{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XUDA3LURBLX5YC6QMOZWTY6VHM","short_pith_number":"pith:XUDA3LUR","schema_version":"1.0","canonical_sha256":"bd060dae910aefdc0bd063b369e3d53b0804578348e5e444cee1f5e5b61de8f2","source":{"kind":"arxiv","id":"2507.08567","version":2},"attestation_state":"computed","paper":{"title":"AbbIE: Autoregressive Block-Based Iterative Encoder for Efficient Sequence Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Iacob, David O'Shea, Fernando Garcia Redondo, Lorenzo Sani, Meghdad Kurmanji, Nicholas D. Lane, Nicola Cancedda, Preslav Aleksandrov, William Shen, Xinchi Qiu","submitted_at":"2025-07-11T13:11:11Z","abstract_excerpt":"We introduce the Autoregressive Block-Based Iterative Encoder (AbbIE), a novel recursive generalization of the encoder-only Transformer architecture, which achieves better perplexity than a standard Transformer and allows for the dynamic scaling of compute resources at test time. This simple, recursive approach is a complement to scaling large language model (LLM) performance through parameter and token counts. AbbIE performs its iterations in latent space, but unlike latent reasoning models, does not require a specialized dataset or training protocol. We show that AbbIE upward generalizes (ab"},"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":"2507.08567","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-11T13:11:11Z","cross_cats_sorted":[],"title_canon_sha256":"25e099e096bb16f995fb03b43ec0c1962b70fb6c50fd753763c5dcbc7551f2a4","abstract_canon_sha256":"5f4f751ef98c112ca5a66b07e7448d2a15e292d1521fd86e142fda1ef7686dbc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:56.915671Z","signature_b64":"KIQz3ACjzpKxeXv23sQjuS4G9swLXtX+U1YvX0lMzrXuRoKwOZcaxFADLyCuoSXXnWw3kaL3dDLrwXZ1pYyaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd060dae910aefdc0bd063b369e3d53b0804578348e5e444cee1f5e5b61de8f2","last_reissued_at":"2026-07-05T11:49:56.915147Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:56.915147Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AbbIE: Autoregressive Block-Based Iterative Encoder for Efficient Sequence Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alex Iacob, David O'Shea, Fernando Garcia Redondo, Lorenzo Sani, Meghdad Kurmanji, Nicholas D. Lane, Nicola Cancedda, Preslav Aleksandrov, William Shen, Xinchi Qiu","submitted_at":"2025-07-11T13:11:11Z","abstract_excerpt":"We introduce the Autoregressive Block-Based Iterative Encoder (AbbIE), a novel recursive generalization of the encoder-only Transformer architecture, which achieves better perplexity than a standard Transformer and allows for the dynamic scaling of compute resources at test time. This simple, recursive approach is a complement to scaling large language model (LLM) performance through parameter and token counts. AbbIE performs its iterations in latent space, but unlike latent reasoning models, does not require a specialized dataset or training protocol. We show that AbbIE upward generalizes (ab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08567","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/2507.08567/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":"2507.08567","created_at":"2026-07-05T11:49:56.915210+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.08567v2","created_at":"2026-07-05T11:49:56.915210+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08567","created_at":"2026-07-05T11:49:56.915210+00:00"},{"alias_kind":"pith_short_12","alias_value":"XUDA3LURBLX5","created_at":"2026-07-05T11:49:56.915210+00:00"},{"alias_kind":"pith_short_16","alias_value":"XUDA3LURBLX5YC6Q","created_at":"2026-07-05T11:49:56.915210+00:00"},{"alias_kind":"pith_short_8","alias_value":"XUDA3LUR","created_at":"2026-07-05T11:49:56.915210+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/XUDA3LURBLX5YC6QMOZWTY6VHM","json":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM.json","graph_json":"https://pith.science/api/pith-number/XUDA3LURBLX5YC6QMOZWTY6VHM/graph.json","events_json":"https://pith.science/api/pith-number/XUDA3LURBLX5YC6QMOZWTY6VHM/events.json","paper":"https://pith.science/paper/XUDA3LUR"},"agent_actions":{"view_html":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM","download_json":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM.json","view_paper":"https://pith.science/paper/XUDA3LUR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.08567&json=true","fetch_graph":"https://pith.science/api/pith-number/XUDA3LURBLX5YC6QMOZWTY6VHM/graph.json","fetch_events":"https://pith.science/api/pith-number/XUDA3LURBLX5YC6QMOZWTY6VHM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM/action/storage_attestation","attest_author":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM/action/author_attestation","sign_citation":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM/action/citation_signature","submit_replication":"https://pith.science/pith/XUDA3LURBLX5YC6QMOZWTY6VHM/action/replication_record"}},"created_at":"2026-07-05T11:49:56.915210+00:00","updated_at":"2026-07-05T11:49:56.915210+00:00"}