{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JFZQSWPIVNCHV4WKXPJ5UWAGFW","short_pith_number":"pith:JFZQSWPI","schema_version":"1.0","canonical_sha256":"49730959e8ab447af2cabbd3da58062d899643b7b6cf0398453492ebd321f745","source":{"kind":"arxiv","id":"2502.15885","version":1},"attestation_state":"computed","paper":{"title":"DOEI: Dual Optimization of Embedding Information for Attention-Enhanced Class Activation Maps","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Daji Ergu, Guansong Pang, Hongjie Zhu, Shimin Wen, Xu Wang, Yang Zhao, Ying Cai, Yu Bai, Zeyu Zhang","submitted_at":"2025-02-21T19:06:01Z","abstract_excerpt":"Weakly supervised semantic segmentation (WSSS) typically utilizes limited semantic annotations to obtain initial Class Activation Maps (CAMs). However, due to the inadequate coupling between class activation responses and semantic information in high-dimensional space, the CAM is prone to object co-occurrence or under-activation, resulting in inferior recognition accuracy. To tackle this issue, we propose DOEI, Dual Optimization of Embedding Information, a novel approach that reconstructs embedding representations through semantic-aware attention weight matrices to optimize the expression capa"},"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":"2502.15885","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-21T19:06:01Z","cross_cats_sorted":[],"title_canon_sha256":"16682748e5117be18538d993348f9462ca8c207ea5ae01205e3bd4c9b4a355da","abstract_canon_sha256":"d642f111c896cf4298b8674dbd61a112aa5cf16aa909017af8e1416cae867f85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:17.815482Z","signature_b64":"0hyas9PAFkHt8knyiGZEkxAKhsDY8VkfGMvYejXHjbHcWlTJ/w1WPqx3CFPiQCzMjq9UXvl12B6DqePOvlRODw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"49730959e8ab447af2cabbd3da58062d899643b7b6cf0398453492ebd321f745","last_reissued_at":"2026-07-05T10:18:17.815069Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:17.815069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DOEI: Dual Optimization of Embedding Information for Attention-Enhanced Class Activation Maps","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Daji Ergu, Guansong Pang, Hongjie Zhu, Shimin Wen, Xu Wang, Yang Zhao, Ying Cai, Yu Bai, Zeyu Zhang","submitted_at":"2025-02-21T19:06:01Z","abstract_excerpt":"Weakly supervised semantic segmentation (WSSS) typically utilizes limited semantic annotations to obtain initial Class Activation Maps (CAMs). However, due to the inadequate coupling between class activation responses and semantic information in high-dimensional space, the CAM is prone to object co-occurrence or under-activation, resulting in inferior recognition accuracy. To tackle this issue, we propose DOEI, Dual Optimization of Embedding Information, a novel approach that reconstructs embedding representations through semantic-aware attention weight matrices to optimize the expression capa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15885","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/2502.15885/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":"2502.15885","created_at":"2026-07-05T10:18:17.815126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15885v1","created_at":"2026-07-05T10:18:17.815126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15885","created_at":"2026-07-05T10:18:17.815126+00:00"},{"alias_kind":"pith_short_12","alias_value":"JFZQSWPIVNCH","created_at":"2026-07-05T10:18:17.815126+00:00"},{"alias_kind":"pith_short_16","alias_value":"JFZQSWPIVNCHV4WK","created_at":"2026-07-05T10:18:17.815126+00:00"},{"alias_kind":"pith_short_8","alias_value":"JFZQSWPI","created_at":"2026-07-05T10:18:17.815126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17451","citing_title":"SegTTA: Training-Free Test-Time Augmentation for Zero-Shot Medical Imaging Segmentation","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW","json":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW.json","graph_json":"https://pith.science/api/pith-number/JFZQSWPIVNCHV4WKXPJ5UWAGFW/graph.json","events_json":"https://pith.science/api/pith-number/JFZQSWPIVNCHV4WKXPJ5UWAGFW/events.json","paper":"https://pith.science/paper/JFZQSWPI"},"agent_actions":{"view_html":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW","download_json":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW.json","view_paper":"https://pith.science/paper/JFZQSWPI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15885&json=true","fetch_graph":"https://pith.science/api/pith-number/JFZQSWPIVNCHV4WKXPJ5UWAGFW/graph.json","fetch_events":"https://pith.science/api/pith-number/JFZQSWPIVNCHV4WKXPJ5UWAGFW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW/action/storage_attestation","attest_author":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW/action/author_attestation","sign_citation":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW/action/citation_signature","submit_replication":"https://pith.science/pith/JFZQSWPIVNCHV4WKXPJ5UWAGFW/action/replication_record"}},"created_at":"2026-07-05T10:18:17.815126+00:00","updated_at":"2026-07-05T10:18:17.815126+00:00"}