{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZV6PA7WWZVZVKRG27277SO5OMK","short_pith_number":"pith:ZV6PA7WW","canonical_record":{"source":{"id":"2503.21817","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-26T04:16:48Z","cross_cats_sorted":[],"title_canon_sha256":"2df4422026c7ebc861ffa7315763170f9b615557cafc4e5d0fd9e5481c5044ab","abstract_canon_sha256":"a04b9e5983cff6ee95c29ff771de5916fab0e8cd12b5e344b250d20b21f26716"},"schema_version":"1.0"},"canonical_sha256":"cd7cf07ed6cd735544dafebff93bae6299cfafbc9d4488c13af8e4a4e8e5d192","source":{"kind":"arxiv","id":"2503.21817","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21817","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21817v3","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21817","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"pith_short_12","alias_value":"ZV6PA7WWZVZV","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"pith_short_16","alias_value":"ZV6PA7WWZVZVKRG2","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"pith_short_8","alias_value":"ZV6PA7WW","created_at":"2026-07-05T11:31:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZV6PA7WWZVZVKRG27277SO5OMK","target":"record","payload":{"canonical_record":{"source":{"id":"2503.21817","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-26T04:16:48Z","cross_cats_sorted":[],"title_canon_sha256":"2df4422026c7ebc861ffa7315763170f9b615557cafc4e5d0fd9e5481c5044ab","abstract_canon_sha256":"a04b9e5983cff6ee95c29ff771de5916fab0e8cd12b5e344b250d20b21f26716"},"schema_version":"1.0"},"canonical_sha256":"cd7cf07ed6cd735544dafebff93bae6299cfafbc9d4488c13af8e4a4e8e5d192","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:06.006909Z","signature_b64":"5cdZW30iuQXAlFT8Bwa0M9whtmAK+blikRD0cwAD6lnajnuUzh2S2y+Qlm0pqwTaVmrDW4/4oYpndP9JyoIvDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd7cf07ed6cd735544dafebff93bae6299cfafbc9d4488c13af8e4a4e8e5d192","last_reissued_at":"2026-07-05T11:31:06.006174Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:06.006174Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.21817","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:31:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LMl0RD7HPIxLCtYWCIIn6GEC/Yp8B4D293Scb8yo7JWFMV8u+vHyiv683Id78jcoGzOuJE+4XhtZgD3/11HWBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T07:38:02.737305Z"},"content_sha256":"67a871ed6f24006c2534b5121e10ba6e0daf146584e40d6ebee4ae16a50f846d","schema_version":"1.0","event_id":"sha256:67a871ed6f24006c2534b5121e10ba6e0daf146584e40d6ebee4ae16a50f846d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZV6PA7WWZVZVKRG27277SO5OMK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skipping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Kaixiang Ji, Weili Zeng, Yichao Yan, Ziyuan Huang","submitted_at":"2025-03-26T04:16:48Z","abstract_excerpt":"Transformer-based models have driven significant advancements in Multimodal Large Language Models (MLLMs), yet their computational costs surge drastically when scaling resolution, training data, and model parameters. A key bottleneck stems from the proliferation of visual tokens required for fine-grained image understanding. We propose Skip-Vision, a unified framework addressing both training and inference inefficiencies in vision-language models. On top of conventional token compression approaches, our method introduces two complementary acceleration strategies. For training acceleration, we "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21817","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/2503.21817/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:31:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mEE5+1830xMMiX5MWjBcuvrbByfvvbbxQxL2ntLxeIH6uuhFyBGJQycab4xDx0HIeXlaYp79HudlktQrlXQAAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T07:38:02.737798Z"},"content_sha256":"5cc19211cf60e133d129b48cf28aaae70934938c400496f698aab533d20292cb","schema_version":"1.0","event_id":"sha256:5cc19211cf60e133d129b48cf28aaae70934938c400496f698aab533d20292cb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZV6PA7WWZVZVKRG27277SO5OMK/bundle.json","state_url":"https://pith.science/pith/ZV6PA7WWZVZVKRG27277SO5OMK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZV6PA7WWZVZVKRG27277SO5OMK/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":"2026-08-18T07:38:02Z","links":{"resolver":"https://pith.science/pith/ZV6PA7WWZVZVKRG27277SO5OMK","bundle":"https://pith.science/pith/ZV6PA7WWZVZVKRG27277SO5OMK/bundle.json","state":"https://pith.science/pith/ZV6PA7WWZVZVKRG27277SO5OMK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZV6PA7WWZVZVKRG27277SO5OMK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZV6PA7WWZVZVKRG27277SO5OMK","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":"a04b9e5983cff6ee95c29ff771de5916fab0e8cd12b5e344b250d20b21f26716","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-26T04:16:48Z","title_canon_sha256":"2df4422026c7ebc861ffa7315763170f9b615557cafc4e5d0fd9e5481c5044ab"},"schema_version":"1.0","source":{"id":"2503.21817","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21817","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21817v3","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21817","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"pith_short_12","alias_value":"ZV6PA7WWZVZV","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"pith_short_16","alias_value":"ZV6PA7WWZVZVKRG2","created_at":"2026-07-05T11:31:06Z"},{"alias_kind":"pith_short_8","alias_value":"ZV6PA7WW","created_at":"2026-07-05T11:31:06Z"}],"graph_snapshots":[{"event_id":"sha256:5cc19211cf60e133d129b48cf28aaae70934938c400496f698aab533d20292cb","target":"graph","created_at":"2026-07-05T11:31:06Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2503.21817/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformer-based models have driven significant advancements in Multimodal Large Language Models (MLLMs), yet their computational costs surge drastically when scaling resolution, training data, and model parameters. A key bottleneck stems from the proliferation of visual tokens required for fine-grained image understanding. We propose Skip-Vision, a unified framework addressing both training and inference inefficiencies in vision-language models. On top of conventional token compression approaches, our method introduces two complementary acceleration strategies. For training acceleration, we ","authors_text":"Kaixiang Ji, Weili Zeng, Yichao Yan, Ziyuan Huang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-26T04:16:48Z","title":"Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skipping"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21817","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:67a871ed6f24006c2534b5121e10ba6e0daf146584e40d6ebee4ae16a50f846d","target":"record","created_at":"2026-07-05T11:31:06Z","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":"a04b9e5983cff6ee95c29ff771de5916fab0e8cd12b5e344b250d20b21f26716","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-26T04:16:48Z","title_canon_sha256":"2df4422026c7ebc861ffa7315763170f9b615557cafc4e5d0fd9e5481c5044ab"},"schema_version":"1.0","source":{"id":"2503.21817","kind":"arxiv","version":3}},"canonical_sha256":"cd7cf07ed6cd735544dafebff93bae6299cfafbc9d4488c13af8e4a4e8e5d192","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cd7cf07ed6cd735544dafebff93bae6299cfafbc9d4488c13af8e4a4e8e5d192","first_computed_at":"2026-07-05T11:31:06.006174Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:31:06.006174Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5cdZW30iuQXAlFT8Bwa0M9whtmAK+blikRD0cwAD6lnajnuUzh2S2y+Qlm0pqwTaVmrDW4/4oYpndP9JyoIvDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:31:06.006909Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.21817","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67a871ed6f24006c2534b5121e10ba6e0daf146584e40d6ebee4ae16a50f846d","sha256:5cc19211cf60e133d129b48cf28aaae70934938c400496f698aab533d20292cb"],"state_sha256":"61c08c7006fd09efefb8789e61db2bcebc224b0060e68a54a1102c7469a4885c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8hGvr19boWyb7WMdSk7aZABlGnolBMMwvZ5iFkYbAxb0+Rx4+06DO+W630javTw5nWEPT8c6FPKcvfqTMtEACA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T07:38:02.742300Z","bundle_sha256":"d32da9cdbef549c71224b737b1fbc7227969bda6367c7389ec7706b25d26c3e1"}}