{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:5G7EMA6M54PXEC3FHYUWNYTHOI","short_pith_number":"pith:5G7EMA6M","schema_version":"1.0","canonical_sha256":"e9be4603ccef1f720b653e2966e2677218b8208a2a4d922003c87f82f9bf730f","source":{"kind":"arxiv","id":"1909.03434","version":1},"attestation_state":"computed","paper":{"title":"Order-free Learning Alleviating Exposure Bias in Multi-label Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Che-Ping Tsai, Hung-yi Lee","submitted_at":"2019-09-08T11:53:24Z","abstract_excerpt":"Multi-label classification (MLC) assigns multiple labels to each sample. Prior studies show that MLC can be transformed to a sequence prediction problem with a recurrent neural network (RNN) decoder to model the label dependency. However, training a RNN decoder requires a predefined order of labels, which is not directly available in the MLC specification. Besides, RNN thus trained tends to overfit the label combinations in the training set and have difficulty generating unseen label sequences. In this paper, we propose a new framework for MLC which does not rely on a predefined label order an"},"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":"1909.03434","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-08T11:53:24Z","cross_cats_sorted":["cs.CL","cs.SD","eess.AS","stat.ML"],"title_canon_sha256":"d0b01ae2798c22623d8e3c262eab99cb34aaa83c561c224309774906b7778cbb","abstract_canon_sha256":"4e01285a095bf0a4839e0ee9b4642fc260752798015583e573a5bc9afaab63d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:03:05.586886Z","signature_b64":"bn814oGcjlSe2xYCjYvNgphtHX4CHNHhMCmMRULUrCroS+CjgWL9g1K0TzKZ0FHuzsQtLETu2bKoDl92/F4ACg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e9be4603ccef1f720b653e2966e2677218b8208a2a4d922003c87f82f9bf730f","last_reissued_at":"2026-07-05T00:03:05.586548Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:03:05.586548Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Order-free Learning Alleviating Exposure Bias in Multi-label Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS","stat.ML"],"primary_cat":"cs.LG","authors_text":"Che-Ping Tsai, Hung-yi Lee","submitted_at":"2019-09-08T11:53:24Z","abstract_excerpt":"Multi-label classification (MLC) assigns multiple labels to each sample. Prior studies show that MLC can be transformed to a sequence prediction problem with a recurrent neural network (RNN) decoder to model the label dependency. However, training a RNN decoder requires a predefined order of labels, which is not directly available in the MLC specification. Besides, RNN thus trained tends to overfit the label combinations in the training set and have difficulty generating unseen label sequences. In this paper, we propose a new framework for MLC which does not rely on a predefined label order an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.03434","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/1909.03434/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":"1909.03434","created_at":"2026-07-05T00:03:05.586606+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.03434v1","created_at":"2026-07-05T00:03:05.586606+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.03434","created_at":"2026-07-05T00:03:05.586606+00:00"},{"alias_kind":"pith_short_12","alias_value":"5G7EMA6M54PX","created_at":"2026-07-05T00:03:05.586606+00:00"},{"alias_kind":"pith_short_16","alias_value":"5G7EMA6M54PXEC3F","created_at":"2026-07-05T00:03:05.586606+00:00"},{"alias_kind":"pith_short_8","alias_value":"5G7EMA6M","created_at":"2026-07-05T00:03:05.586606+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.14508","citing_title":"Machine learning revolution for exoplanet direct imaging detection: transformer architectures","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI","json":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI.json","graph_json":"https://pith.science/api/pith-number/5G7EMA6M54PXEC3FHYUWNYTHOI/graph.json","events_json":"https://pith.science/api/pith-number/5G7EMA6M54PXEC3FHYUWNYTHOI/events.json","paper":"https://pith.science/paper/5G7EMA6M"},"agent_actions":{"view_html":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI","download_json":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI.json","view_paper":"https://pith.science/paper/5G7EMA6M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.03434&json=true","fetch_graph":"https://pith.science/api/pith-number/5G7EMA6M54PXEC3FHYUWNYTHOI/graph.json","fetch_events":"https://pith.science/api/pith-number/5G7EMA6M54PXEC3FHYUWNYTHOI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI/action/storage_attestation","attest_author":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI/action/author_attestation","sign_citation":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI/action/citation_signature","submit_replication":"https://pith.science/pith/5G7EMA6M54PXEC3FHYUWNYTHOI/action/replication_record"}},"created_at":"2026-07-05T00:03:05.586606+00:00","updated_at":"2026-07-05T00:03:05.586606+00:00"}