{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AQSBO7BXCAHDDDTKVLKRKWLUHJ","short_pith_number":"pith:AQSBO7BX","schema_version":"1.0","canonical_sha256":"0424177c37100e318e6aaad51559743a4e8ba1421cc16fd0573eaca7138f7f7b","source":{"kind":"arxiv","id":"2505.15649","version":1},"attestation_state":"computed","paper":{"title":"The Devil is in Fine-tuning and Long-tailed Problems:A New Benchmark for Scene Text Detection","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiahao Lyu, Tianjiao Cao, Weichao Zeng, Weimin Mu, Yu Zhou","submitted_at":"2025-05-21T15:26:46Z","abstract_excerpt":"Scene text detection has seen the emergence of high-performing methods that excel on academic benchmarks. However, these detectors often fail to replicate such success in real-world scenarios. We uncover two key factors contributing to this discrepancy through extensive experiments. First, a \\textit{Fine-tuning Gap}, where models leverage \\textit{Dataset-Specific Optimization} (DSO) paradigm for one domain at the cost of reduced effectiveness in others, leads to inflated performances on academic benchmarks. Second, the suboptimal performance in practical settings is primarily attributed to the"},"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":"2505.15649","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-21T15:26:46Z","cross_cats_sorted":[],"title_canon_sha256":"e1c611db1ff1fa6799b95f08f9a6e7cbae38e461b67f476a96bd4d8fcde599aa","abstract_canon_sha256":"c3909b1c19ad11dc345b5a61c87c54ea77b52393ff2e221ebeb6b0f609451c88"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:49.021849Z","signature_b64":"xSl21bcPM6DvUIAkUS9YAoIEuPT49DNMLtgI4hgHThV8quM/sBos2N1/Pbx1MzfbTSutPJZKskdGzhp6XvAvBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0424177c37100e318e6aaad51559743a4e8ba1421cc16fd0573eaca7138f7f7b","last_reissued_at":"2026-07-05T11:06:49.021398Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:49.021398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Devil is in Fine-tuning and Long-tailed Problems:A New Benchmark for Scene Text Detection","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiahao Lyu, Tianjiao Cao, Weichao Zeng, Weimin Mu, Yu Zhou","submitted_at":"2025-05-21T15:26:46Z","abstract_excerpt":"Scene text detection has seen the emergence of high-performing methods that excel on academic benchmarks. However, these detectors often fail to replicate such success in real-world scenarios. We uncover two key factors contributing to this discrepancy through extensive experiments. First, a \\textit{Fine-tuning Gap}, where models leverage \\textit{Dataset-Specific Optimization} (DSO) paradigm for one domain at the cost of reduced effectiveness in others, leads to inflated performances on academic benchmarks. Second, the suboptimal performance in practical settings is primarily attributed to the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15649","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/2505.15649/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":"2505.15649","created_at":"2026-07-05T11:06:49.021457+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15649v1","created_at":"2026-07-05T11:06:49.021457+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15649","created_at":"2026-07-05T11:06:49.021457+00:00"},{"alias_kind":"pith_short_12","alias_value":"AQSBO7BXCAHD","created_at":"2026-07-05T11:06:49.021457+00:00"},{"alias_kind":"pith_short_16","alias_value":"AQSBO7BXCAHDDDTK","created_at":"2026-07-05T11:06:49.021457+00:00"},{"alias_kind":"pith_short_8","alias_value":"AQSBO7BX","created_at":"2026-07-05T11:06:49.021457+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14708","citing_title":"StyleTextGen: Style-Conditioned Multilingual Scene Text Generation","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ","json":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ.json","graph_json":"https://pith.science/api/pith-number/AQSBO7BXCAHDDDTKVLKRKWLUHJ/graph.json","events_json":"https://pith.science/api/pith-number/AQSBO7BXCAHDDDTKVLKRKWLUHJ/events.json","paper":"https://pith.science/paper/AQSBO7BX"},"agent_actions":{"view_html":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ","download_json":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ.json","view_paper":"https://pith.science/paper/AQSBO7BX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15649&json=true","fetch_graph":"https://pith.science/api/pith-number/AQSBO7BXCAHDDDTKVLKRKWLUHJ/graph.json","fetch_events":"https://pith.science/api/pith-number/AQSBO7BXCAHDDDTKVLKRKWLUHJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ/action/storage_attestation","attest_author":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ/action/author_attestation","sign_citation":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ/action/citation_signature","submit_replication":"https://pith.science/pith/AQSBO7BXCAHDDDTKVLKRKWLUHJ/action/replication_record"}},"created_at":"2026-07-05T11:06:49.021457+00:00","updated_at":"2026-07-05T11:06:49.021457+00:00"}