{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:GQ7AFIA2KIS563MEOYXCERTLKJ","short_pith_number":"pith:GQ7AFIA2","canonical_record":{"source":{"id":"2509.02560","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T17:54:21Z","cross_cats_sorted":[],"title_canon_sha256":"775dee16a98c51d2dca3277c76c502537f8c521554eeb5ce1246ba918d1be400","abstract_canon_sha256":"b5183e4a8ae8fe8f06347e2d74565b2c0c9067442ce470bb14af321b1d5f9d74"},"schema_version":"1.0"},"canonical_sha256":"343e02a01a5225df6d84762e22466b52617a475fbcaf461a2972c81236a1a4b4","source":{"kind":"arxiv","id":"2509.02560","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.02560","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"arxiv_version","alias_value":"2509.02560v2","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02560","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"pith_short_12","alias_value":"GQ7AFIA2KIS5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"GQ7AFIA2KIS563ME","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"GQ7AFIA2","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:GQ7AFIA2KIS563MEOYXCERTLKJ","target":"record","payload":{"canonical_record":{"source":{"id":"2509.02560","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T17:54:21Z","cross_cats_sorted":[],"title_canon_sha256":"775dee16a98c51d2dca3277c76c502537f8c521554eeb5ce1246ba918d1be400","abstract_canon_sha256":"b5183e4a8ae8fe8f06347e2d74565b2c0c9067442ce470bb14af321b1d5f9d74"},"schema_version":"1.0"},"canonical_sha256":"343e02a01a5225df6d84762e22466b52617a475fbcaf461a2972c81236a1a4b4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:49.752619Z","signature_b64":"6BJv8XeoB39XZT0+VlmWo2+QkzxZPYFzYu1RhlmB1uUjky/iZmNf4E2YpnhEMFC1e4xREbr6+8MZ9DeWhfNtCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"343e02a01a5225df6d84762e22466b52617a475fbcaf461a2972c81236a1a4b4","last_reissued_at":"2026-05-17T23:38:49.752098Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:49.752098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.02560","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kie/N1Nesc4AH/tYC51SyA2Gh1nIwbKxSxezZJbED8ZZx7oPi1UtXmKm6yChObdg2tuicz+DBdMRee5X3k3iCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T13:51:45.522276Z"},"content_sha256":"c34ca3e34fb195cb6011a5f99e290810dc705ed6f6e59835493c5677d9fcf3d0","schema_version":"1.0","event_id":"sha256:c34ca3e34fb195cb6011a5f99e290810dc705ed6f6e59835493c5677d9fcf3d0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:GQ7AFIA2KIS563MEOYXCERTLKJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FastVGGT: Training-Free Acceleration of Visual Geometry Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A 3D-specific token partitioning strategy lets token merging accelerate VGGT fourfold on thousand-image sequences without retraining.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiayi Ji, Liujuan Cao, Shengchuan Zhang, Xiawu Zheng, Yansong Qu, You Shen, Zhipeng Zhang","submitted_at":"2025-09-02T17:54:21Z","abstract_excerpt":"Foundation models for 3D vision have recently demonstrated remarkable capabilities in 3D perception. However, scaling these models to long-sequence image inputs remains a significant challenge due to inference-time inefficiency. In this work, we present a detailed analysis of VGGT, a state-of-the-art feed-forward visual geometry model and identify its primary bottleneck. Visualization further reveals a token collapse phenomenon in the attention maps. Motivated by these findings, we explore the potential of token merging in the feed-forward visual geometry model. Owing to the unique architectur"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Notably, with 1000 input images, FastVGGT achieves a 4x speedup over VGGT while mitigating error accumulation in long-sequence scenarios.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the newly devised 3D-specific token partitioning strategy can remove redundant computation while fully preserving VGGT's reconstruction capacity, even though directly applying existing merging techniques is stated to be challenging due to architectural and task-specific properties.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"FastVGGT achieves 4x speedup on VGGT for 1000-image inputs using training-free token merging tailored to 3D architectures while reducing error accumulation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A 3D-specific token partitioning strategy lets token merging accelerate VGGT fourfold on thousand-image sequences without retraining.