{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:P4OCF7WMJLDLRWIGAZYGCCWEFX","short_pith_number":"pith:P4OCF7WM","schema_version":"1.0","canonical_sha256":"7f1c22fecc4ac6b8d9060670610ac42dedea86a076a6349f13d4b61685910280","source":{"kind":"arxiv","id":"2207.10951","version":1},"attestation_state":"computed","paper":{"title":"Hyper-Representations for Pre-Training and Transfer Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Boris Knyazev, Damian Borth, Konstantin Sch\\\"urholt, Xavier Gir\\'o-i-Nieto","submitted_at":"2022-07-22T09:01:21Z","abstract_excerpt":"Learning representations of neural network weights given a model zoo is an emerging and challenging area with many potential applications from model inspection, to neural architecture search or knowledge distillation. Recently, an autoencoder trained on a model zoo was able to learn a hyper-representation, which captures intrinsic and extrinsic properties of the models in the zoo. In this work, we extend hyper-representations for generative use to sample new model weights as pre-training. We propose layer-wise loss normalization which we demonstrate is key to generate high-performing models an"},"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":"2207.10951","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-07-22T09:01:21Z","cross_cats_sorted":[],"title_canon_sha256":"827bf9fada24aab8613271e79b5ede633c593b3db9079f60bfe9dd36d7115c28","abstract_canon_sha256":"7db6d8fa14d6b7504a3a63f8ea98ec013f2951065bd465b0ece3312a3ac07343"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:42:41.834771Z","signature_b64":"NCoNUy14La5JTZ9Ie+mevwkk+xGbB2aUfQuMxLB1vYFQTCmt+YElQ/G5ChJSQ66oygB1Fs90GsRIQu5JQEK5AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f1c22fecc4ac6b8d9060670610ac42dedea86a076a6349f13d4b61685910280","last_reissued_at":"2026-07-05T04:42:41.834376Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:42:41.834376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hyper-Representations for Pre-Training and Transfer Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Boris Knyazev, Damian Borth, Konstantin Sch\\\"urholt, Xavier Gir\\'o-i-Nieto","submitted_at":"2022-07-22T09:01:21Z","abstract_excerpt":"Learning representations of neural network weights given a model zoo is an emerging and challenging area with many potential applications from model inspection, to neural architecture search or knowledge distillation. Recently, an autoencoder trained on a model zoo was able to learn a hyper-representation, which captures intrinsic and extrinsic properties of the models in the zoo. In this work, we extend hyper-representations for generative use to sample new model weights as pre-training. We propose layer-wise loss normalization which we demonstrate is key to generate high-performing models an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.10951","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/2207.10951/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":"2207.10951","created_at":"2026-07-05T04:42:41.834435+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.10951v1","created_at":"2026-07-05T04:42:41.834435+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.10951","created_at":"2026-07-05T04:42:41.834435+00:00"},{"alias_kind":"pith_short_12","alias_value":"P4OCF7WMJLDL","created_at":"2026-07-05T04:42:41.834435+00:00"},{"alias_kind":"pith_short_16","alias_value":"P4OCF7WMJLDLRWIG","created_at":"2026-07-05T04:42:41.834435+00:00"},{"alias_kind":"pith_short_8","alias_value":"P4OCF7WM","created_at":"2026-07-05T04:42:41.834435+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.11587","citing_title":"Recurrent Diffusion for Large-Scale Parameter Generation","ref_index":51,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX","json":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX.json","graph_json":"https://pith.science/api/pith-number/P4OCF7WMJLDLRWIGAZYGCCWEFX/graph.json","events_json":"https://pith.science/api/pith-number/P4OCF7WMJLDLRWIGAZYGCCWEFX/events.json","paper":"https://pith.science/paper/P4OCF7WM"},"agent_actions":{"view_html":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX","download_json":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX.json","view_paper":"https://pith.science/paper/P4OCF7WM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.10951&json=true","fetch_graph":"https://pith.science/api/pith-number/P4OCF7WMJLDLRWIGAZYGCCWEFX/graph.json","fetch_events":"https://pith.science/api/pith-number/P4OCF7WMJLDLRWIGAZYGCCWEFX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX/action/storage_attestation","attest_author":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX/action/author_attestation","sign_citation":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX/action/citation_signature","submit_replication":"https://pith.science/pith/P4OCF7WMJLDLRWIGAZYGCCWEFX/action/replication_record"}},"created_at":"2026-07-05T04:42:41.834435+00:00","updated_at":"2026-07-05T04:42:41.834435+00:00"}