{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SMKQB5RWDZSFISOEFZZ6OCCE5Z","short_pith_number":"pith:SMKQB5RW","schema_version":"1.0","canonical_sha256":"931500f6361e645449c42e73e70844ee707acc9c8c6ccb4b2c87c1d0e894e214","source":{"kind":"arxiv","id":"2505.18764","version":2},"attestation_state":"computed","paper":{"title":"Efficient Differentiable Hardware Rasterization for 3D Gaussian Splatting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.GR","authors_text":"Qianyue He, Yitian Yuan","submitted_at":"2025-05-24T16:07:33Z","abstract_excerpt":"Recent works demonstrate the advantages of hardware rasterization for 3D Gaussian Splatting (3DGS) in forward-pass rendering through fast GPU-optimized graphics and fixed memory footprint. However, extending these benefits to backward-pass gradient computation remains challenging due to graphics pipeline constraints. We present a differentiable hardware rasterizer for 3DGS that overcomes the memory and performance limitations of tile-based software rasterization. Our solution employs programmable blending for per-pixel gradient computation combined with a hybrid gradient reduction strategy (qu"},"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":"2505.18764","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.GR","submitted_at":"2025-05-24T16:07:33Z","cross_cats_sorted":[],"title_canon_sha256":"d2447bd6e37251f90e78a8e61ebe38562014690d390a173cfeb0430c61f6873f","abstract_canon_sha256":"3abb6117af1bc60315f0c1d06c61365407730fc879f49e1bde0f1e0842a4af52"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:03.537779Z","signature_b64":"nAyUAuWtmda4HGuJd9sDvit/JXkqZRqryF2bf/EVb0J8K0SaoZMkrEz8Gsh7S+EUNFVuIOJVoBQdpHCaJoDBCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"931500f6361e645449c42e73e70844ee707acc9c8c6ccb4b2c87c1d0e894e214","last_reissued_at":"2026-07-05T11:53:03.537346Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:03.537346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Differentiable Hardware Rasterization for 3D Gaussian Splatting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.GR","authors_text":"Qianyue He, Yitian Yuan","submitted_at":"2025-05-24T16:07:33Z","abstract_excerpt":"Recent works demonstrate the advantages of hardware rasterization for 3D Gaussian Splatting (3DGS) in forward-pass rendering through fast GPU-optimized graphics and fixed memory footprint. However, extending these benefits to backward-pass gradient computation remains challenging due to graphics pipeline constraints. We present a differentiable hardware rasterizer for 3DGS that overcomes the memory and performance limitations of tile-based software rasterization. Our solution employs programmable blending for per-pixel gradient computation combined with a hybrid gradient reduction strategy (qu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18764","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/2505.18764/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":"2505.18764","created_at":"2026-07-05T11:53:03.537400+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18764v2","created_at":"2026-07-05T11:53:03.537400+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18764","created_at":"2026-07-05T11:53:03.537400+00:00"},{"alias_kind":"pith_short_12","alias_value":"SMKQB5RWDZSF","created_at":"2026-07-05T11:53:03.537400+00:00"},{"alias_kind":"pith_short_16","alias_value":"SMKQB5RWDZSFISOE","created_at":"2026-07-05T11:53:03.537400+00:00"},{"alias_kind":"pith_short_8","alias_value":"SMKQB5RW","created_at":"2026-07-05T11:53:03.537400+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00352","citing_title":"HiGS: A Hierarchical Rendering Architecture for Real-Time 3D Gaussian Splatting","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00219","citing_title":"VkSplat: High-Performance 3DGS Training in Vulkan Compute","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z","json":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z.json","graph_json":"https://pith.science/api/pith-number/SMKQB5RWDZSFISOEFZZ6OCCE5Z/graph.json","events_json":"https://pith.science/api/pith-number/SMKQB5RWDZSFISOEFZZ6OCCE5Z/events.json","paper":"https://pith.science/paper/SMKQB5RW"},"agent_actions":{"view_html":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z","download_json":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z.json","view_paper":"https://pith.science/paper/SMKQB5RW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18764&json=true","fetch_graph":"https://pith.science/api/pith-number/SMKQB5RWDZSFISOEFZZ6OCCE5Z/graph.json","fetch_events":"https://pith.science/api/pith-number/SMKQB5RWDZSFISOEFZZ6OCCE5Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z/action/storage_attestation","attest_author":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z/action/author_attestation","sign_citation":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z/action/citation_signature","submit_replication":"https://pith.science/pith/SMKQB5RWDZSFISOEFZZ6OCCE5Z/action/replication_record"}},"created_at":"2026-07-05T11:53:03.537400+00:00","updated_at":"2026-07-05T11:53:03.537400+00:00"}