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3d10fba8fe40e5c7ecbf125beace250215d7db342aa9ac59f305d10df01611b1"},"source":{"id":"2509.02560","kind":"arxiv","version":2},"verdict":{"id":"fad21ddd-9b7c-410d-ba62-094860d8edd2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T23:32:44.742619Z","strongest_claim":"Notably, with 1000 input images, FastVGGT achieves a 4x speedup over VGGT while mitigating error accumulation in long-sequence scenarios.","one_line_summary":"FastVGGT achieves 4x speedup on VGGT for 1000-image inputs using training-free token merging tailored to 3D architectures while reducing error accumulation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the newly devised 3D-specific token partitioning strategy can remove redundant computation while fully preserving VGGT's reconstruction capacity, even though directly applying existing merging techniques is stated to be challenging due to architectural and task-specific properties.","pith_extraction_headline":"A 3D-specific token partitioning strategy lets token merging accelerate VGGT fourfold on thousand-image sequences without retraining."},"references":{"count":33,"sample":[{"doi":"","year":null,"title":"Token merging for fast sta- ble diffusion","work_id":"c2a6db5b-1e5d-46b2-bd17-02ef709ec2b4","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Token Merging: Your ViT But Faster","work_id":"528509bc-2611-4e7f-a772-ea14d25b6dae","ref_index":2,"cited_arxiv_id":"2210.09461","is_internal_anchor":true},{"doi":"","year":null,"title":"Pumer: Pruning and merging tokens for efficient vision language models.arXiv preprint arXiv:2305.17530,","work_id":"91361f0f-8572-41d1-a5eb-06f62fd5b464","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Emerg- ing properties in self-supervised vision transformers","work_id":"8a5a47d5-320f-431e-953a-e774104b82a0","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"vid-tldr: Training free token merging for light-weight video transformer","work_id":"aba45603-5ed7-40ed-9c87-a8fc5e7de13d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":33,"snapshot_sha256":"203434c8ade2d746ee1f4491a2948ffb92dbaba8e303131c946409924889260a","internal_anchors":8},"formal_canon":{"evidence_count":3,"snapshot_sha256":"25739931bf32a146934c67e37139adaf5374be9b2e37ecfe90b2541c0e4150c1"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"fad21ddd-9b7c-410d-ba62-094860d8edd2"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QaZ9BL+BxdnArrQS98n+7SU3VfCka3mkMr7/1qHk8fLz3DeiKa+EyKQtFCSAbvdMcbiqLrcnHXgYdR8x8lTBBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T13:51:45.522900Z"},"content_sha256":"ae0f5a33050c83e9a6c91f581694b1394db82158b026c69a3db099ca4d140b1a","schema_version":"1.0","event_id":"sha256:ae0f5a33050c83e9a6c91f581694b1394db82158b026c69a3db099ca4d140b1a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GQ7AFIA2KIS563MEOYXCERTLKJ/bundle.json","state_url":"https://pith.science/pith/GQ7AFIA2KIS563MEOYXCERTLKJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GQ7AFIA2KIS563MEOYXCERTLKJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-04T13:51:45Z","links":{"resolver":"https://pith.science/pith/GQ7AFIA2KIS563MEOYXCERTLKJ","bundle":"https://pith.science/pith/GQ7AFIA2KIS563MEOYXCERTLKJ/bundle.json","state":"https://pith.science/pith/GQ7AFIA2KIS563MEOYXCERTLKJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GQ7AFIA2KIS563MEOYXCERTLKJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:GQ7AFIA2KIS563MEOYXCERTLKJ","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":"b5183e4a8ae8fe8f06347e2d74565b2c0c9067442ce470bb14af321b1d5f9d74","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T17:54:21Z","title_canon_sha256":"775dee16a98c51d2dca3277c76c502537f8c521554eeb5ce1246ba918d1be400"},"schema_version":"1.0","source":{"id":"2509.02560","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.02560","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"arxiv_version","alias_value":"2509.02560v2","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02560","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"pith_short_12","alias_value":"GQ7AFIA2KIS5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"GQ7AFIA2KIS563ME","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"GQ7AFIA2","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:ae0f5a33050c83e9a6c91f581694b1394db82158b026c69a3db099ca4d140b1a","target":"graph","created_at":"2026-05-17T23:38:49Z","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":"Notably, with 1000 input images, FastVGGT achieves a 4x speedup over VGGT while mitigating error accumulation in long-sequence scenarios."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the newly devised 3D-specific token partitioning strategy can remove redundant computation while fully preserving VGGT's reconstruction capacity, even though directly applying existing merging techniques is stated to be challenging due to architectural and task-specific properties."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"FastVGGT achieves 4x speedup on VGGT for 1000-image inputs using training-free token merging tailored to 3D architectures while reducing error accumulation."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A 3D-specific token partitioning strategy lets token merging accelerate VGGT fourfold on thousand-image sequences without retraining."