{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FUWAHLJHOK7FMZQVQU66Q62ENP","short_pith_number":"pith:FUWAHLJH","schema_version":"1.0","canonical_sha256":"2d2c03ad2772be566615853de87b446be3b067e5e4db9a44b9a691e3d79d5fc2","source":{"kind":"arxiv","id":"2607.09164","version":1},"attestation_state":"computed","paper":{"title":"What Pixels Are Enough? SEAMS: Sufficiency Saliency via MSE-Preservation Soft-Masks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Andrzej Grudzie\\'n, Arkadiusz Lewicki, Jacek Tabor, Karolina Pachota, {\\L}ukasz Struski, Magdalena Tr\\k{e}dowicz, Mateusz Jag{\\l}a","submitted_at":"2026-07-10T07:41:13Z","abstract_excerpt":"Saliency maps are most useful when they identify the image regions that are sufficient to preserve a model's behaviour. We introduce SEAMS, a sufficiency-based saliency method that directly optimises a soft mask using a preservation objective. Given a frozen differentiable model output, such as a class probability, CLS embedding, or token representation, SEAMS searches for a compact mask that preserves the selected output. The approach relies on a simple optimisation framework based on soft masks, a learnable budget, and a three-way image composite generated entirely from the query image. As a"},"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":"2607.09164","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-10T07:41:13Z","cross_cats_sorted":[],"title_canon_sha256":"9ea8b029f0228209b214ba7eaea24196eb2bb16e8f83a61fec2d63aa28d35e7f","abstract_canon_sha256":"303fa725b738ea18fb895b6caeecf57d1ba98a7fc995e19188bb5283fb05f082"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-13T01:18:47.798097Z","signature_b64":"FZeG03aH41+aoMdCyec97fSUdZNgE4kEpDUmvM+AByMHnSRiHWBqctN3yUAN57Oya1KyU96RPOh9c7q/0uXsCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d2c03ad2772be566615853de87b446be3b067e5e4db9a44b9a691e3d79d5fc2","last_reissued_at":"2026-07-13T01:18:47.795979Z","signature_status":"signed_v1","first_computed_at":"2026-07-13T01:18:47.795979Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Pixels Are Enough? SEAMS: Sufficiency Saliency via MSE-Preservation Soft-Masks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Andrzej Grudzie\\'n, Arkadiusz Lewicki, Jacek Tabor, Karolina Pachota, {\\L}ukasz Struski, Magdalena Tr\\k{e}dowicz, Mateusz Jag{\\l}a","submitted_at":"2026-07-10T07:41:13Z","abstract_excerpt":"Saliency maps are most useful when they identify the image regions that are sufficient to preserve a model's behaviour. We introduce SEAMS, a sufficiency-based saliency method that directly optimises a soft mask using a preservation objective. Given a frozen differentiable model output, such as a class probability, CLS embedding, or token representation, SEAMS searches for a compact mask that preserves the selected output. The approach relies on a simple optimisation framework based on soft masks, a learnable budget, and a three-way image composite generated entirely from the query image. As a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.09164","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/2607.09164/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":"2607.09164","created_at":"2026-07-13T01:18:47.797512+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.09164v1","created_at":"2026-07-13T01:18:47.797512+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.09164","created_at":"2026-07-13T01:18:47.797512+00:00"},{"alias_kind":"pith_short_12","alias_value":"FUWAHLJHOK7F","created_at":"2026-07-13T01:18:47.797512+00:00"},{"alias_kind":"pith_short_16","alias_value":"FUWAHLJHOK7FMZQV","created_at":"2026-07-13T01:18:47.797512+00:00"},{"alias_kind":"pith_short_8","alias_value":"FUWAHLJH","created_at":"2026-07-13T01:18:47.797512+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/FUWAHLJHOK7FMZQVQU66Q62ENP","json":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP.json","graph_json":"https://pith.science/api/pith-number/FUWAHLJHOK7FMZQVQU66Q62ENP/graph.json","events_json":"https://pith.science/api/pith-number/FUWAHLJHOK7FMZQVQU66Q62ENP/events.json","paper":"https://pith.science/paper/FUWAHLJH"},"agent_actions":{"view_html":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP","download_json":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP.json","view_paper":"https://pith.science/paper/FUWAHLJH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.09164&json=true","fetch_graph":"https://pith.science/api/pith-number/FUWAHLJHOK7FMZQVQU66Q62ENP/graph.json","fetch_events":"https://pith.science/api/pith-number/FUWAHLJHOK7FMZQVQU66Q62ENP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP/action/storage_attestation","attest_author":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP/action/author_attestation","sign_citation":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP/action/citation_signature","submit_replication":"https://pith.science/pith/FUWAHLJHOK7FMZQVQU66Q62ENP/action/replication_record"}},"created_at":"2026-07-13T01:18:47.797512+00:00","updated_at":"2026-07-13T01:18:47.797512+00:00"}