{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:X2XC2RBRMDN52VK3PKH6BPCQBQ","short_pith_number":"pith:X2XC2RBR","canonical_record":{"source":{"id":"1606.06160","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2016-06-20T15:02:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f5c5654d85aae4aeed5ac597cebe9538f2eb9f0d9ac4260b3eda570dac417a2b","abstract_canon_sha256":"afb5cd0d291695dd979479df901d570d131776ba3455fc85ae2d3de11168c3f3"},"schema_version":"1.0"},"canonical_sha256":"beae2d443160dbdd555b7a8fe0bc500c30d7835e2acc296d2e4dc4fdfab4c732","source":{"kind":"arxiv","id":"1606.06160","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1606.06160","created_at":"2026-05-18T00:24:35Z"},{"alias_kind":"arxiv_version","alias_value":"1606.06160v3","created_at":"2026-05-18T00:24:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.06160","created_at":"2026-05-18T00:24:35Z"},{"alias_kind":"pith_short_12","alias_value":"X2XC2RBRMDN5","created_at":"2026-05-18T12:30:51Z"},{"alias_kind":"pith_short_16","alias_value":"X2XC2RBRMDN52VK3","created_at":"2026-05-18T12:30:51Z"},{"alias_kind":"pith_short_8","alias_value":"X2XC2RBR","created_at":"2026-05-18T12:30:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:X2XC2RBRMDN52VK3PKH6BPCQBQ","target":"record","payload":{"canonical_record":{"source":{"id":"1606.06160","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2016-06-20T15:02:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f5c5654d85aae4aeed5ac597cebe9538f2eb9f0d9ac4260b3eda570dac417a2b","abstract_canon_sha256":"afb5cd0d291695dd979479df901d570d131776ba3455fc85ae2d3de11168c3f3"},"schema_version":"1.0"},"canonical_sha256":"beae2d443160dbdd555b7a8fe0bc500c30d7835e2acc296d2e4dc4fdfab4c732","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:24:35.593192Z","signature_b64":"Y5QG2ZoGWsmzQ4uhWb1ibNm6ACl4ga5yPViIYf9QcCsp+QIaZIvZuNdlVgugGtZZJSRFD89BZ7dtUIlU12WDAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"beae2d443160dbdd555b7a8fe0bc500c30d7835e2acc296d2e4dc4fdfab4c732","last_reissued_at":"2026-05-18T00:24:35.592739Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:24:35.592739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1606.06160","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":"2026-05-18T00:24:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yMs7OELCECBwsYnEp5SG0HU96sQLg/KNzgUuGK3KCnovx7lgLvPPH0p3u89HWTIGPeBkD+W1vGsTKdcAvI5qCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T01:19:47.610001Z"},"content_sha256":"a8bb602428c7bc0534473794c2158593212b146997f80f665c120c6c610cf87c","schema_version":"1.0","event_id":"sha256:a8bb602428c7bc0534473794c2158593212b146997f80f665c120c6c610cf87c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:X2XC2RBRMDN52VK3PKH6BPCQBQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.NE","authors_text":"He Wen, Shuchang Zhou, Xinyu Zhou, Yuheng Zou, Yuxin Wu, Zekun Ni","submitted_at":"2016-06-20T15:02:31Z","abstract_excerpt":"We propose DoReFa-Net, a method to train convolutional neural networks that have low bitwidth weights and activations using low bitwidth parameter gradients. In particular, during backward pass, parameter gradients are stochastically quantized to low bitwidth numbers before being propagated to convolutional layers. As convolutions during forward/backward passes can now operate on low bitwidth weights and activations/gradients respectively, DoReFa-Net can use bit convolution kernels to accelerate both training and inference. Moreover, as bit convolutions can be efficiently implemented on CPU, F"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.06160","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":""},"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":"2026-05-18T00:24:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"O27pzryLpBviircjzWQZOWI/oTY7+YbBWQD3Appk6du8BNjnxRjf1ygf6LMKTi44u8zgy/sDjZw4hCJGS8lgBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T01:19:47.610755Z"},"content_sha256":"62cef8334d432d69a022be46f51bfeea4e496eff41cf623aae456f7e853ff848","schema_version":"1.0","event_id":"sha256:62cef8334d432d69a022be46f51bfeea4e496eff41cf623aae456f7e853ff848"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ/bundle.json","state_url":"https