{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:PMKAOJCGFQOYGXLBVTZI7IPJHJ","short_pith_number":"pith:PMKAOJCG","canonical_record":{"source":{"id":"2306.16788","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-29T08:49:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2cf7478643fc80cad9f85d898f9863238e515f2d6e296d3425a2416cfcc9748b","abstract_canon_sha256":"0824bf64c4f0d21d565784484493bd759fb3b6727f734ce95e0c3db43289ce60"},"schema_version":"1.0"},"canonical_sha256":"7b140724462c1d835d61acf28fa1e93a5b51ca7c63de453de5c60f5265f3edc3","source":{"kind":"arxiv","id":"2306.16788","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.16788","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"arxiv_version","alias_value":"2306.16788v3","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.16788","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"pith_short_12","alias_value":"PMKAOJCGFQOY","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"pith_short_16","alias_value":"PMKAOJCGFQOYGXLB","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"pith_short_8","alias_value":"PMKAOJCG","created_at":"2026-07-05T07:59:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:PMKAOJCGFQOYGXLBVTZI7IPJHJ","target":"record","payload":{"canonical_record":{"source":{"id":"2306.16788","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-29T08:49:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2cf7478643fc80cad9f85d898f9863238e515f2d6e296d3425a2416cfcc9748b","abstract_canon_sha256":"0824bf64c4f0d21d565784484493bd759fb3b6727f734ce95e0c3db43289ce60"},"schema_version":"1.0"},"canonical_sha256":"7b140724462c1d835d61acf28fa1e93a5b51ca7c63de453de5c60f5265f3edc3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:59:51.328687Z","signature_b64":"m8RHgPFyu7YQPaaLhFUBypdwh57OzFIIlF4DRn7YpnymPxndPCPzSe2/Nc2uU+3aYJwqg0ypOA9rVu+mp35kAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b140724462c1d835d61acf28fa1e93a5b51ca7c63de453de5c60f5265f3edc3","last_reissued_at":"2026-07-05T07:59:51.328305Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:59:51.328305Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.16788","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:59:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l48zbEU4z6tEgTbL1Gx946dZYrn5VymSiM/RzP0MUcNmSlOSF2TiuaRZWNzNpmcFdRGZm0SpEGs9LVdoZ/wSCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:59:05.313233Z"},"content_sha256":"1f3b832317de746d7579b23ae15983a79891ce2d5a31335f826607aae976d107","schema_version":"1.0","event_id":"sha256:1f3b832317de746d7579b23ae15983a79891ce2d5a31335f826607aae976d107"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:PMKAOJCGFQOYGXLBVTZI7IPJHJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sparse Model Soups: A Recipe for Improved Pruning via Model Averaging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christoph Spiegel, Max Zimmer, Sebastian Pokutta","submitted_at":"2023-06-29T08:49:41Z","abstract_excerpt":"Neural networks can be significantly compressed by pruning, yielding sparse models with reduced storage and computational demands while preserving predictive performance. Model soups (Wortsman et al., 2022) enhance generalization and out-of-distribution (OOD) performance by averaging the parameters of multiple models into a single one, without increasing inference time. However, achieving both sparsity and parameter averaging is challenging as averaging arbitrary sparse models reduces the overall sparsity due to differing sparse connectivities. This work addresses these challenges by demonstra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.16788","kind":"arxiv","version":3},"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/2306.16788/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:59:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FXYcXX59ACU6T89creqwTt7YN4QWwjtkYUDztQL+ISc7RNA4SxVkjnuGsRQyF2IfOUkDtAQOeFbJ6vcdGow+Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:59:05.313629Z"},"content_sha256":"4ae46e21aa738a15f7528bf8845e6904aa874f32cbd6846d235d39127f41292d","schema_version":"1.0","event_id":"sha256:4ae46e21aa738a15f7528bf8845e6904aa874f32cbd6846d235d39127f41292d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ/bundle.json","state_url":"https://pith.science/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T15:59:05Z","links":{"resolver":"https://pith.science/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ","bundle":"https://pith.science/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ/bundle.json","state":"https://pith.science/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PMKAOJCGFQOYGXLBVTZI7IPJHJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:PMKAOJCGFQOYGXLBVTZI7IPJHJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0824bf64c4f0d21d565784484493bd759fb3b6727f734ce95e0c3db43289ce60","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-29T08:49:41Z","title_canon_sha256":"2cf7478643fc80cad9f85d898f9863238e515f2d6e296d3425a2416cfcc9748b"},"schema_version":"1.0","source":{"id":"2306.16788","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.16788","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"arxiv_version","alias_value":"2306.16788v3","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.16788","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"pith_short_12","alias_value":"PMKAOJCGFQOY","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"pith_short_16","alias_value":"PMKAOJCGFQOYGXLB","created_at":"2026-07-05T07:59:51Z"},{"alias_kind":"pith_short_8","alias_value":"PMKAOJCG","created_at":"2026-07-05T07:59:51Z"}],"graph_snapshots":[{"event_id":"sha256:4ae46e21aa738a15f7528bf8845e6904aa874f32cbd6846d235d39127f41292d","target":"graph","created_at":"2026-07-05T07:59:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2306.16788/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural networks can be significantly compressed by pruning, yielding sparse models with reduced storage and computational demands while preserving predictive performance. Model soups (Wortsman et al., 2022) enhance generalization and out-of-distribution (OOD) performance by averaging the parameters of multiple models into a single one, without increasing inference time. However, achieving both sparsity and parameter averaging is challenging as averaging arbitrary sparse models reduces the overall sparsity due to differing sparse connectivities. This work addresses these challenges by demonstra","authors_text":"Christoph Spiegel, Max Zimmer, Sebastian Pokutta","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-29T08:49:41Z","title":"Sparse Model Soups: A Recipe for Improved Pruning via Model Averaging"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.16788","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1f3b832317de746d7579b23ae15983a79891ce2d5a31335f826607aae976d107","target":"record","created_at":"2026-07-05T07:59:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0824bf64c4f0d21d565784484493bd759fb3b6727f734ce95e0c3db43289ce60","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-29T08:49:41Z","title_canon_sha256":"2cf7478643fc80cad9f85d898f9863238e515f2d6e296d3425a2416cfcc9748b"},"schema_version":"1.0","source":{"id":"2306.16788","kind":"arxiv","version":3}},"canonical_sha256":"7b140724462c1d835d61acf28fa1e93a5b51ca7c63de453de5c60f5265f3edc3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7b140724462c1d835d61acf28fa1e93a5b51ca7c63de453de5c60f5265f3edc3","first_computed_at":"2026-07-05T07:59:51.328305Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:59:51.328305Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"m8RHgPFyu7YQPaaLhFUBypdwh57OzFIIlF4DRn7YpnymPxndPCPzSe2/Nc2uU+3aYJwqg0ypOA9rVu+mp35kAA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:59:51.328687Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.16788","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1f3b832317de746d7579b23ae15983a79891ce2d5a31335f826607aae976d107","sha256:4ae46e21aa738a15f7528bf8845e6904aa874f32cbd6846d235d39127f41292d"],"state_sha256":"ae19214d7bff9df7f098f628f26c6523e20f7e1e8014239aee50d4c483869ea6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"70xaVwKiIBTegRf+tPrawj+0SmYaZcQjnlQDJlCk80v+lyx+yVPabqH+2iAlPbfItAi65NQqcRwUS1IUhUSICw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T15:59:05.315623Z","bundle_sha256":"7fc16c76afd61fe9a66a734f815f34d9ab495caa8c440ae7afa50c261ecbb140"}}