{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:ZMLQ7VUXDI56FE3LPEVKFSU7WJ","short_pith_number":"pith:ZMLQ7VUX","schema_version":"1.0","canonical_sha256":"cb170fd6971a3be2936b792aa2ca9fb277fb041306fed00d696006a969612cc4","source":{"kind":"arxiv","id":"2006.04096","version":1},"attestation_state":"computed","paper":{"title":"Robust Learning Through Cross-Task Consistency","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Sax, Amir Zamir, Jitendra Malik, Leonidas Guibas, Nikhil Cheerla, O\\u{g}uzhan Kar, Rohan Suri, Teresa Yeo, Zhangjie Cao","submitted_at":"2020-06-07T09:24:33Z","abstract_excerpt":"Visual perception entails solving a wide set of tasks, e.g., object detection, depth estimation, etc. The predictions made for multiple tasks from the same image are not independent, and therefore, are expected to be consistent. We propose a broadly applicable and fully computational method for augmenting learning with Cross-Task Consistency. The proposed formulation is based on inference-path invariance over a graph of arbitrary tasks. We observe that learning with cross-task consistency leads to more accurate predictions and better generalization to out-of-distribution inputs. This framework"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2006.04096","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-06-07T09:24:33Z","cross_cats_sorted":["cs.GR","cs.LG"],"title_canon_sha256":"53a0eeae768fa025fbf8c8c4a07ae251794aaded68219e7253d29a4f31ae4d04","abstract_canon_sha256":"5bba2bde7bd5244ef59769e27d1056de9513bb5d030ed1c3b4c5d0b442d6af82"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:08:47.545839Z","signature_b64":"AkUTAz6AvPBZo6iisPcbOmO/2WSTedx/XFSzQu1KtxuikdP4bx+LrQyo6ASRwW51MGsyyaadRxYptj8nX4zIAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb170fd6971a3be2936b792aa2ca9fb277fb041306fed00d696006a969612cc4","last_reissued_at":"2026-07-05T01:08:47.545441Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:08:47.545441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust Learning Through Cross-Task Consistency","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Sax, Amir Zamir, Jitendra Malik, Leonidas Guibas, Nikhil Cheerla, O\\u{g}uzhan Kar, Rohan Suri, Teresa Yeo, Zhangjie Cao","submitted_at":"2020-06-07T09:24:33Z","abstract_excerpt":"Visual perception entails solving a wide set of tasks, e.g., object detection, depth estimation, etc. The predictions made for multiple tasks from the same image are not independent, and therefore, are expected to be consistent. We propose a broadly applicable and fully computational method for augmenting learning with Cross-Task Consistency. The proposed formulation is based on inference-path invariance over a graph of arbitrary tasks. We observe that learning with cross-task consistency leads to more accurate predictions and better generalization to out-of-distribution inputs. This framework"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.04096","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/2006.04096/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2006.04096","created_at":"2026-07-05T01:08:47.545500+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.04096v1","created_at":"2026-07-05T01:08:47.545500+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.04096","created_at":"2026-07-05T01:08:47.545500+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZMLQ7VUXDI56","created_at":"2026-07-05T01:08:47.545500+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZMLQ7VUXDI56FE3L","created_at":"2026-07-05T01:08:47.545500+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZMLQ7VUX","created_at":"2026-07-05T01:08:47.545500+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01557","citing_title":"Everywhere Learning: Artificial Intelligence with Pointwise Constraints","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14055","citing_title":"PEML: Parameter-efficient Multi-Task Learning with Optimized Continuous Prompts","ref_index":87,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ","json":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ.json","graph_json":"https://pith.science/api/pith-number/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/graph.json","events_json":"https://pith.science/api/pith-number/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/events.json","paper":"https://pith.science/paper/ZMLQ7VUX"},"agent_actions":{"view_html":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ","download_json":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ.json","view_paper":"https://pith.science/paper/ZMLQ7VUX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.04096&json=true","fetch_graph":"https://pith.science/api/pith-number/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/action/storage_attestation","attest_author":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/action/author_attestation","sign_citation":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/action/citation_signature","submit_replication":"https://pith.science/pith/ZMLQ7VUXDI56FE3LPEVKFSU7WJ/action/replication_record"}},"created_at":"2026-07-05T01:08:47.545500+00:00","updated_at":"2026-07-05T01:08:47.545500+00:00"}