{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6GRWJT5YACCEET7O63TU7KTVBV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"bca6ad25013db74fa5f3ca07a1af9b3f14b36d4f9f5c2b1c4828dc4add42e335","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-07T13:45:37Z","title_canon_sha256":"e4392de7f8550d57ae5355fcfd2f370b2d111f19fda7e6ebb29670ab8b1a43d5"},"schema_version":"1.0","source":{"id":"2605.06270","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06270","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06270v2","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06270","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_12","alias_value":"6GRWJT5YACCE","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_16","alias_value":"6GRWJT5YACCEET7O","created_at":"2026-05-20T01:05:15Z"},{"alias_kind":"pith_short_8","alias_value":"6GRWJT5Y","created_at":"2026-05-20T01:05:15Z"}],"graph_snapshots":[{"event_id":"sha256:f215487890d0f4e233face8f3b3cca529f9ffa71a513ad174842278734db59eb","target":"graph","created_at":"2026-05-20T01:05:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining."}],"snapshot_sha256":"1b52a6f3622165dbca41dd4b57a5e92e09236db79c62c5e425b5f200faa43aa3"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2605.06270/integrity.json","findings":[],"snapshot_sha256":"69fa156b166f40e5eba7fa76e121be7e56b08211501b85eeb40f5a39ecdc900b","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Haijie Li, Jian Zhang, Jiaqi Zhang, Jiaye Fu, Qiankun Gao, Siwei Ma, Yanmin Wu, Zecheng Tang","cross_cats":[],"headline":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining.","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-07T13:45:37Z","title":"Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.06270","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T13:36:27.852912Z","id":"ab2ef0a5-f0b5-477e-aaa2-f566ddaa820d","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"Asymmetric compression of query versus key-value tokens speeds feed-forward 3D reconstruction up to 28 times without retraining.","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.","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."}},"verdict_id":"ab2ef0a5-f0b5-477e-aaa2-f566ddaa820d"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a27a5481a2b696514a849312a0e85f295fc5baf554161fe69d6f2c82a8331f69","target":"record","created_at":"2026-05-20T01:05:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"bca6ad25013db74fa5f3ca07a1af9b3f14b36d4f9f5c2b1c4828dc4add42e335","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-07T13:45:37Z","title_canon_sha256":"e4392de7f8550d57ae5355fcfd2f370b2d111f19fda7e6ebb29670ab8b1a43d5"},"schema_version":"1.0","source":{"id":"2605.06270","kind":"arxiv","version":2}},"canonical_sha256":"f1a364cfb80084424feef6e74faa750d4352b8eba4eca8562a796148763f3f21","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f1a364cfb80084424feef6e74faa750d4352b8eba4eca8562a796148763f3f21","first_computed_at":"2026-05-20T01:05:15.580584Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:15.580584Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7lLNYlSwv0oWfXzbIwBPCiy6U8fSRza405m+0MtadV2wOW+LYtxgw9wrLM2YB0B+VRLhAYKAuvgBbaALQOIMDg==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:15.581256Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.06270","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a27a5481a2b696514a849312a0e85f295fc5baf554161fe69d6f2c82a8331f69","sha256:f215487890d0f4e233face8f3b3cca529f9ffa71a513ad174842278734db59eb"],"state_sha256":"54214cc89a462e5b48212edd3833acca999c3b0c6d4daee943eff80e788fce01"}