://pith.science/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ/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":"2026-05-31T01:19:47Z","links":{"resolver":"https://pith.science/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ","bundle":"https://pith.science/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ/bundle.json","state":"https://pith.science/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X2XC2RBRMDN52VK3PKH6BPCQBQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:X2XC2RBRMDN52VK3PKH6BPCQBQ","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":"afb5cd0d291695dd979479df901d570d131776ba3455fc85ae2d3de11168c3f3","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2016-06-20T15:02:31Z","title_canon_sha256":"f5c5654d85aae4aeed5ac597cebe9538f2eb9f0d9ac4260b3eda570dac417a2b"},"schema_version":"1.0","source":{"id":"1606.06160","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1606.06160","created_at":"2026-05-18T00:24:35Z"},{"alias_kind":"arxiv_version","alias_value":"1606.06160v3","created_at":"2026-05-18T00:24:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.06160","created_at":"2026-05-18T00:24:35Z"},{"alias_kind":"pith_short_12","alias_value":"X2XC2RBRMDN5","created_at":"2026-05-18T12:30:51Z"},{"alias_kind":"pith_short_16","alias_value":"X2XC2RBRMDN52VK3","created_at":"2026-05-18T12:30:51Z"},{"alias_kind":"pith_short_8","alias_value":"X2XC2RBR","created_at":"2026-05-18T12:30:51Z"}],"graph_snapshots":[{"event_id":"sha256:62cef8334d432d69a022be46f51bfeea4e496eff41cf623aae456f7e853ff848","target":"graph","created_at":"2026-05-18T00:24:35Z","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":"We propose DoReFa-Net, a method to train convolutional neural networks that have low bitwidth weights and activations using low bitwidth parameter gradients. In particular, during backward pass, parameter gradients are stochastically quantized to low bitwidth numbers before being propagated to convolutional layers. As convolutions during forward/backward passes can now operate on low bitwidth weights and activations/gradients respectively, DoReFa-Net can use bit convolution kernels to accelerate both training and inference. Moreover, as bit convolutions can be efficiently implemented on CPU, F","authors_text":"He Wen, Shuchang Zhou, Xinyu Zhou, Yuheng Zou, Yuxin Wu, Zekun Ni","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2016-06-20T15:02:31Z","title":"DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.06160","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:a8bb602428c7bc0534473794c2158593212b146997f80f665c120c6c610cf87c","target":"record","created_at":"2026-05-18T00:24:35Z","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":"afb5cd0d291695dd979479df901d570d131776ba3455fc85ae2d3de11168c3f3","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NE","submitted_at":"2016-06-20T15:02:31Z","title_canon_sha256":"f5c5654d85aae4aeed5ac597cebe9538f2eb9f0d9ac4260b3eda570dac417a2b"},"schema_version":"1.0","source":{"id":"1606.06160","kind":"arxiv","version":3}},"canonical_sha256":"beae2d443160dbdd555b7a8fe0bc500c30d7835e2acc296d2e4dc4fdfab4c732","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"beae2d443160dbdd555b7a8fe0bc500c30d7835e2acc296d2e4dc4fdfab4c732","first_computed_at":"2026-05-18T00:24:35.592739Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:24:35.592739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Y5QG2ZoGWsmzQ4uhWb1ibNm6ACl4ga5yPViIYf9QcCsp+QIaZIvZuNdlVgugGtZZJSRFD89BZ7dtUIlU12WDAg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:24:35.593192Z","signed_message":"canonical_sha256_bytes"},"source_id":"1606.06160","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a8bb602428c7bc0534473794c2158593212b146997f80f665c120c6c610cf87c","sha256:62cef8334d432d69a022be46f51bfeea4e496eff41cf623aae456f7e853ff848"],"state_sha256":"be63b4f77a9d09c343d9616004daffbf785fe2b97186dc024fc24fb3817a73e3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"42jx3jLKAB6ydJDxyhp4SAVPQcgLqPTLngEUL4uo/ojs4CiU+YoqcKtxVl9OOXXEDpRN9b23e6uMyKsSjIM1BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T01:19:47.614680Z","bundle_sha256":"646c11531d39e6bb1e135e6efa76e18ef1052108dad85753fe17354a1f182451"}}