{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:MXJZKLTFIWGWNSFLGUMZFWW5YS","short_pith_number":"pith:MXJZKLTF","canonical_record":{"source":{"id":"1912.06844","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-14T13:35:32Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"654286f0470f1e6da7fd57e82dfd29ed267ab044533c1d32a98c637ae4180508","abstract_canon_sha256":"6bf3eb185add7c1f90bb046b1a61b2240c559091b090ad4398ecdc7e0981d485"},"schema_version":"1.0"},"canonical_sha256":"65d3952e65458d66c8ab351992daddc4a4c0f5d1a7ae3c70f864a491b6d03bcc","source":{"kind":"arxiv","id":"1912.06844","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1912.06844","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"arxiv_version","alias_value":"1912.06844v1","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.06844","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"pith_short_12","alias_value":"MXJZKLTFIWGW","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"pith_short_16","alias_value":"MXJZKLTFIWGWNSFL","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"pith_short_8","alias_value":"MXJZKLTF","created_at":"2026-07-05T00:26:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:MXJZKLTFIWGWNSFLGUMZFWW5YS","target":"record","payload":{"canonical_record":{"source":{"id":"1912.06844","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-14T13:35:32Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"654286f0470f1e6da7fd57e82dfd29ed267ab044533c1d32a98c637ae4180508","abstract_canon_sha256":"6bf3eb185add7c1f90bb046b1a61b2240c559091b090ad4398ecdc7e0981d485"},"schema_version":"1.0"},"canonical_sha256":"65d3952e65458d66c8ab351992daddc4a4c0f5d1a7ae3c70f864a491b6d03bcc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:26:15.685076Z","signature_b64":"aofNVWaXeb0QXbpwAoc8/AroS6hGQo4mshjbUtHQeoryH7W3FRt+pVWgCzEH4k8ZHT5Ec6w9o5xQffOOGuXvCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"65d3952e65458d66c8ab351992daddc4a4c0f5d1a7ae3c70f864a491b6d03bcc","last_reissued_at":"2026-07-05T00:26:15.684405Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:26:15.684405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1912.06844","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":"2026-07-05T00:26:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uEEAfKMoTizZ7rVgsxaCkpPCQ5Mw32iJBwRx7o9VVVRfz+vXspgsDe18PHb+5aBUfzUREKTVbEg39cKYYCfACA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T00:30:26.707219Z"},"content_sha256":"8428fd1b13171971821a5bf3e5ab673cf9b6f2eb136f598da703ac5eb94ab960","schema_version":"1.0","event_id":"sha256:8428fd1b13171971821a5bf3e5ab673cf9b6f2eb136f598da703ac5eb94ab960"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:MXJZKLTFIWGWNSFLGUMZFWW5YS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Regularizing Deep Multi-Task Networks using Orthogonal Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Mihai Suteu, Yike Guo","submitted_at":"2019-12-14T13:35:32Z","abstract_excerpt":"Deep neural networks are a promising approach towards multi-task learning because of their capability to leverage knowledge across domains and learn general purpose representations. Nevertheless, they can fail to live up to these promises as tasks often compete for a model's limited resources, potentially leading to lower overall performance. In this work we tackle the issue of interfering tasks through a comprehensive analysis of their training, derived from looking at the interaction between gradients within their shared parameters. Our empirical results show that well-performing models have"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.06844","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/1912.06844/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":"2026-07-05T00:26:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zzlj+Tn8HfNTBfRTP/2aUJB+rmzjrwagz3aFMoxnjmZuCPSuCZeEoKxpaYw41BGivoS5x7Ml7n2xKClPVN22Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T00:30:26.707733Z"},"content_sha256":"67491552169e21a9691d5e7a24e2ee4e7634056523bffe3fa494fa1c5e9a680b","schema_version":"1.0","event_id":"sha256:67491552169e21a9691d5e7a24e2ee4e7634056523bffe3fa494fa1c5e9a680b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS/bundle.json","state_url":"https://pith