{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:BUPG5XYQCLACMSOVP5OMJY37EW","short_pith_number":"pith:BUPG5XYQ","canonical_record":{"source":{"id":"2006.07253","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-06-12T15:07:08Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"617aefe6112e32c77041558f6589e26b0db2573cfed3654846799fae4f29aad9","abstract_canon_sha256":"1d7d7d5fdd61d34079afe3d3c57d50a16439501c1cfeefe54c49d5d198a58577"},"schema_version":"1.0"},"canonical_sha256":"0d1e6edf1012c02649d57f5cc4e37f25be4e547f576a37732da931eaf9e12176","source":{"kind":"arxiv","id":"2006.07253","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.07253","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"arxiv_version","alias_value":"2006.07253v1","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.07253","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"pith_short_12","alias_value":"BUPG5XYQCLAC","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"pith_short_16","alias_value":"BUPG5XYQCLACMSOV","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"pith_short_8","alias_value":"BUPG5XYQ","created_at":"2026-07-05T01:09:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:BUPG5XYQCLACMSOVP5OMJY37EW","target":"record","payload":{"canonical_record":{"source":{"id":"2006.07253","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-06-12T15:07:08Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"617aefe6112e32c77041558f6589e26b0db2573cfed3654846799fae4f29aad9","abstract_canon_sha256":"1d7d7d5fdd61d34079afe3d3c57d50a16439501c1cfeefe54c49d5d198a58577"},"schema_version":"1.0"},"canonical_sha256":"0d1e6edf1012c02649d57f5cc4e37f25be4e547f576a37732da931eaf9e12176","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:09:53.174893Z","signature_b64":"TBHLxNJUDXbnH+tzrsY8OnpcKjJSaMD0afsWanfjDnIqubS6hErOOAK69Kwj9e1O3pi1YKG8lB3Hb7HtMqv/Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d1e6edf1012c02649d57f5cc4e37f25be4e547f576a37732da931eaf9e12176","last_reissued_at":"2026-07-05T01:09:53.174338Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:09:53.174338Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2006.07253","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T01:09:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h+hvHpktsqx9oJVjSGYcf6CSew4sQ+EBAyw2BeNZlBv4akG8aoqt3y1q4mIKxstYQS51wMtBk7/B6Kz11BlBAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T03:05:23.349197Z"},"content_sha256":"9785ecf1aa8907673e664a06e19c401d1448bceecf7180437977d413d384f55a","schema_version":"1.0","event_id":"sha256:9785ecf1aa8907673e664a06e19c401d1448bceecf7180437977d413d384f55a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:BUPG5XYQCLACMSOVP5OMJY37EW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Dynamic Model Pruning with Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniil Dmitriev, Luis Barba, Martin Jaggi, Sebastian U. Stich, Tao Lin","submitted_at":"2020-06-12T15:07:08Z","abstract_excerpt":"Deep neural networks often have millions of parameters. This can hinder their deployment to low-end devices, not only due to high memory requirements but also because of increased latency at inference. We propose a novel model compression method that generates a sparse trained model without additional overhead: by allowing (i) dynamic allocation of the sparsity pattern and (ii) incorporating feedback signal to reactivate prematurely pruned weights we obtain a performant sparse model in one single training pass (retraining is not needed, but can further improve the performance). We evaluate our"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.07253","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/2006.07253/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":"2026-07-05T01:09:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DuAJ98X5CbNm4lE1weZISkxTZg3ItMJq71wP3OfZgOzerobqzVvtLOwS/aL6hF/ot6lADftjRuKRszBQZ2B7DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T03:05:23.349709Z"},"content_sha256":"e2793f1cd9c290460ef6a2f0ea2bc5aae82e5a11ea9781b59806b29571d90c5f","schema_version":"1.0","event_id":"sha256:e2793f1cd9c290460ef6a2f0ea2bc5aae82e5a11ea9781b59806b29571d90c5f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BUPG5XYQCLACMSOVP5OMJY37EW/bundle.json","state_url":"https://pith.science/pith/BUPG5XYQCLACMSOVP5OMJY37EW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BUPG5XYQCLACMSOVP5OMJY37EW/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":"2026-08-05T03:05:23Z","links":{"resolver":"https://pith.science/pith/BUPG5XYQCLACMSOVP5OMJY37EW","bundle":"https://pith.science/pith/BUPG5XYQCLACMSOVP5OMJY37EW/bundle.json","state":"https://pith.science/pith/BUPG5XYQCLACMSOVP5OMJY37EW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BUPG5XYQCLACMSOVP5OMJY37EW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:BUPG5XYQCLACMSOVP5OMJY37EW","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":"1d7d7d5fdd61d34079afe3d3c57d50a16439501c1cfeefe54c49d5d198a58577","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-06-12T15:07:08Z","title_canon_sha256":"617aefe6112e32c77041558f6589e26b0db2573cfed3654846799fae4f29aad9"},"schema_version":"1.0","source":{"id":"2006.07253","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.07253","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"arxiv_version","alias_value":"2006.07253v1","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.07253","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"pith_short_12","alias_value":"BUPG5XYQCLAC","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"pith_short_16","alias_value":"BUPG5XYQCLACMSOV","created_at":"2026-07-05T01:09:53Z"},{"alias_kind":"pith_short_8","alias_value":"BUPG5XYQ","created_at":"2026-07-05T01:09:53Z"}],"graph_snapshots":[{"event_id":"sha256:e2793f1cd9c290460ef6a2f0ea2bc5aae82e5a11ea9781b59806b29571d90c5f","target":"graph","created_at":"2026-07-05T01:09:53Z","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/2006.07253/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep neural networks often have millions of parameters. This can hinder their deployment to low-end devices, not only due to high memory requirements but also because of increased latency at inference. We propose a novel model compression method that generates a sparse trained model without additional overhead: by allowing (i) dynamic allocation of the sparsity pattern and (ii) incorporating feedback signal to reactivate prematurely pruned weights we obtain a performant sparse model in one single training pass (retraining is not needed, but can further improve the performance). We evaluate our","authors_text":"Daniil Dmitriev, Luis Barba, Martin Jaggi, Sebastian U. Stich, Tao Lin","cross_cats":["stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-06-12T15:07:08Z","title":"Dynamic Model Pruning with Feedback"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.07253","kind":"arxiv","version":1},"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":"sha256:9785ecf1aa8907673e664a06e19c401d1448bceecf7180437977d413d384f55a","target":"record","created_at":"2026-07-05T01:09:53Z","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":"1d7d7d5fdd61d34079afe3d3c57d50a16439501c1cfeefe54c49d5d198a58577","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-06-12T15:07:08Z","title_canon_sha256":"617aefe6112e32c77041558f6589e26b0db2573cfed3654846799fae4f29aad9"},"schema_version":"1.0","source":{"id":"2006.07253","kind":"arxiv","version":1}},"canonical_sha256":"0d1e6edf1012c02649d57f5cc4e37f25be4e547f576a37732da931eaf9e12176","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0d1e6edf1012c02649d57f5cc4e37f25be4e547f576a37732da931eaf9e12176","first_computed_at":"2026-07-05T01:09:53.174338Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:09:53.174338Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TBHLxNJUDXbnH+tzrsY8OnpcKjJSaMD0afsWanfjDnIqubS6hErOOAK69Kwj9e1O3pi1YKG8lB3Hb7HtMqv/Bw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:09:53.174893Z","signed_message":"canonical_sha256_bytes"},"source_id":"2006.07253","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9785ecf1aa8907673e664a06e19c401d1448bceecf7180437977d413d384f55a","sha256:e2793f1cd9c290460ef6a2f0ea2bc5aae82e5a11ea9781b59806b29571d90c5f"],"state_sha256":"323395ff77fdfc9cc233bc7858d1be319920d5ce50092c26782cb726c84d7464"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i4Br3oRY1fgTVUKKX+aWT3Po29Wct+Pf5oY9RPrchsnYh5p/ppJnYvbrp9oXBb9ZXHBKdSKF81oSm8J7cNBYBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T03:05:23.353583Z","bundle_sha256":"75b2cac01b8b7bbf80f5da2ef5d358cc3a2926c713c83d4b31a8b2a4ce1d85f7"}}