{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BEAVHY7MFN75GFPRLTBHPTTICD","short_pith_number":"pith:BEAVHY7M","schema_version":"1.0","canonical_sha256":"090153e3ec2b7fd315f15cc277ce6810dc51c3b23f151cda0b99447ccb7424b1","source":{"kind":"arxiv","id":"2601.15275","version":3},"attestation_state":"computed","paper":{"title":"RayRoPE: Projective Ray Positional Encoding for Multi-view Attention","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jen-Hao Rick Chang, Minsik Jeon, Oncel Tuzel, Shubham Tulsiani, Yu Wu","submitted_at":"2026-01-21T18:55:51Z","abstract_excerpt":"We study positional encodings for multi-view transformers that process tokens from a set of posed input images, and seek a mechanism that encodes patches uniquely, allows SE(3)-invariant attention with multi-frequency similarity, and can adapt to the geometry of the underlying 3D scene. We find that prior (absolute or relative) encoding schemes for multi-view attention do not meet these desiderata, and present RayRoPE to address this gap. RayRoPE represents patch positions based on associated rays and computes query-frame projective coordinates to ensure SE(3) invariance. To adapt to scene geo"},"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":"2601.15275","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-01-21T18:55:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e0337153c02deffc0c68f2e61736a13db00ba37263ea06e0f194cfc7a9ebb786","abstract_canon_sha256":"75ae67b0ca82dcbdbe44fdd1837eadc16ea3b783f59eec3bd43ae92afa395932"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:24.619090Z","signature_b64":"jbLYRlhYzVPr/WumjGLO4cXNcJKMtKtuPCYjEculFQAIlALkNhsmC6YEgyq87mfqkFXdYBVbFCLjWSY90yxuDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"090153e3ec2b7fd315f15cc277ce6810dc51c3b23f151cda0b99447ccb7424b1","last_reissued_at":"2026-07-08T01:18:24.618392Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:24.618392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RayRoPE: Projective Ray Positional Encoding for Multi-view Attention","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jen-Hao Rick Chang, Minsik Jeon, Oncel Tuzel, Shubham Tulsiani, Yu Wu","submitted_at":"2026-01-21T18:55:51Z","abstract_excerpt":"We study positional encodings for multi-view transformers that process tokens from a set of posed input images, and seek a mechanism that encodes patches uniquely, allows SE(3)-invariant attention with multi-frequency similarity, and can adapt to the geometry of the underlying 3D scene. We find that prior (absolute or relative) encoding schemes for multi-view attention do not meet these desiderata, and present RayRoPE to address this gap. RayRoPE represents patch positions based on associated rays and computes query-frame projective coordinates to ensure SE(3) invariance. To adapt to scene geo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.15275","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/2601.15275/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":"2601.15275","created_at":"2026-07-08T01:18:24.618565+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.15275v3","created_at":"2026-07-08T01:18:24.618565+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.15275","created_at":"2026-07-08T01:18:24.618565+00:00"},{"alias_kind":"pith_short_12","alias_value":"BEAVHY7MFN75","created_at":"2026-07-08T01:18:24.618565+00:00"},{"alias_kind":"pith_short_16","alias_value":"BEAVHY7MFN75GFPR","created_at":"2026-07-08T01:18:24.618565+00:00"},{"alias_kind":"pith_short_8","alias_value":"BEAVHY7M","created_at":"2026-07-08T01:18:24.618565+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":6,"sample":[{"citing_arxiv_id":"2606.11275","citing_title":"RoVE: Rotary Value Embeddings Attention for Relative Position-dependent Value Pathways","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2607.00417","citing_title":"EO-VGGT: Orbital Ray-Conditioned 3D Foundation Models for Satellite Multi-View Reconstruction","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01590","citing_title":"Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2606.31585","citing_title":"DPPE: Rethinking Camera-Based Positional Encoding for Scaling Multi-View Transformers","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12938","citing_title":"CRePE: Curved Ray Expectation Positional Encoding for Unified-Camera-Controlled Video Generation","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2604.18747","citing_title":"URoPE: Universal Relative Position Embedding across Geometric Spaces","ref_index":36,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD","json":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD.json","graph_json":"https://pith.science/api/pith-number/BEAVHY7MFN75GFPRLTBHPTTICD/graph.json","events_json":"https://pith.science/api/pith-number/BEAVHY7MFN75GFPRLTBHPTTICD/events.json","paper":"https://pith.science/paper/BEAVHY7M"},"agent_actions":{"view_html":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD","download_json":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD.json","view_paper":"https://pith.science/paper/BEAVHY7M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.15275&json=true","fetch_graph":"https://pith.science/api/pith-number/BEAVHY7MFN75GFPRLTBHPTTICD/graph.json","fetch_events":"https://pith.science/api/pith-number/BEAVHY7MFN75GFPRLTBHPTTICD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD/action/storage_attestation","attest_author":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD/action/author_attestation","sign_citation":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD/action/citation_signature","submit_replication":"https://pith.science/pith/BEAVHY7MFN75GFPRLTBHPTTICD/action/replication_record"}},"created_at":"2026-07-08T01:18:24.618565+00:00","updated_at":"2026-07-08T01:18:24.618565+00:00"}