{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:NOCHXPETIKLAX4ENFVIBEMZ63M","short_pith_number":"pith:NOCHXPET","canonical_record":{"source":{"id":"1702.08597","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-02-28T01:37:57Z","cross_cats_sorted":[],"title_canon_sha256":"c7f8edef4ede2bc732e7c9eabc4dd89eeb6632686044e42d8ec57129fdfc0e54","abstract_canon_sha256":"6536202643d7568f83610d688b9e9048c76f2bc9a436bdb5d0c8d4663a48d72f"},"schema_version":"1.0"},"canonical_sha256":"6b847bbc9342960bf08d2d5012333edb237ffb7ef230bcd0c3c01507b2abb6e2","source":{"kind":"arxiv","id":"1702.08597","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1702.08597","created_at":"2026-05-18T00:32:55Z"},{"alias_kind":"arxiv_version","alias_value":"1702.08597v2","created_at":"2026-05-18T00:32:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.08597","created_at":"2026-05-18T00:32:55Z"},{"alias_kind":"pith_short_12","alias_value":"NOCHXPETIKLA","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_16","alias_value":"NOCHXPETIKLAX4EN","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_8","alias_value":"NOCHXPET","created_at":"2026-05-18T12:31:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:NOCHXPETIKLAX4ENFVIBEMZ63M","target":"record","payload":{"canonical_record":{"source":{"id":"1702.08597","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-02-28T01:37:57Z","cross_cats_sorted":[],"title_canon_sha256":"c7f8edef4ede2bc732e7c9eabc4dd89eeb6632686044e42d8ec57129fdfc0e54","abstract_canon_sha256":"6536202643d7568f83610d688b9e9048c76f2bc9a436bdb5d0c8d4663a48d72f"},"schema_version":"1.0"},"canonical_sha256":"6b847bbc9342960bf08d2d5012333edb237ffb7ef230bcd0c3c01507b2abb6e2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:32:55.329428Z","signature_b64":"lyKGUYagyRQCZiMXPeQlmkSXroWoMUhjfMKBnx2i1D4q/T1/gMrKOY1SRInp3+rTBKaAKTcma3ek5FCcBI/RAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b847bbc9342960bf08d2d5012333edb237ffb7ef230bcd0c3c01507b2abb6e2","last_reissued_at":"2026-05-18T00:32:55.328702Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:32:55.328702Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1702.08597","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:32:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d5oNm2VFvXk6yYxfsTlcdZh6dAhpl/PU1+837msdeN5Nwvplncvi4FL6lDTQkEwMHC6qKE5T43S704+t5ev1AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T10:27:18.646616Z"},"content_sha256":"7f7ff3ccca87ecbf5c8c6e09b63421037e667bba146cc3308e1957dcb6eb4199","schema_version":"1.0","event_id":"sha256:7f7ff3ccca87ecbf5c8c6e09b63421037e667bba146cc3308e1957dcb6eb4199"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:NOCHXPETIKLAX4ENFVIBEMZ63M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Enabling Sparse Winograd Convolution by Native Pruning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jongsoo Park, Ping Tak Peter Tang, Sheng Li","submitted_at":"2017-02-28T01:37:57Z","abstract_excerpt":"Sparse methods and the use of Winograd convolutions are two orthogonal approaches, each of which significantly accelerates convolution computations in modern CNNs. Sparse Winograd merges these two and thus has the potential to offer a combined performance benefit. Nevertheless, training convolution layers so that the resulting Winograd kernels are sparse has not hitherto been very successful. By introducing a Winograd layer in place of a standard convolution layer, we can learn and prune Winograd coefficients \"natively\" and obtain sparsity level beyond 90% with only 0.1% accuracy loss with Ale"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.08597","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":""},"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":"2026-05-18T00:32:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k2W6j0IDK9pHmJnNvVJ8SgOFaEdPzxw+RRliTGnCZczALSRA7b4hXP3RuD//piNp3PBCuQe6XdcT0C591lTJDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T10:27:18.647024Z"},"content_sha256":"2d03581006e3eebd6288128bae097ef444356840cc149b966ad295530f81c851","schema_version":"1.0","event_id":"sha256:2d03581006e3eebd6288128bae097ef444356840cc149b966ad295530f81c851"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NOCHXPETIKLAX4ENFVIBEMZ63M/bundle.json","state_url":"https://pith.science/pith/NOCHXPETIKLAX4ENFVIBEMZ63M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NOCHXPETIKLAX4ENFVIBEMZ63M/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":"2026-05-21T10:27:18Z","links":{"resolver":"https://pith.science/pith/NOCHXPETIKLAX4ENFVIBEMZ63M","bundle":"https://pith.science/pith/NOCHXPETIKLAX4ENFVIBEMZ63M/bundle.json","state":"https://pith.science/pith/NOCHXPETIKLAX4ENFVIBEMZ63M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NOCHXPETIKLAX4ENFVIBEMZ63M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:NOCHXPETIKLAX4ENFVIBEMZ63M","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":"6536202643d7568f83610d688b9e9048c76f2bc9a436bdb5d0c8d4663a48d72f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-02-28T01:37:57Z","title_canon_sha256":"c7f8edef4ede2bc732e7c9eabc4dd89eeb6632686044e42d8ec57129fdfc0e54"},"schema_version":"1.0","source":{"id":"1702.08597","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1702.08597","created_at":"2026-05-18T00:32:55Z"},{"alias_kind":"arxiv_version","alias_value":"1702.08597v2","created_at":"2026-05-18T00:32:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.08597","created_at":"2026-05-18T00:32:55Z"},{"alias_kind":"pith_short_12","alias_value":"NOCHXPETIKLA","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_16","alias_value":"NOCHXPETIKLAX4EN","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_8","alias_value":"NOCHXPET","created_at":"2026-05-18T12:31:34Z"}],"graph_snapshots":[{"event_id":"sha256:2d03581006e3eebd6288128bae097ef444356840cc149b966ad295530f81c851","target":"graph","created_at":"2026-05-18T00:32:55Z","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"},"paper":{"abstract_excerpt":"Sparse methods and the use of Winograd convolutions are two orthogonal approaches, each of which significantly accelerates convolution computations in modern CNNs. Sparse Winograd merges these two and thus has the potential to offer a combined performance benefit. Nevertheless, training convolution layers so that the resulting Winograd kernels are sparse has not hitherto been very successful. By introducing a Winograd layer in place of a standard convolution layer, we can learn and prune Winograd coefficients \"natively\" and obtain sparsity level beyond 90% with only 0.1% accuracy loss with Ale","authors_text":"Jongsoo Park, Ping Tak Peter Tang, Sheng Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-02-28T01:37:57Z","title":"Enabling Sparse Winograd Convolution by Native Pruning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.08597","kind":"arxiv","version":2},"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":"sha256:7f7ff3ccca87ecbf5c8c6e09b63421037e667bba146cc3308e1957dcb6eb4199","target":"record","created_at":"2026-05-18T00:32:55Z","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":"6536202643d7568f83610d688b9e9048c76f2bc9a436bdb5d0c8d4663a48d72f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-02-28T01:37:57Z","title_canon_sha256":"c7f8edef4ede2bc732e7c9eabc4dd89eeb6632686044e42d8ec57129fdfc0e54"},"schema_version":"1.0","source":{"id":"1702.08597","kind":"arxiv","version":2}},"canonical_sha256":"6b847bbc9342960bf08d2d5012333edb237ffb7ef230bcd0c3c01507b2abb6e2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6b847bbc9342960bf08d2d5012333edb237ffb7ef230bcd0c3c01507b2abb6e2","first_computed_at":"2026-05-18T00:32:55.328702Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:32:55.328702Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lyKGUYagyRQCZiMXPeQlmkSXroWoMUhjfMKBnx2i1D4q/T1/gMrKOY1SRInp3+rTBKaAKTcma3ek5FCcBI/RAA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:32:55.329428Z","signed_message":"canonical_sha256_bytes"},"source_id":"1702.08597","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7f7ff3ccca87ecbf5c8c6e09b63421037e667bba146cc3308e1957dcb6eb4199","sha256:2d03581006e3eebd6288128bae097ef444356840cc149b966ad295530f81c851"],"state_sha256":"bf9488e7ba0dd56e9b2ea69ecda79ebfb0336e6d3b2aba3a9c74b1adead9d591"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lXgsyvyLdXnNUhFV9PZA1Y02Hyw75BuaAzj9P3ReLANu2EDqblAzP3l+tPEMzem01XynXb7/u2Pf53LV7PE1Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T10:27:18.649828Z","bundle_sha256":"bbb8336f08ca722cdb2a7198ab07a70470aab7e90e77b1eec19b5e4ad2e63c15"}}