{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6GRWJT5YACCEET7O63TU7KTVBV","short_pith_number":"pith:6GRWJT5Y","schema_version":"1.0","canonical_sha256":"f1a364cfb80084424feef6e74faa750d4352b8eba4eca8562a796148763f3f21","source":{"kind":"arxiv","id":"2605.06270","version":2},"attestation_state":"computed","paper":{"title":"Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Haijie Li, Jian Zhang, Jiaqi Zhang, Jiaye Fu, Qiankun Gao, Siwei Ma, Yanmin Wu, Zecheng Tang","submitted_at":"2026-05-07T13:45:37Z","abstract_excerpt":"Feed-forward 3D reconstruction models based on Vision Transformers can directly estimate scene geometry and camera poses from a small set of input images, but scaling them to video inputs with hundreds or thousands of frames remains challenging due to the quadratic cost of global attention layers. Recent token-merging methods accelerate these models by compressing the token sequence within the global attention layers, but they apply a uniform reduction to query tokens and key-value tokens, ignoring their functionally distinct roles in 3D reconstruction. In this work, we identify a key property"},"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":"2605.06270","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-07T13:45:37Z","cross_cats_sorted":[],"title_canon_sha256":"e4392de7f8550d57ae5355fcfd2f370b2d111f19fda7e6ebb29670ab8b1a43d5","abstract_canon_sha256":"bca6ad25013db74fa5f3ca07a1af9b3f14b36d4f9f5c2b1c4828dc4add42e335"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:15.581256Z","signature_b64":"7lLNYlSwv0oWfXzbIwBPCiy6U8fSRza405m+0MtadV2wOW+LYtxgw9wrLM2YB0B+VRLhAYKAuvgBbaALQOIMDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1a364cfb80084424feef6e74faa750d4352b8eba4eca8562a796148763f3f21","last_reissued_at":"2026-05-20T01:05:15.580584Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:15.580584Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Haijie Li, Jian Zhang, Jiaqi Zhang, Jiaye Fu, Qiankun Gao, Siwei Ma, Yanmin Wu, Zecheng Tang","submitted_at":"2026-05-07T13:45:37Z","abstract_excerpt":"Feed-forward 3D reconstruction models based on Vision Transformers can directly estimate scene geometry and camera poses from a small set of input images, but scaling them to video inputs with hundreds or thousands of frames remains challenging due to the quadratic cost of global attention layers. Recent token-merging methods accelerate these models by compressing the token sequence within the global attention layers, but they apply a uniform reduction to query tokens and key-value tokens, ignoring their functionally distinct roles in 3D reconstruction. In this work, we identify a key property"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Spark3R achieves up to 28× speedup on 1,000-frame inputs while maintaining competitive reconstruction quality by decoupling the compression of query tokens and key-value tokens with intra-group merging and lightweight pruning.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That query tokens encode view-specific geometric requests sensitive to compression while key-value tokens represent shared scene context tolerant to aggressive compression, and that this functional distinction generalizes across pretrained models and input sequences without retraining.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Asymmetric token reduction, with distinct merging for queries and pruning for key-values plus layer-wise adaptation, delivers up to 28x speedup on 1000-frame 3D reconstruction inputs while preserving competitive quality.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1b52a6f3622165dbca41dd4b57a5e92e09236db79c62c5e425b5f200faa43aa3"},"source":{"id":"2605.06270","kind":"arxiv","version":2},"verdict":{"id":"ab2ef0a5-f0b5-477e-aaa2-f566ddaa820d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T13:36:27.852912Z","strongest_claim":"Spark3R achieves up to 28× speedup on 1,000-frame inputs while maintaining competitive reconstruction quality by decoupling the compression of query tokens and key-value tokens with intra-group merging and lightweight pruning.","one_line_summary":"Asymmetric token reduction, with distinct merging for queries and pruning for key-values plus layer-wise adaptation, delivers up to 28x speedup on 1000-frame 3D reconstruction inputs while preserving competitive quality.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That query tokens encode view-specific geometric requests sensitive to compression while key-value tokens represent shared scene context tolerant to aggressive compression, and that this functional distinction generalizes across pretrained models and input sequences without retraining.","pith_extraction_headline":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.06270/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T18:31:19.611942Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:50:09.833483Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"69fa156b166f40e5eba7fa76e121be7e56b08211501b85eeb40f5a39ecdc900b"},"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":"2605.06270","created_at":"2026-05-20T01:05:15.580678+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.06270v2","created_at":"2026-05-20T01:05:15.580678+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06270","created_at":"2026-05-20T01:05:15.580678+00:00"},{"alias_kind":"pith_short_12","alias_value":"6GRWJT5YACCE","created_at":"2026-05-20T01:05:15.580678+00:00"},{"alias_kind":"pith_short_16","alias_value":"6GRWJT5YACCEET7O","created_at":"2026-05-20T01:05:15.580678+00:00"},{"alias_kind":"pith_short_8","alias_value":"6GRWJT5Y","created_at":"2026-05-20T01:05:15.580678+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/6GRWJT5YACCEET7O63TU7KTVBV","json":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV.json","graph_json":"https://pith.science/api/pith-number/6GRWJT5YACCEET7O63TU7KTVBV/graph.json","events_json":"https://pith.science/api/pith-number/6GRWJT5YACCEET7O63TU7KTVBV/events.json","paper":"https://pith.science/paper/6GRWJT5Y"},"agent_actions":{"view_html":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV","download_json":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV.json","view_paper":"https://pith.science/paper/6GRWJT5Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.06270&json=true","fetch_graph":"https://pith.science/api/pith-number/6GRWJT5YACCEET7O63TU7KTVBV/graph.json","fetch_events":"https://pith.science/api/pith-number/6GRWJT5YACCEET7O63TU7KTVBV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV/action/storage_attestation","attest_author":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV/action/author_attestation","sign_citation":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV/action/citation_signature","submit_replication":"https://pith.science/pith/6GRWJT5YACCEET7O63TU7KTVBV/action/replication_record"}},"created_at":"2026-05-20T01:05:15.580678+00:00","updated_at":"2026-05-20T01:05:15.580678+00:00"}