{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:L5PHGKUGLHW2SM4GITEDY5YX3M","short_pith_number":"pith:L5PHGKUG","canonical_record":{"source":{"id":"2604.25886","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2026-04-28T17:29:19Z","cross_cats_sorted":[],"title_canon_sha256":"51d8b39789f4318fafd2b6043e90646d262748a203019096625715b049668428","abstract_canon_sha256":"c8bdc0b57a9b6a5da6d69776a0134df8a9fe9bc4aeefa42a95e808315190d409"},"schema_version":"1.0"},"canonical_sha256":"5f5e732a8659eda9338644c83c7717db26355e2d99a64bb28a688d0e9bda3051","source":{"kind":"arxiv","id":"2604.25886","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.25886","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"arxiv_version","alias_value":"2604.25886v3","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.25886","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"pith_short_12","alias_value":"L5PHGKUGLHW2","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"pith_short_16","alias_value":"L5PHGKUGLHW2SM4G","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"pith_short_8","alias_value":"L5PHGKUG","created_at":"2026-06-19T16:09:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:L5PHGKUGLHW2SM4GITEDY5YX3M","target":"record","payload":{"canonical_record":{"source":{"id":"2604.25886","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2026-04-28T17:29:19Z","cross_cats_sorted":[],"title_canon_sha256":"51d8b39789f4318fafd2b6043e90646d262748a203019096625715b049668428","abstract_canon_sha256":"c8bdc0b57a9b6a5da6d69776a0134df8a9fe9bc4aeefa42a95e808315190d409"},"schema_version":"1.0"},"canonical_sha256":"5f5e732a8659eda9338644c83c7717db26355e2d99a64bb28a688d0e9bda3051","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:09:58.630122Z","signature_b64":"shJcmqIRqUABgUu+TMqZ2GGeSKu3jDdCsn5WlfSQp/eweoKC22S7MIFHAEp+5aulDhHDaUI00zec93EtXherDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f5e732a8659eda9338644c83c7717db26355e2d99a64bb28a688d0e9bda3051","last_reissued_at":"2026-06-19T16:09:58.629736Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:09:58.629736Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.25886","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":"2026-06-19T16:09:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6jcRhCEplhuVhFOAv4/WYyZYacUrEJninB+25q2CMg8k0nYk5gsMW2CIQZ9edzWB4RE8VQxUrO2Bcm5ES8tKBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T00:22:22.891036Z"},"content_sha256":"46077d38d8c920f9bc4dc9d46304473656033bdcf6c11e9f75e7a8f7f4367d2f","schema_version":"1.0","event_id":"sha256:46077d38d8c920f9bc4dc9d46304473656033bdcf6c11e9f75e7a8f7f4367d2f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:L5PHGKUGLHW2SM4GITEDY5YX3M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MarkIt: Training-Free Visual Markers for Precise Video Temporal Grounding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"MarkIt converts videos into query-marked versions with instance masks and frame indices to let Vid-LLMs output more accurate start and end times.","cross_cats":[],"primary_cat":"cs.MM","authors_text":"Pengcheng Fang, Xiaohao Cai, Yuxia Chen","submitted_at":"2026-04-28T17:29:19Z","abstract_excerpt":"Video temporal grounding (VTG) aims to localize the start and end timestamps of the event described by a given query within an untrimmed video. Despite the strong open-world video understanding and recognition ability of video language large models (Vid-LLMs), outputting precise temporal grounding information remains challenging, since explicit temporal cues are scarce in untrimmed videos, and query-relevant entities are hard to track consistently across the video timeline. In this paper, we present \\MarkIt{}, a training-free framework that transforms an input video into a query-conditioned ma"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"MarkIt achieves state-of-the-art results on multiple mainstream moment retrieval and highlight detection benchmarks, delivering consistent temporal grounding improvements across a wide range of existing models.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The annotation-free query-to-mask grounding bridge can reliably extract canonical subject tags via linguistic parsing and map them to accurate instance masks using open-vocabulary segmentation without introducing errors that mislead the Vid-LLM.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MarkIt uses a query-to-mask bridge with open-vocabulary segmentation to add visual markers and frame indices to videos, enabling Vid-LLMs to achieve state-of-the-art temporal grounding on moment retrieval and highlight detection benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"MarkIt converts videos into query-marked versions with instance masks and frame indices to let Vid-LLMs output more accurate start and end times.