{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4TLZJRLWNFDL2FEAFJ6JJCWU3Y","short_pith_number":"pith:4TLZJRLW","schema_version":"1.0","canonical_sha256":"e4d794c5766946bd14802a7c948ad4de244ed06b7367277c8ea630a3d519f5da","source":{"kind":"arxiv","id":"2606.27372","version":1},"attestation_state":"computed","paper":{"title":"DnA: Denoising Attention for Visual Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aritra Dutta, Ron Campos, Srijan Das, Subhajit Maity, Xin Li","submitted_at":"2026-06-25T17:59:48Z","abstract_excerpt":"The softmax activation in multihead attention (MHA) is the de facto standard for attention-based models in visual perception tasks. However, standard softmax can produce noisy attention patterns that dilute relevant features and degrade its performance. In this paper, we propose Denoising Attention or DnA, in which, first, a positive query identifies which image features belong to the correct class, and a negative query identifies closely associated but irrelevant image features. DnA then projects these interactions into two distinct subspaces with larger principal angles, promoting subspace s"},"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":"2606.27372","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-25T17:59:48Z","cross_cats_sorted":[],"title_canon_sha256":"dadd069161ae6c77ed583e0c4d153c51b79d7c56e7b4ac5f8b88f82cb61956c5","abstract_canon_sha256":"389db34d1714fbc24e14fe1889b9fce7b7deb749d287b6d88467ad327f40e106"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:16:20.158787Z","signature_b64":"XMV7byX7GTQ+qT8nOR8BaqIjeqRMSqEJhQ+AduwXhci576vuVa8bxNZzLxb/t2qiwbrlPfH6Oh9DTEyeG1u3AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4d794c5766946bd14802a7c948ad4de244ed06b7367277c8ea630a3d519f5da","last_reissued_at":"2026-06-26T01:16:20.158385Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:16:20.158385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DnA: Denoising Attention for Visual Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aritra Dutta, Ron Campos, Srijan Das, Subhajit Maity, Xin Li","submitted_at":"2026-06-25T17:59:48Z","abstract_excerpt":"The softmax activation in multihead attention (MHA) is the de facto standard for attention-based models in visual perception tasks. However, standard softmax can produce noisy attention patterns that dilute relevant features and degrade its performance. In this paper, we propose Denoising Attention or DnA, in which, first, a positive query identifies which image features belong to the correct class, and a negative query identifies closely associated but irrelevant image features. DnA then projects these interactions into two distinct subspaces with larger principal angles, promoting subspace s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.27372","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/2606.27372/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":"2606.27372","created_at":"2026-06-26T01:16:20.158449+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.27372v1","created_at":"2026-06-26T01:16:20.158449+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.27372","created_at":"2026-06-26T01:16:20.158449+00:00"},{"alias_kind":"pith_short_12","alias_value":"4TLZJRLWNFDL","created_at":"2026-06-26T01:16:20.158449+00:00"},{"alias_kind":"pith_short_16","alias_value":"4TLZJRLWNFDL2FEA","created_at":"2026-06-26T01:16:20.158449+00:00"},{"alias_kind":"pith_short_8","alias_value":"4TLZJRLW","created_at":"2026-06-26T01:16:20.158449+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/4TLZJRLWNFDL2FEAFJ6JJCWU3Y","json":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y.json","graph_json":"https://pith.science/api/pith-number/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/graph.json","events_json":"https://pith.science/api/pith-number/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/events.json","paper":"https://pith.science/paper/4TLZJRLW"},"agent_actions":{"view_html":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y","download_json":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y.json","view_paper":"https://pith.science/paper/4TLZJRLW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.27372&json=true","fetch_graph":"https://pith.science/api/pith-number/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/graph.json","fetch_events":"https://pith.science/api/pith-number/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/action/storage_attestation","attest_author":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/action/author_attestation","sign_citation":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/action/citation_signature","submit_replication":"https://pith.science/pith/4TLZJRLWNFDL2FEAFJ6JJCWU3Y/action/replication_record"}},"created_at":"2026-06-26T01:16:20.158449+00:00","updated_at":"2026-06-26T01:16:20.158449+00:00"}