{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SSCTPGT3AVQ4GFLNHQ5WRCXXI3","short_pith_number":"pith:SSCTPGT3","schema_version":"1.0","canonical_sha256":"9485379a7b0561c3156d3c3b688af746c82952d44fc3d229e096c72b4addfd4d","source":{"kind":"arxiv","id":"2605.20613","version":1},"attestation_state":"computed","paper":{"title":"HRM-Text: Efficient Pretraining Beyond Scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cai Zhou, Changling Liu, Chenyu Wang, Guan Wang, Luca Scimeca, Shuai Zhen, Yasin Abbasi Yadkori, Yifei Wu, Yuhao Sun","submitted_at":"2026-05-20T01:59:50Z","abstract_excerpt":"The current pretraining paradigm for large language models relies on massive compute and internet-scale raw text, creating a significant barrier to foundational research. In contrast, biological systems demonstrate highly sample-efficient learning through multi-timescale processing, such as the functional organization of the frontoparietal loop. Taking this as inspiration, we introduce HRM-Text, which replaces standard Transformers with a Hierarchical Recurrent Model (HRM) that decouples computation into slow-evolving strategic and fast-evolving execution layers. To stabilize this deep recurre"},"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":"2605.20613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T01:59:50Z","cross_cats_sorted":[],"title_canon_sha256":"95963fff264d247dc693f292f946f618743ae3067e68aeed9ce79b28628daedb","abstract_canon_sha256":"a77aa44cbe5ef72bbd3ca8f087626d2e930952315ab0a6203547145d4733b48f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:44.616563Z","signature_b64":"7gEOf1NXLD/fJRHVteRUixKVBRumB+SelHmvmPAsm7DeBZBOSs5XaGrq4cjocoQT363M/z3dZvex8UTxQNw9DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9485379a7b0561c3156d3c3b688af746c82952d44fc3d229e096c72b4addfd4d","last_reissued_at":"2026-05-21T01:04:44.615816Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:44.615816Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HRM-Text: Efficient Pretraining Beyond Scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cai Zhou, Changling Liu, Chenyu Wang, Guan Wang, Luca Scimeca, Shuai Zhen, Yasin Abbasi Yadkori, Yifei Wu, Yuhao Sun","submitted_at":"2026-05-20T01:59:50Z","abstract_excerpt":"The current pretraining paradigm for large language models relies on massive compute and internet-scale raw text, creating a significant barrier to foundational research. In contrast, biological systems demonstrate highly sample-efficient learning through multi-timescale processing, such as the functional organization of the frontoparietal loop. Taking this as inspiration, we introduce HRM-Text, which replaces standard Transformers with a Hierarchical Recurrent Model (HRM) that decouples computation into slow-evolving strategic and fast-evolving execution layers. To stabilize this deep recurre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20613","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/2605.20613/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":"2605.20613","created_at":"2026-05-21T01:04:44.615930+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.20613v1","created_at":"2026-05-21T01:04:44.615930+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20613","created_at":"2026-05-21T01:04:44.615930+00:00"},{"alias_kind":"pith_short_12","alias_value":"SSCTPGT3AVQ4","created_at":"2026-05-21T01:04:44.615930+00:00"},{"alias_kind":"pith_short_16","alias_value":"SSCTPGT3AVQ4GFLN","created_at":"2026-05-21T01:04:44.615930+00:00"},{"alias_kind":"pith_short_8","alias_value":"SSCTPGT3","created_at":"2026-05-21T01:04:44.615930+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/SSCTPGT3AVQ4GFLNHQ5WRCXXI3","json":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3.json","graph_json":"https://pith.science/api/pith-number/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/graph.json","events_json":"https://pith.science/api/pith-number/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/events.json","paper":"https://pith.science/paper/SSCTPGT3"},"agent_actions":{"view_html":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3","download_json":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3.json","view_paper":"https://pith.science/paper/SSCTPGT3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.20613&json=true","fetch_graph":"https://pith.science/api/pith-number/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/graph.json","fetch_events":"https://pith.science/api/pith-number/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/action/storage_attestation","attest_author":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/action/author_attestation","sign_citation":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/action/citation_signature","submit_replication":"https://pith.science/pith/SSCTPGT3AVQ4GFLNHQ5WRCXXI3/action/replication_record"}},"created_at":"2026-05-21T01:04:44.615930+00:00","updated_at":"2026-05-21T01:04:44.615930+00:00"}