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"36558f638236f5f73373f2636faed8dfe04abca71311924f129e186b1db57f3c"},"source":{"id":"2604.25886","kind":"arxiv","version":3},"verdict":{"id":"ced1ec09-602d-4732-a0c0-395e5696db51","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T14:04:14.961772Z","strongest_claim":"MarkIt achieves state-of-the-art results on multiple mainstream moment retrieval and highlight detection benchmarks, delivering consistent temporal grounding improvements across a wide range of existing models.","one_line_summary":"MarkIt uses a query-to-mask bridge with open-vocabulary segmentation to add visual markers and frame indices to videos, enabling Vid-LLMs to achieve state-of-the-art temporal grounding on moment retrieval and highlight detection benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The annotation-free query-to-mask grounding bridge can reliably extract canonical subject tags via linguistic parsing and map them to accurate instance masks using open-vocabulary segmentation without introducing errors that mislead the Vid-LLM.","pith_extraction_headline":"MarkIt converts videos into query-marked versions with instance masks and frame indices to let Vid-LLMs output more accurate start and end times."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.25886/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T03:39:07.429305Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:37:10.203654Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"f2d114bc65ed4ea9a99edc2e17e64ccdc88d18fab66f50f5453a82649c6e2049"},"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":"ced1ec09-602d-4732-a0c0-395e5696db51"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-19T16:09:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f8+blq6btlIDEs7BJWpd45uO+dLYZV88VT0/QBZh5oZ93YVst74IE/pgPfNMU1bPKiUoASApXMVJSY3m2vcUBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T00:22:22.891966Z"},"content_sha256":"77c91fe06fecd033fb79deb5f2597b2351a8898cf13dc457fc8c83cda5db7205","schema_version":"1.0","event_id":"sha256:77c91fe06fecd033fb79deb5f2597b2351a8898cf13dc457fc8c83cda5db7205"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L5PHGKUGLHW2SM4GITEDY5YX3M/bundle.json","state_url":"https://pith.science/pith/L5PHGKUGLHW2SM4GITEDY5YX3M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L5PHGKUGLHW2SM4GITEDY5YX3M/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":"2026-08-06T00:22:22Z","links":{"resolver":"https://pith.science/pith/L5PHGKUGLHW2SM4GITEDY5YX3M","bundle":"https://pith.science/pith/L5PHGKUGLHW2SM4GITEDY5YX3M/bundle.json","state":"https://pith.science/pith/L5PHGKUGLHW2SM4GITEDY5YX3M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L5PHGKUGLHW2SM4GITEDY5YX3M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:L5PHGKUGLHW2SM4GITEDY5YX3M","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":"c8bdc0b57a9b6a5da6d69776a0134df8a9fe9bc4aeefa42a95e808315190d409","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2026-04-28T17:29:19Z","title_canon_sha256":"51d8b39789f4318fafd2b6043e90646d262748a203019096625715b049668428"},"schema_version":"1.0","source":{"id":"2604.25886","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.25886","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"arxiv_version","alias_value":"2604.25886v3","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.25886","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"pith_short_12","alias_value":"L5PHGKUGLHW2","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"pith_short_16","alias_value":"L5PHGKUGLHW2SM4G","created_at":"2026-06-19T16:09:58Z"},{"alias_kind":"pith_short_8","alias_value":"L5PHGKUG","created_at":"2026-06-19T16:09:58Z"}],"graph_snapshots":[{"event_id":"sha256:77c91fe06fecd033fb79deb5f2597b2351a8898cf13dc457fc8c83cda5db7205","target":"graph","created_at":"2026-06-19T16:09:58Z","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":"MarkIt achieves state-of-the-art results on multiple mainstream moment retrieval and highlight detection benchmarks, delivering consistent temporal grounding improvements across a wide range of existing models."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The annotation-free query-to-mask grounding bridge can reliably extract canonical subject tags via linguistic parsing and map them to accurate instance masks using open-vocabulary segmentation without introducing errors that mislead the Vid-LLM."