{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HVJBB6CXQNFHDKI5LYOC6NOV4S","short_pith_number":"pith:HVJBB6CX","schema_version":"1.0","canonical_sha256":"3d5210f857834a71a91d5e1c2f35d5e4b254a14ada7990b7d9b0f21906b974cd","source":{"kind":"arxiv","id":"2606.18721","version":1},"attestation_state":"computed","paper":{"title":"Rethinking the Pointer Loss in Table Structure Recognition: Geometry-Aware Pointer Loss for Spatial Locality","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hong-Jun Choi, Jaeyoung Kim, Jongho Lee","submitted_at":"2026-06-17T05:59:31Z","abstract_excerpt":"Table Structure Recognition (TSR) using a pointer network achieves impressive results by predicting HTML sequences while aligning tags to detected text (or cell) regions. However, our analysis reveals that when pointer networks fail, 79.6% of errors occur between spatially adjacent cells (Manhattan distance <= 2). Despite this, standard cross-entropy loss weights all negative candidates equally. In this work, we propose Geometry-Aware Pointer (GAP) Loss, which reweights the cross-entropy objective based on spatial proximity to ground truth. By applying inverse distance weighting, GAP focuses g"},"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":"2606.18721","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-17T05:59:31Z","cross_cats_sorted":[],"title_canon_sha256":"1f479598bae3346bae2484f2c990687c4c298f964a21ab4dd3eb8e8a36ffe488","abstract_canon_sha256":"fa85ff4c43e6c32de709a41a36a481ebd78e1a22937db8c18a100c04ff0bf3ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:11:45.613077Z","signature_b64":"fYHM33Xelq3X8qRvMa/ZRY1W2mJVZyJcLOeDeHyFfhLkStTX8+Gb+wClrIp6xzJPA4R1Z/YnmUGFEaczuLdkCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d5210f857834a71a91d5e1c2f35d5e4b254a14ada7990b7d9b0f21906b974cd","last_reissued_at":"2026-06-19T16:11:45.612712Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:11:45.612712Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking the Pointer Loss in Table Structure Recognition: Geometry-Aware Pointer Loss for Spatial Locality","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hong-Jun Choi, Jaeyoung Kim, Jongho Lee","submitted_at":"2026-06-17T05:59:31Z","abstract_excerpt":"Table Structure Recognition (TSR) using a pointer network achieves impressive results by predicting HTML sequences while aligning tags to detected text (or cell) regions. However, our analysis reveals that when pointer networks fail, 79.6% of errors occur between spatially adjacent cells (Manhattan distance <= 2). Despite this, standard cross-entropy loss weights all negative candidates equally. In this work, we propose Geometry-Aware Pointer (GAP) Loss, which reweights the cross-entropy objective based on spatial proximity to ground truth. By applying inverse distance weighting, GAP focuses g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18721","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/2606.18721/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":"2606.18721","created_at":"2026-06-19T16:11:45.612768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.18721v1","created_at":"2026-06-19T16:11:45.612768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18721","created_at":"2026-06-19T16:11:45.612768+00:00"},{"alias_kind":"pith_short_12","alias_value":"HVJBB6CXQNFH","created_at":"2026-06-19T16:11:45.612768+00:00"},{"alias_kind":"pith_short_16","alias_value":"HVJBB6CXQNFHDKI5","created_at":"2026-06-19T16:11:45.612768+00:00"},{"alias_kind":"pith_short_8","alias_value":"HVJBB6CX","created_at":"2026-06-19T16:11:45.612768+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S","json":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S.json","graph_json":"https://pith.science/api/pith-number/HVJBB6CXQNFHDKI5LYOC6NOV4S/graph.json","events_json":"https://pith.science/api/pith-number/HVJBB6CXQNFHDKI5LYOC6NOV4S/events.json","paper":"https://pith.science/paper/HVJBB6CX"},"agent_actions":{"view_html":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S","download_json":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S.json","view_paper":"https://pith.science/paper/HVJBB6CX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.18721&json=true","fetch_graph":"https://pith.science/api/pith-number/HVJBB6CXQNFHDKI5LYOC6NOV4S/graph.json","fetch_events":"https://pith.science/api/pith-number/HVJBB6CXQNFHDKI5LYOC6NOV4S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S/action/storage_attestation","attest_author":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S/action/author_attestation","sign_citation":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S/action/citation_signature","submit_replication":"https://pith.science/pith/HVJBB6CXQNFHDKI5LYOC6NOV4S/action/replication_record"}},"created_at":"2026-06-19T16:11:45.612768+00:00","updated_at":"2026-06-19T16:11:45.612768+00:00"}