{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:N7ACWPVARFXD7P2HVU3DQILU2P","short_pith_number":"pith:N7ACWPVA","schema_version":"1.0","canonical_sha256":"6fc02b3ea0896e3fbf47ad36382174d3e169b7822fc980ba30a6d207c1a04e1b","source":{"kind":"arxiv","id":"2503.06671","version":2},"attestation_state":"computed","paper":{"title":"Emulating Self-attention with Convolution for Efficient Image Super-Resolution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongheon Lee, Seokju Yun, Youngmin Ro","submitted_at":"2025-03-09T15:45:53Z","abstract_excerpt":"In this paper, we tackle the high computational overhead of Transformers for efficient image super-resolution~(SR). Motivated by the observations of self-attention's inter-layer repetition, we introduce a convolutionized self-attention module named Convolutional Attention~(ConvAttn) that emulates self-attention's long-range modeling capability and instance-dependent weighting with a single shared large kernel and dynamic kernels. By utilizing the ConvAttn module, we significantly reduce the reliance on self-attention and its involved memory-bound operations while maintaining the representation"},"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":"2503.06671","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-09T15:45:53Z","cross_cats_sorted":[],"title_canon_sha256":"15dd5889104cf78691f738d059bbbf25f66c760f9ff1499c22cca9e0b8c8c047","abstract_canon_sha256":"489e45c26010c6a82a743e7ae118def30e4f7c381098b8f934afc2c4d925d9f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:29:08.047206Z","signature_b64":"GqHX/xxQTgIfniU5tfZUuEjY+RjpcQ0RCIQN1g8DI0YdFjBs4y7JLVXlOFs4/VndEunGf5Sb8UChaDr3LO1lCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6fc02b3ea0896e3fbf47ad36382174d3e169b7822fc980ba30a6d207c1a04e1b","last_reissued_at":"2026-07-05T11:29:08.046720Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:29:08.046720Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Emulating Self-attention with Convolution for Efficient Image Super-Resolution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongheon Lee, Seokju Yun, Youngmin Ro","submitted_at":"2025-03-09T15:45:53Z","abstract_excerpt":"In this paper, we tackle the high computational overhead of Transformers for efficient image super-resolution~(SR). Motivated by the observations of self-attention's inter-layer repetition, we introduce a convolutionized self-attention module named Convolutional Attention~(ConvAttn) that emulates self-attention's long-range modeling capability and instance-dependent weighting with a single shared large kernel and dynamic kernels. By utilizing the ConvAttn module, we significantly reduce the reliance on self-attention and its involved memory-bound operations while maintaining the representation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06671","kind":"arxiv","version":2},"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/2503.06671/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":"2503.06671","created_at":"2026-07-05T11:29:08.046778+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.06671v2","created_at":"2026-07-05T11:29:08.046778+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06671","created_at":"2026-07-05T11:29:08.046778+00:00"},{"alias_kind":"pith_short_12","alias_value":"N7ACWPVARFXD","created_at":"2026-07-05T11:29:08.046778+00:00"},{"alias_kind":"pith_short_16","alias_value":"N7ACWPVARFXD7P2H","created_at":"2026-07-05T11:29:08.046778+00:00"},{"alias_kind":"pith_short_8","alias_value":"N7ACWPVA","created_at":"2026-07-05T11:29:08.046778+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.11680","citing_title":"UCAN: Unified Convolutional Attention Network for Expansive Receptive Fields in Lightweight Super-Resolution","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P","json":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P.json","graph_json":"https://pith.science/api/pith-number/N7ACWPVARFXD7P2HVU3DQILU2P/graph.json","events_json":"https://pith.science/api/pith-number/N7ACWPVARFXD7P2HVU3DQILU2P/events.json","paper":"https://pith.science/paper/N7ACWPVA"},"agent_actions":{"view_html":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P","download_json":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P.json","view_paper":"https://pith.science/paper/N7ACWPVA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.06671&json=true","fetch_graph":"https://pith.science/api/pith-number/N7ACWPVARFXD7P2HVU3DQILU2P/graph.json","fetch_events":"https://pith.science/api/pith-number/N7ACWPVARFXD7P2HVU3DQILU2P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P/action/storage_attestation","attest_author":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P/action/author_attestation","sign_citation":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P/action/citation_signature","submit_replication":"https://pith.science/pith/N7ACWPVARFXD7P2HVU3DQILU2P/action/replication_record"}},"created_at":"2026-07-05T11:29:08.046778+00:00","updated_at":"2026-07-05T11:29:08.046778+00:00"}