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"MarkIt uses a query-to-mask bridge with open-vocabulary segmentation to add visual markers and frame indices to videos, enabling Vid-LLMs to achieve state-of-the-art temporal grounding on moment retrieval and highlight detection benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"MarkIt converts videos into query-marked versions with instance masks and frame indices to let Vid-LLMs output more accurate start and end times."}],"snapshot_sha256":"36558f638236f5f73373f2636faed8dfe04abca71311924f129e186b1db57f3c"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-21T03:39:07.429305Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T20:37:10.203654Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.25886/integrity.json","findings":[],"snapshot_sha256":"f2d114bc65ed4ea9a99edc2e17e64ccdc88d18fab66f50f5453a82649c6e2049","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Video temporal grounding (VTG) aims to localize the start and end timestamps of the event described by a given query within an untrimmed video. Despite the strong open-world video understanding and recognition ability of video language large models (Vid-LLMs), outputting precise temporal grounding information remains challenging, since explicit temporal cues are scarce in untrimmed videos, and query-relevant entities are hard to track consistently across the video timeline. In this paper, we present \\MarkIt{}, a training-free framework that transforms an input video into a query-conditioned ma","authors_text":"Pengcheng Fang, Xiaohao Cai, Yuxia Chen","cross_cats":[],"headline":"MarkIt converts videos into query-marked versions with instance masks and frame indices to let Vid-LLMs output more accurate start and end times.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2026-04-28T17:29:19Z","title":"MarkIt: Training-Free Visual Markers for Precise Video Temporal Grounding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.25886","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-07T14:04:14.961772Z","id":"ced1ec09-602d-4732-a0c0-395e5696db51","model_set":{"reader":"grok-4.3"},"one_line_summary":"MarkIt uses a query-to-mask bridge with open-vocabulary segmentation to add visual markers and frame indices to videos, enabling Vid-LLMs to achieve state-of-the-art temporal grounding on moment retrieval and highlight detection benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"MarkIt converts videos into query-marked versions with instance masks and frame indices to let Vid-LLMs output more accurate start and end times.","strongest_claim":"MarkIt achieves state-of-the-art results on multiple mainstream moment retrieval and highlight detection benchmarks, delivering consistent temporal grounding improvements across a wide range of existing models.","weakest_assumption":"The annotation-free query-to-mask grounding bridge can reliably extract canonical subject tags via linguistic parsing and map them to accurate instance masks using open-vocabulary segmentation without introducing errors that mislead the Vid-LLM."}},"verdict_id":"ced1ec09-602d-4732-a0c0-395e5696db51"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:46077d38d8c920f9bc4dc9d46304473656033bdcf6c11e9f75e7a8f7f4367d2f","target":"record","created_at":"2026-06-19T16:09:58Z","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":"c8bdc0b57a9b6a5da6d69776a0134df8a9fe9bc4aeefa42a95e808315190d409","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MM","submitted_at":"2026-04-28T17:29:19Z","title_canon_sha256":"51d8b39789f4318fafd2b6043e90646d262748a203019096625715b049668428"},"schema_version":"1.0","source":{"id":"2604.25886","kind":"arxiv","version":3}},"canonical_sha256":"5f5e732a8659eda9338644c83c7717db26355e2d99a64bb28a688d0e9bda3051","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5f5e732a8659eda9338644c83c7717db26355e2d99a64bb28a688d0e9bda3051","first_computed_at":"2026-06-19T16:09:58.629736Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:09:58.629736Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"shJcmqIRqUABgUu+TMqZ2GGeSKu3jDdCsn5WlfSQp/eweoKC22S7MIFHAEp+5aulDhHDaUI00zec93EtXherDA==","signature_status":"signed_v1","signed_at":"2026-06-19T16:09:58.630122Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.25886","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:46077d38d8c920f9bc4dc9d46304473656033bdcf6c11e9f75e7a8f7f4367d2f","sha256:77c91fe06fecd033fb79deb5f2597b2351a8898cf13dc457fc8c83cda5db7205"],"state_sha256":"53fcfc4dbdd27567aab0ca89914336ba518e630c580822e969dc087990d14e2b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hY4qya447//ie2Z5Of9ytVFKk/6D7x3lrGnQ7MigG64mj0HkkrmlBeEOOkLVmxUpn9V1hOLm2rm2nBcZvkRLAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T00:22:22.899799Z","bundle_sha256":"4d61618521122ba437fcae852bec7f7447fa59d535cbb2f121379da1844f8482"}}