}],"snapshot_sha256":"3d10fba8fe40e5c7ecbf125beace250215d7db342aa9ac59f305d10df01611b1"},"formal_canon":{"evidence_count":3,"snapshot_sha256":"25739931bf32a146934c67e37139adaf5374be9b2e37ecfe90b2541c0e4150c1"},"paper":{"abstract_excerpt":"Foundation models for 3D vision have recently demonstrated remarkable capabilities in 3D perception. However, scaling these models to long-sequence image inputs remains a significant challenge due to inference-time inefficiency. In this work, we present a detailed analysis of VGGT, a state-of-the-art feed-forward visual geometry model and identify its primary bottleneck. Visualization further reveals a token collapse phenomenon in the attention maps. Motivated by these findings, we explore the potential of token merging in the feed-forward visual geometry model. Owing to the unique architectur","authors_text":"Jiayi Ji, Liujuan Cao, Shengchuan Zhang, Xiawu Zheng, Yansong Qu, You Shen, Zhipeng Zhang","cross_cats":[],"headline":"A 3D-specific token partitioning strategy lets token merging accelerate VGGT fourfold on thousand-image sequences without retraining.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T17:54:21Z","title":"FastVGGT: Training-Free Acceleration of Visual Geometry Transformer"},"references":{"count":33,"internal_anchors":8,"resolved_work":33,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Token merging for fast sta- ble diffusion","work_id":"c2a6db5b-1e5d-46b2-bd17-02ef709ec2b4","year":null},{"cited_arxiv_id":"2210.09461","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Token Merging: Your ViT But Faster","work_id":"528509bc-2611-4e7f-a772-ea14d25b6dae","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Pumer: Pruning and merging tokens for efficient vision language models.arXiv preprint arXiv:2305.17530,","work_id":"91361f0f-8572-41d1-a5eb-06f62fd5b464","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Emerg- ing properties in self-supervised vision transformers","work_id":"8a5a47d5-320f-431e-953a-e774104b82a0","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"vid-tldr: Training free token merging for light-weight video transformer","work_id":"aba45603-5ed7-40ed-9c87-a8fc5e7de13d","year":2024}],"snapshot_sha256":"203434c8ade2d746ee1f4491a2948ffb92dbaba8e303131c946409924889260a"},"source":{"id":"2509.02560","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-15T23:32:44.742619Z","id":"fad21ddd-9b7c-410d-ba62-094860d8edd2","model_set":{"reader":"grok-4.3"},"one_line_summary":"FastVGGT achieves 4x speedup on VGGT for 1000-image inputs using training-free token merging tailored to 3D architectures while reducing error accumulation.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A 3D-specific token partitioning strategy lets token merging accelerate VGGT fourfold on thousand-image sequences without retraining.","strongest_claim":"Notably, with 1000 input images, FastVGGT achieves a 4x speedup over VGGT while mitigating error accumulation in long-sequence scenarios.","weakest_assumption":"That the newly devised 3D-specific token partitioning strategy can remove redundant computation while fully preserving VGGT's reconstruction capacity, even though directly applying existing merging techniques is stated to be challenging due to architectural and task-specific properties."}},"verdict_id":"fad21ddd-9b7c-410d-ba62-094860d8edd2"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c34ca3e34fb195cb6011a5f99e290810dc705ed6f6e59835493c5677d9fcf3d0","target":"record","created_at":"2026-05-17T23:38:49Z","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":"b5183e4a8ae8fe8f06347e2d74565b2c0c9067442ce470bb14af321b1d5f9d74","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T17:54:21Z","title_canon_sha256":"775dee16a98c51d2dca3277c76c502537f8c521554eeb5ce1246ba918d1be400"},"schema_version":"1.0","source":{"id":"2509.02560","kind":"arxiv","version":2}},"canonical_sha256":"343e02a01a5225df6d84762e22466b52617a475fbcaf461a2972c81236a1a4b4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"343e02a01a5225df6d84762e22466b52617a475fbcaf461a2972c81236a1a4b4","first_computed_at":"2026-05-17T23:38:49.752098Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:49.752098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6BJv8XeoB39XZT0+VlmWo2+QkzxZPYFzYu1RhlmB1uUjky/iZmNf4E2YpnhEMFC1e4xREbr6+8MZ9DeWhfNtCA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:49.752619Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.02560","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c34ca3e34fb195cb6011a5f99e290810dc705ed6f6e59835493c5677d9fcf3d0","sha256:ae0f5a33050c83e9a6c91f581694b1394db82158b026c69a3db099ca4d140b1a"],"state_sha256":"6b4c7802be72002f0dfb47fde6bdc95c26f95d5d32c1e6986ec43a99fa0ac7d0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rgYevhqc8inWTll3+BwC7l67CRM9SFhKGl+co67OWeNXJMkbMV9iP9udRsqBaF6+f6+d7BCOSnaIBSmTfTwlAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T13:51:45.525555Z","bundle_sha256":"e6c5a71d7fde9d41ed1be3abe427755e83c2c52af35f3f39310015820f7ff3a1"}}