13cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:34.882805Z","signature_b64":"1axDVm8mkeKvTB/YKh4NQosVePra929o3POumOUwhx/aoFFCpIIzinvY5ZHa6IUhgwcy0Ve7BPDdEV8MAV7eAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f3f18fab579391628a7505bf87ea2739adbf2b0076b89b47ae9e8ff40d6f9232","last_reissued_at":"2026-06-01T01:02:34.881637Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:34.881637Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the \"Induction Bias\" in Sequence Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Micha\\\"el Defferrard, M.Reza Ebrahimi, Roland Memisevic, Sunny Panchal","submitted_at":"2026-02-20T16:39:07Z","abstract_excerpt":"Despite the remarkable practical success of transformer-based language models, recent work has raised concerns about their ability to perform state tracking. In particular, a growing body of literature has shown this limitation primarily through failures in out-of-distribution (OOD) generalization, such as length extrapolation. In this work, we shift attention to the in-distribution implications of these limitations. We conduct a large-scale experimental study of the data efficiency of transformers and recurrent neural networks (RNNs) across multiple supervision regimes. We find that the amoun"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.18333","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/2602.18333/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":"2602.18333","created_at":"2026-06-01T01:02:34.881754+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.18333v2","created_at":"2026-06-01T01:02:34.881754+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.18333","created_at":"2026-06-01T01:02:34.881754+00:00"},{"alias_kind":"pith_short_12","alias_value":"6PYY7K2XSOIW","created_at":"2026-06-01T01:02:34.881754+00:00"},{"alias_kind":"pith_short_16","alias_value":"6PYY7K2XSOIWFCTV","created_at":"2026-06-01T01:02:34.881754+00:00"},{"alias_kind":"pith_short_8","alias_value":"6PYY7K2X","created_at":"2026-06-01T01:02:34.881754+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.08539","citing_title":"Continuity Laws for Sequential Models","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG","json":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG.json","graph_json":"https://pith.science/api/pith-number/6PYY7K2XSOIWFCTVAW7YP2RHHG/graph.json","events_json":"https://pith.science/api/pith-number/6PYY7K2XSOIWFCTVAW7YP2RHHG/events.json","paper":"https://pith.science/paper/6PYY7K2X"},"agent_actions":{"view_html":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG","download_json":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG.json","view_paper":"https://pith.science/paper/6PYY7K2X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.18333&json=true","fetch_graph":"https://pith.science/api/pith-number/6PYY7K2XSOIWFCTVAW7YP2RHHG/graph.json","fetch_events":"https://pith.science/api/pith-number/6PYY7K2XSOIWFCTVAW7YP2RHHG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG/action/storage_attestation","attest_author":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG/action/author_attestation","sign_citation":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG/action/citation_signature","submit_replication":"https://pith.science/pith/6PYY7K2XSOIWFCTVAW7YP2RHHG/action/replication_record"}},"created_at":"2026-06-01T01:02:34.881754+00:00","updated_at":"2026-06-01T01:02:34.881754+00:00"}