{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:ZOF3QYPLXTGTDVBKVOC7SGGV2S","short_pith_number":"pith:ZOF3QYPL","canonical_record":{"source":{"id":"2007.06889","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-07-14T08:02:42Z","cross_cats_sorted":[],"title_canon_sha256":"e7e9458736fff77ae7448720c706e88f0b8cf461ddb88691220037e1745c832a","abstract_canon_sha256":"f9dbd6eb1db3189de5f0f9fe8e1ebbcd0e26dc0e146bd4ef1d472add524cc2dd"},"schema_version":"1.0"},"canonical_sha256":"cb8bb861ebbccd31d42aab85f918d5d4acf5a27c1cae8169728b055de0c9e2eb","source":{"kind":"arxiv","id":"2007.06889","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.06889","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"arxiv_version","alias_value":"2007.06889v2","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.06889","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"pith_short_12","alias_value":"ZOF3QYPLXTGT","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"pith_short_16","alias_value":"ZOF3QYPLXTGTDVBK","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"pith_short_8","alias_value":"ZOF3QYPL","created_at":"2026-07-05T01:37:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:ZOF3QYPLXTGTDVBKVOC7SGGV2S","target":"record","payload":{"canonical_record":{"source":{"id":"2007.06889","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-07-14T08:02:42Z","cross_cats_sorted":[],"title_canon_sha256":"e7e9458736fff77ae7448720c706e88f0b8cf461ddb88691220037e1745c832a","abstract_canon_sha256":"f9dbd6eb1db3189de5f0f9fe8e1ebbcd0e26dc0e146bd4ef1d472add524cc2dd"},"schema_version":"1.0"},"canonical_sha256":"cb8bb861ebbccd31d42aab85f918d5d4acf5a27c1cae8169728b055de0c9e2eb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:37:42.900918Z","signature_b64":"z7FoXfqqCbZxCXXqopxRPwHRw0J+LFpH90imQ6FHBkp5fTNIL5o5/Ol0FcGWZrAd6ytSOpIK+mUN6uOjfbxkAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb8bb861ebbccd31d42aab85f918d5d4acf5a27c1cae8169728b055de0c9e2eb","last_reissued_at":"2026-07-05T01:37:42.900538Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:37:42.900538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2007.06889","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":"2026-07-05T01:37:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z7EbAWkEHnO1lC4R2+sfEfi5SBh5NzVfPuo42CeOvnRNXTTUUIaAnVdSZnI5IfVRbJ64X31ggjukWzCKUkAACQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:12:35.634655Z"},"content_sha256":"3130882bfd5dba4d7f46ebac746e1d09f6507c158a09535daf606662f1ad79db","schema_version":"1.0","event_id":"sha256:3130882bfd5dba4d7f46ebac746e1d09f6507c158a09535daf606662f1ad79db"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:ZOF3QYPLXTGTDVBKVOC7SGGV2S","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Knowledge Distillation for Multi-task Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hakan Bilen, Wei-Hong Li","submitted_at":"2020-07-14T08:02:42Z","abstract_excerpt":"Multi-task learning (MTL) is to learn one single model that performs multiple tasks for achieving good performance on all tasks and lower cost on computation. Learning such a model requires to jointly optimize losses of a set of tasks with different difficulty levels, magnitudes, and characteristics (e.g. cross-entropy, Euclidean loss), leading to the imbalance problem in multi-task learning. To address the imbalance problem, we propose a knowledge distillation based method in this work. We first learn a task-specific model for each task. We then learn the multi-task model for minimizing task-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.06889","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2007.06889/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":"2026-07-05T01:37:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WWH/vrd5JHiSCqcV4o2QMOccg48QihGaFloNdbS7hfQ7g4FL4w1fVrDaCx08oDaFUZpucIPAFceOw/2pVWzaAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:12:35.635158Z"},"content_sha256":"ac94b61234ec33e3c68a9c82da25e56bce8b4119b80c15282f3a7df7f9106e13","schema_version":"1.0","event_id":"sha256:ac94b61234ec33e3c68a9c82da25e56bce8b4119b80c15282f3a7df7f9106e13"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S/bundle.json","state_url":"https://pith