{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KOVJDIIDHITDYWKSRKJH7KCCZ5","short_pith_number":"pith:KOVJDIID","schema_version":"1.0","canonical_sha256":"53aa91a1033a263c59528a927fa842cf5a28a0d288c91069c67a8f4118a77257","source":{"kind":"arxiv","id":"2505.15516","version":3},"attestation_state":"computed","paper":{"title":"Explainable embeddings with Distance Explainer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Christiaan Meijer, E. G. Patrick Bos","submitted_at":"2025-05-21T13:42:28Z","abstract_excerpt":"While eXplainable AI (XAI) has advanced significantly, few methods address interpretability in embedded vector spaces where dimensions represent complex abstractions. We introduce Distance Explainer, a novel method for generating local, post-hoc explanations of embedded spaces in machine learning models. Our approach adapts saliency-based techniques from RISE to explain the distance between two embedded data points by assigning attribution values through selective masking and distance-ranked mask filtering. We evaluate Distance Explainer on cross-modal embeddings (image-image and image-caption"},"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":"2505.15516","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-21T13:42:28Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV"],"title_canon_sha256":"81f89d0ed2864de2daa41c72dd918f7d9fbfe6130f7b404082109011bd5bc1a0","abstract_canon_sha256":"d40565fcb412a04c24542e276b79ef025d3b3d81e28e67b429c4f8f6882d58c3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:19:05.354929Z","signature_b64":"FUYfpz/fqNvwNL6Nu9WmCSukXJId8Jy3OO9lxqfTAJxwP5hFtq3mb+4UbEe/p50Jzxg6QmkkQZOLNkKjb98KBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53aa91a1033a263c59528a927fa842cf5a28a0d288c91069c67a8f4118a77257","last_reissued_at":"2026-07-08T01:19:05.354372Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:19:05.354372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Explainable embeddings with Distance Explainer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Christiaan Meijer, E. G. Patrick Bos","submitted_at":"2025-05-21T13:42:28Z","abstract_excerpt":"While eXplainable AI (XAI) has advanced significantly, few methods address interpretability in embedded vector spaces where dimensions represent complex abstractions. We introduce Distance Explainer, a novel method for generating local, post-hoc explanations of embedded spaces in machine learning models. Our approach adapts saliency-based techniques from RISE to explain the distance between two embedded data points by assigning attribution values through selective masking and distance-ranked mask filtering. We evaluate Distance Explainer on cross-modal embeddings (image-image and image-caption"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15516","kind":"arxiv","version":3},"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/2505.15516/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":"2505.15516","created_at":"2026-07-08T01:19:05.354442+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15516v3","created_at":"2026-07-08T01:19:05.354442+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15516","created_at":"2026-07-08T01:19:05.354442+00:00"},{"alias_kind":"pith_short_12","alias_value":"KOVJDIIDHITD","created_at":"2026-07-08T01:19:05.354442+00:00"},{"alias_kind":"pith_short_16","alias_value":"KOVJDIIDHITDYWKS","created_at":"2026-07-08T01:19:05.354442+00:00"},{"alias_kind":"pith_short_8","alias_value":"KOVJDIID","created_at":"2026-07-08T01:19:05.354442+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.30836","citing_title":"Partition-Guided Distance Saliency: Bridging Decision and Objective Spaces in Many-Objective Optimization","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5","json":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5.json","graph_json":"https://pith.science/api/pith-number/KOVJDIIDHITDYWKSRKJH7KCCZ5/graph.json","events_json":"https://pith.science/api/pith-number/KOVJDIIDHITDYWKSRKJH7KCCZ5/events.json","paper":"https://pith.science/paper/KOVJDIID"},"agent_actions":{"view_html":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5","download_json":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5.json","view_paper":"https://pith.science/paper/KOVJDIID","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15516&json=true","fetch_graph":"https://pith.science/api/pith-number/KOVJDIIDHITDYWKSRKJH7KCCZ5/graph.json","fetch_events":"https://pith.science/api/pith-number/KOVJDIIDHITDYWKSRKJH7KCCZ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5/action/storage_attestation","attest_author":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5/action/author_attestation","sign_citation":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5/action/citation_signature","submit_replication":"https://pith.science/pith/KOVJDIIDHITDYWKSRKJH7KCCZ5/action/replication_record"}},"created_at":"2026-07-08T01:19:05.354442+00:00","updated_at":"2026-07-08T01:19:05.354442+00:00"}