{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4TNGNSSO42CS6MJNYSHRZN64PI","short_pith_number":"pith:4TNGNSSO","schema_version":"1.0","canonical_sha256":"e4da66ca4ee6852f312dc48f1cb7dc7a0cb97ea7cdea08cdc870a589a2470e11","source":{"kind":"arxiv","id":"2501.11587","version":2},"attestation_state":"computed","paper":{"title":"Recurrent Diffusion for Large-Scale Parameter Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongwen Tang, Kai Wang, Konstantin Sch\\\"urholt, Wangbo Zhao, Yang You, Zhangyang Wang","submitted_at":"2025-01-20T16:46:26Z","abstract_excerpt":"Parameter generation has long struggled to match the scale of today large vision and language models, curbing its broader utility. In this paper, we introduce Recurrent Diffusion for Large Scale Parameter Generation (RPG), a novel framework that generates full neural network parameters up to hundreds of millions on a single GPU. Our approach first partitions a networks parameters into non-overlapping tokens, each corresponding to a distinct portion of the model. A recurrent mechanism then learns the inter token relationships, producing prototypes which serve as conditions for a diffusion proce"},"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":"2501.11587","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-20T16:46:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2f886adc35fc53c45d2022609a9e1a4a1db6a112837d8568ab52e0056c680639","abstract_canon_sha256":"e08d8fe9a23aeca62722f7f909621f3ab431f38ee38fe9bdb7fbeb65a1f7fff2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:25.418868Z","signature_b64":"6YpfkB4TR2n6wqAPYtjxs/uBQdg6yln57/8bBTG83IItOBJu5OLZMPT04INm4CAJN/wPvyPObImffL3JpHPgAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4da66ca4ee6852f312dc48f1cb7dc7a0cb97ea7cdea08cdc870a589a2470e11","last_reissued_at":"2026-07-05T10:12:25.418387Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:25.418387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Recurrent Diffusion for Large-Scale Parameter Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongwen Tang, Kai Wang, Konstantin Sch\\\"urholt, Wangbo Zhao, Yang You, Zhangyang Wang","submitted_at":"2025-01-20T16:46:26Z","abstract_excerpt":"Parameter generation has long struggled to match the scale of today large vision and language models, curbing its broader utility. In this paper, we introduce Recurrent Diffusion for Large Scale Parameter Generation (RPG), a novel framework that generates full neural network parameters up to hundreds of millions on a single GPU. Our approach first partitions a networks parameters into non-overlapping tokens, each corresponding to a distinct portion of the model. A recurrent mechanism then learns the inter token relationships, producing prototypes which serve as conditions for a diffusion proce"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.11587","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/2501.11587/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":"2501.11587","created_at":"2026-07-05T10:12:25.418446+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.11587v2","created_at":"2026-07-05T10:12:25.418446+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.11587","created_at":"2026-07-05T10:12:25.418446+00:00"},{"alias_kind":"pith_short_12","alias_value":"4TNGNSSO42CS","created_at":"2026-07-05T10:12:25.418446+00:00"},{"alias_kind":"pith_short_16","alias_value":"4TNGNSSO42CS6MJN","created_at":"2026-07-05T10:12:25.418446+00:00"},{"alias_kind":"pith_short_8","alias_value":"4TNGNSSO","created_at":"2026-07-05T10:12:25.418446+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07217","citing_title":"Robotic Policy Adaptation via Weight-Space Meta-Learning","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2505.13919","citing_title":"Generative Adaptation of Dynamics to Environmental Shifts via Weight-space Diffusion","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20189","citing_title":"SOLAR: A Self-Optimizing Open-Ended Autonomous Agent for Lifelong Learning and Continual Adaptation","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18474","citing_title":"Prompt2Fingerprint: Plug-and-Play LLM Fingerprinting via Text-to-Weight Generation","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI","json":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI.json","graph_json":"https://pith.science/api/pith-number/4TNGNSSO42CS6MJNYSHRZN64PI/graph.json","events_json":"https://pith.science/api/pith-number/4TNGNSSO42CS6MJNYSHRZN64PI/events.json","paper":"https://pith.science/paper/4TNGNSSO"},"agent_actions":{"view_html":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI","download_json":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI.json","view_paper":"https://pith.science/paper/4TNGNSSO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.11587&json=true","fetch_graph":"https://pith.science/api/pith-number/4TNGNSSO42CS6MJNYSHRZN64PI/graph.json","fetch_events":"https://pith.science/api/pith-number/4TNGNSSO42CS6MJNYSHRZN64PI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI/action/storage_attestation","attest_author":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI/action/author_attestation","sign_citation":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI/action/citation_signature","submit_replication":"https://pith.science/pith/4TNGNSSO42CS6MJNYSHRZN64PI/action/replication_record"}},"created_at":"2026-07-05T10:12:25.418446+00:00","updated_at":"2026-07-05T10:12:25.418446+00:00"}