{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7LF56GBIFFTYG5PCGULKWNGNFR","short_pith_number":"pith:7LF56GBI","schema_version":"1.0","canonical_sha256":"facbdf182829678375e23516ab34cd2c67bd4168d6ae7678ef1200474d9eb753","source":{"kind":"arxiv","id":"2504.00476","version":1},"attestation_state":"computed","paper":{"title":"4th PVUW MeViS 3rd Place Report: Sa2VA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Haobo Yuan, Jiashi Feng, Lu Qi, Ming-Hsuan Yang, Shilin Xu, Tao Zhang, Xiangtai Li, Zilong Huang","submitted_at":"2025-04-01T07:06:47Z","abstract_excerpt":"Referring video object segmentation (RVOS) is a challenging task that requires the model to segment the object in a video given the language description. MeViS is a recently proposed dataset that contains motion expressions of the target objects, leading to a challenging benchmark, compared with existing RVOS benchmarks. On the other hand, for referring expression tasks, a new trend is to adopt multi-modal large language model (MLLM) to achieve better image and text alignment. In this report, we show that with a simple modification to the test time inference method on stronger MLLMs, we can le"},"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":"2504.00476","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-01T07:06:47Z","cross_cats_sorted":[],"title_canon_sha256":"876c5841c9c64bab8a302a7a4c17825273d185aa839c49005709e7e68eb91dbe","abstract_canon_sha256":"4e5f355862d34493456232f8de5e6c4a71733b40187746cf6c67c143c7ad1548"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:42:27.132760Z","signature_b64":"bFfHuDMsSD4mCRiYzGZo5JXXNc3aQyYxsfJYbToM8/ZNje8PuEWcnb35mnVEqIiBJAig/iyYbovGLqGHqsrADg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"facbdf182829678375e23516ab34cd2c67bd4168d6ae7678ef1200474d9eb753","last_reissued_at":"2026-07-05T10:42:27.132292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:42:27.132292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"4th PVUW MeViS 3rd Place Report: Sa2VA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Haobo Yuan, Jiashi Feng, Lu Qi, Ming-Hsuan Yang, Shilin Xu, Tao Zhang, Xiangtai Li, Zilong Huang","submitted_at":"2025-04-01T07:06:47Z","abstract_excerpt":"Referring video object segmentation (RVOS) is a challenging task that requires the model to segment the object in a video given the language description. MeViS is a recently proposed dataset that contains motion expressions of the target objects, leading to a challenging benchmark, compared with existing RVOS benchmarks. On the other hand, for referring expression tasks, a new trend is to adopt multi-modal large language model (MLLM) to achieve better image and text alignment. In this report, we show that with a simple modification to the test time inference method on stronger MLLMs, we can le"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.00476","kind":"arxiv","version":1},"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/2504.00476/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.00476","created_at":"2026-07-05T10:42:27.132362+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.00476v1","created_at":"2026-07-05T10:42:27.132362+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.00476","created_at":"2026-07-05T10:42:27.132362+00:00"},{"alias_kind":"pith_short_12","alias_value":"7LF56GBIFFTY","created_at":"2026-07-05T10:42:27.132362+00:00"},{"alias_kind":"pith_short_16","alias_value":"7LF56GBIFFTYG5PC","created_at":"2026-07-05T10:42:27.132362+00:00"},{"alias_kind":"pith_short_8","alias_value":"7LF56GBI","created_at":"2026-07-05T10:42:27.132362+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.14471","citing_title":"Dense360: Dense Understanding from Omnidirectional Panoramas","ref_index":66,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR","json":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR.json","graph_json":"https://pith.science/api/pith-number/7LF56GBIFFTYG5PCGULKWNGNFR/graph.json","events_json":"https://pith.science/api/pith-number/7LF56GBIFFTYG5PCGULKWNGNFR/events.json","paper":"https://pith.science/paper/7LF56GBI"},"agent_actions":{"view_html":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR","download_json":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR.json","view_paper":"https://pith.science/paper/7LF56GBI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.00476&json=true","fetch_graph":"https://pith.science/api/pith-number/7LF56GBIFFTYG5PCGULKWNGNFR/graph.json","fetch_events":"https://pith.science/api/pith-number/7LF56GBIFFTYG5PCGULKWNGNFR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR/action/storage_attestation","attest_author":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR/action/author_attestation","sign_citation":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR/action/citation_signature","submit_replication":"https://pith.science/pith/7LF56GBIFFTYG5PCGULKWNGNFR/action/replication_record"}},"created_at":"2026-07-05T10:42:27.132362+00:00","updated_at":"2026-07-05T10:42:27.132362+00:00"}