.science/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS/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":"2026-08-07T00:30:26Z","links":{"resolver":"https://pith.science/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS","bundle":"https://pith.science/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS/bundle.json","state":"https://pith.science/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MXJZKLTFIWGWNSFLGUMZFWW5YS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:MXJZKLTFIWGWNSFLGUMZFWW5YS","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":"6bf3eb185add7c1f90bb046b1a61b2240c559091b090ad4398ecdc7e0981d485","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-14T13:35:32Z","title_canon_sha256":"654286f0470f1e6da7fd57e82dfd29ed267ab044533c1d32a98c637ae4180508"},"schema_version":"1.0","source":{"id":"1912.06844","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1912.06844","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"arxiv_version","alias_value":"1912.06844v1","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.06844","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"pith_short_12","alias_value":"MXJZKLTFIWGW","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"pith_short_16","alias_value":"MXJZKLTFIWGWNSFL","created_at":"2026-07-05T00:26:15Z"},{"alias_kind":"pith_short_8","alias_value":"MXJZKLTF","created_at":"2026-07-05T00:26:15Z"}],"graph_snapshots":[{"event_id":"sha256:67491552169e21a9691d5e7a24e2ee4e7634056523bffe3fa494fa1c5e9a680b","target":"graph","created_at":"2026-07-05T00:26:15Z","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/1912.06844/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep neural networks are a promising approach towards multi-task learning because of their capability to leverage knowledge across domains and learn general purpose representations. Nevertheless, they can fail to live up to these promises as tasks often compete for a model's limited resources, potentially leading to lower overall performance. In this work we tackle the issue of interfering tasks through a comprehensive analysis of their training, derived from looking at the interaction between gradients within their shared parameters. Our empirical results show that well-performing models have","authors_text":"Mihai Suteu, Yike Guo","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-14T13:35:32Z","title":"Regularizing Deep Multi-Task Networks using Orthogonal Gradients"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.06844","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":"sha256:8428fd1b13171971821a5bf3e5ab673cf9b6f2eb136f598da703ac5eb94ab960","target":"record","created_at":"2026-07-05T00:26:15Z","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":"6bf3eb185add7c1f90bb046b1a61b2240c559091b090ad4398ecdc7e0981d485","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-14T13:35:32Z","title_canon_sha256":"654286f0470f1e6da7fd57e82dfd29ed267ab044533c1d32a98c637ae4180508"},"schema_version":"1.0","source":{"id":"1912.06844","kind":"arxiv","version":1}},"canonical_sha256":"65d3952e65458d66c8ab351992daddc4a4c0f5d1a7ae3c70f864a491b6d03bcc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"65d3952e65458d66c8ab351992daddc4a4c0f5d1a7ae3c70f864a491b6d03bcc","first_computed_at":"2026-07-05T00:26:15.684405Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:26:15.684405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aofNVWaXeb0QXbpwAoc8/AroS6hGQo4mshjbUtHQeoryH7W3FRt+pVWgCzEH4k8ZHT5Ec6w9o5xQffOOGuXvCA==","signature_status":"signed_v1","signed_at":"2026-07-05T00:26:15.685076Z","signed_message":"canonical_sha256_bytes"},"source_id":"1912.06844","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8428fd1b13171971821a5bf3e5ab673cf9b6f2eb136f598da703ac5eb94ab960","sha256:67491552169e21a9691d5e7a24e2ee4e7634056523bffe3fa494fa1c5e9a680b"],"state_sha256":"c39d5342a7c2b42e94db7be13d94bd042a6e65e7f3246554e7d8aee222566822"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FjAcBoUtLMkMaF7FAQnZp+DZuurMN5KwLgSratU25G0iv87LFoZYyytjtLkr4msqnlfOHNV/ITag+7wNch6NDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T00:30:26.711373Z","bundle_sha256":"86aade804e5e0321dc3e51637ee4026a4716dc9edbcb9a1e49be033af0fc7af3"}}