.science/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S/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":"2026-08-08T18:12:35Z","links":{"resolver":"https://pith.science/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S","bundle":"https://pith.science/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S/bundle.json","state":"https://pith.science/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZOF3QYPLXTGTDVBKVOC7SGGV2S/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:ZOF3QYPLXTGTDVBKVOC7SGGV2S","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":"f9dbd6eb1db3189de5f0f9fe8e1ebbcd0e26dc0e146bd4ef1d472add524cc2dd","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-07-14T08:02:42Z","title_canon_sha256":"e7e9458736fff77ae7448720c706e88f0b8cf461ddb88691220037e1745c832a"},"schema_version":"1.0","source":{"id":"2007.06889","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.06889","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"arxiv_version","alias_value":"2007.06889v2","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.06889","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"pith_short_12","alias_value":"ZOF3QYPLXTGT","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"pith_short_16","alias_value":"ZOF3QYPLXTGTDVBK","created_at":"2026-07-05T01:37:42Z"},{"alias_kind":"pith_short_8","alias_value":"ZOF3QYPL","created_at":"2026-07-05T01:37:42Z"}],"graph_snapshots":[{"event_id":"sha256:ac94b61234ec33e3c68a9c82da25e56bce8b4119b80c15282f3a7df7f9106e13","target":"graph","created_at":"2026-07-05T01:37:42Z","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/2007.06889/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-task learning (MTL) is to learn one single model that performs multiple tasks for achieving good performance on all tasks and lower cost on computation. Learning such a model requires to jointly optimize losses of a set of tasks with different difficulty levels, magnitudes, and characteristics (e.g. cross-entropy, Euclidean loss), leading to the imbalance problem in multi-task learning. To address the imbalance problem, we propose a knowledge distillation based method in this work. We first learn a task-specific model for each task. We then learn the multi-task model for minimizing task-","authors_text":"Hakan Bilen, Wei-Hong Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-07-14T08:02:42Z","title":"Knowledge Distillation for Multi-task Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.06889","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":"sha256:3130882bfd5dba4d7f46ebac746e1d09f6507c158a09535daf606662f1ad79db","target":"record","created_at":"2026-07-05T01:37:42Z","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":"f9dbd6eb1db3189de5f0f9fe8e1ebbcd0e26dc0e146bd4ef1d472add524cc2dd","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-07-14T08:02:42Z","title_canon_sha256":"e7e9458736fff77ae7448720c706e88f0b8cf461ddb88691220037e1745c832a"},"schema_version":"1.0","source":{"id":"2007.06889","kind":"arxiv","version":2}},"canonical_sha256":"cb8bb861ebbccd31d42aab85f918d5d4acf5a27c1cae8169728b055de0c9e2eb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb8bb861ebbccd31d42aab85f918d5d4acf5a27c1cae8169728b055de0c9e2eb","first_computed_at":"2026-07-05T01:37:42.900538Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:37:42.900538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z7FoXfqqCbZxCXXqopxRPwHRw0J+LFpH90imQ6FHBkp5fTNIL5o5/Ol0FcGWZrAd6ytSOpIK+mUN6uOjfbxkAg==","signature_status":"signed_v1","signed_at":"2026-07-05T01:37:42.900918Z","signed_message":"canonical_sha256_bytes"},"source_id":"2007.06889","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3130882bfd5dba4d7f46ebac746e1d09f6507c158a09535daf606662f1ad79db","sha256:ac94b61234ec33e3c68a9c82da25e56bce8b4119b80c15282f3a7df7f9106e13"],"state_sha256":"5a4931adf08ba4a372d1c1f62c815364cc19172d6cb527d148d7db170bcdac1e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aR4Qly1/b1/vHnoC/8/QkQK9STLsT8K/eXWO9Je9dPJLCdTc4ql1h8SwTjBp9JtmafFdVYdskNx1sXheY1hDDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T18:12:35.640542Z","bundle_sha256":"1b7a69575b62218f9ea1847be3dc304d7dba40d8a54444c08f0a4015b82b7767"}}