{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UXDRSHU7ER4ZS2AUKLUF3I3FJE","short_pith_number":"pith:UXDRSHU7","schema_version":"1.0","canonical_sha256":"a5c7191e9f247999681452e85da36549058bd510c8d0baafd940b1d6f71ba424","source":{"kind":"arxiv","id":"2606.18524","version":1},"attestation_state":"computed","paper":{"title":"On the Residual Scaling of Looped Transformers: Stability and Transferability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bingrui Li, Ge Zhang, Jian Li, Shaowen Wang, Shen Yan, Wenhao Huang","submitted_at":"2026-06-16T22:39:13Z","abstract_excerpt":"Looped (weight-tied) Transformers apply a shared residual block $N$ times ($h \\leftarrow h + \\varepsilon\\,f(h)$, same $f$ at each step), increasing effective depth without adding parameters. Prior depth-scaling analyses prescribe $\\varepsilon = 1/\\!\\sqrt{L}$ for depth-$L$ residual networks. We show that this is insufficient for looped architectures: weight sharing makes residual updates correlated across iterations, requiring the stronger scaling $\\varepsilon = 1/N$. For multi-layer blocks ($L$ unique layers looped $N$ times), we derive a factored parameterization $\\varepsilon = \\lambda/(N\\!\\s"},"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":"2606.18524","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-16T22:39:13Z","cross_cats_sorted":[],"title_canon_sha256":"3bd51251590bd17bb2057b62501502b7b947515735e540a6a11f5caa37a52538","abstract_canon_sha256":"5b31cd9321d5abc5975d6f32fd1de78529ffedd01b8985b551c1f73e7d73099b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:11:40.053821Z","signature_b64":"MNka+8rFoX17qRK7mYzWGo3b4uYHJvUv6wmh5ah6qmop/Zc/Bb3Gb4QqMeYEfx3yJQdGVAfItq19aY4+hrMkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5c7191e9f247999681452e85da36549058bd510c8d0baafd940b1d6f71ba424","last_reissued_at":"2026-06-19T16:11:40.053430Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:11:40.053430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Residual Scaling of Looped Transformers: Stability and Transferability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bingrui Li, Ge Zhang, Jian Li, Shaowen Wang, Shen Yan, Wenhao Huang","submitted_at":"2026-06-16T22:39:13Z","abstract_excerpt":"Looped (weight-tied) Transformers apply a shared residual block $N$ times ($h \\leftarrow h + \\varepsilon\\,f(h)$, same $f$ at each step), increasing effective depth without adding parameters. Prior depth-scaling analyses prescribe $\\varepsilon = 1/\\!\\sqrt{L}$ for depth-$L$ residual networks. We show that this is insufficient for looped architectures: weight sharing makes residual updates correlated across iterations, requiring the stronger scaling $\\varepsilon = 1/N$. For multi-layer blocks ($L$ unique layers looped $N$ times), we derive a factored parameterization $\\varepsilon = \\lambda/(N\\!\\s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18524","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/2606.18524/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":"2606.18524","created_at":"2026-06-19T16:11:40.053493+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.18524v1","created_at":"2026-06-19T16:11:40.053493+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18524","created_at":"2026-06-19T16:11:40.053493+00:00"},{"alias_kind":"pith_short_12","alias_value":"UXDRSHU7ER4Z","created_at":"2026-06-19T16:11:40.053493+00:00"},{"alias_kind":"pith_short_16","alias_value":"UXDRSHU7ER4ZS2AU","created_at":"2026-06-19T16:11:40.053493+00:00"},{"alias_kind":"pith_short_8","alias_value":"UXDRSHU7","created_at":"2026-06-19T16:11:40.053493+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/UXDRSHU7ER4ZS2AUKLUF3I3FJE","json":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE.json","graph_json":"https://pith.science/api/pith-number/UXDRSHU7ER4ZS2AUKLUF3I3FJE/graph.json","events_json":"https://pith.science/api/pith-number/UXDRSHU7ER4ZS2AUKLUF3I3FJE/events.json","paper":"https://pith.science/paper/UXDRSHU7"},"agent_actions":{"view_html":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE","download_json":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE.json","view_paper":"https://pith.science/paper/UXDRSHU7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.18524&json=true","fetch_graph":"https://pith.science/api/pith-number/UXDRSHU7ER4ZS2AUKLUF3I3FJE/graph.json","fetch_events":"https://pith.science/api/pith-number/UXDRSHU7ER4ZS2AUKLUF3I3FJE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE/action/storage_attestation","attest_author":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE/action/author_attestation","sign_citation":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE/action/citation_signature","submit_replication":"https://pith.science/pith/UXDRSHU7ER4ZS2AUKLUF3I3FJE/action/replication_record"}},"created_at":"2026-06-19T16:11:40.053493+00:00","updated_at":"2026-06-19T16:11:40.053493+00:00"}