{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5TYKRYZ5DDBA6A7AJQFI4EO4AW","short_pith_number":"pith:5TYKRYZ5","schema_version":"1.0","canonical_sha256":"ecf0a8e33d18c20f03e04c0a8e11dc05b5ca3782786a23ca07650317bfccc83c","source":{"kind":"arxiv","id":"2404.19128","version":1},"attestation_state":"computed","paper":{"title":"Q-GroundCAM: Quantifying Grounding in Vision Language Models via GradCAM","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jana Kosecka, Navid Rajabi","submitted_at":"2024-04-29T22:06:17Z","abstract_excerpt":"Vision and Language Models (VLMs) continue to demonstrate remarkable zero-shot (ZS) performance across various tasks. However, many probing studies have revealed that even the best-performing VLMs struggle to capture aspects of compositional scene understanding, lacking the ability to properly ground and localize linguistic phrases in images. Recent VLM advancements include scaling up both model and dataset sizes, additional training objectives and levels of supervision, and variations in the model architectures. To characterize the grounding ability of VLMs, such as phrase grounding, referrin"},"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":"2404.19128","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-29T22:06:17Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"1ab5b59b633bf8d63bbd86a6212e6a8b110513643fcb42c3510406d96bbb92dc","abstract_canon_sha256":"5943675ddc0760097e52b11ca22879262142e65c246ab96793364d4b0550768d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:13:37.959744Z","signature_b64":"NkQ4nNgn9P5SP0vi2EAV52wlhd56ZgKAbt3doY9+MeGvcn7RW36pLDKFnzGxkp4wwxl2zYQ65wSM8v9xq8O0BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ecf0a8e33d18c20f03e04c0a8e11dc05b5ca3782786a23ca07650317bfccc83c","last_reissued_at":"2026-07-05T08:13:37.959256Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:13:37.959256Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Q-GroundCAM: Quantifying Grounding in Vision Language Models via GradCAM","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jana Kosecka, Navid Rajabi","submitted_at":"2024-04-29T22:06:17Z","abstract_excerpt":"Vision and Language Models (VLMs) continue to demonstrate remarkable zero-shot (ZS) performance across various tasks. However, many probing studies have revealed that even the best-performing VLMs struggle to capture aspects of compositional scene understanding, lacking the ability to properly ground and localize linguistic phrases in images. Recent VLM advancements include scaling up both model and dataset sizes, additional training objectives and levels of supervision, and variations in the model architectures. To characterize the grounding ability of VLMs, such as phrase grounding, referrin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.19128","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/2404.19128/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":"2404.19128","created_at":"2026-07-05T08:13:37.959322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.19128v1","created_at":"2026-07-05T08:13:37.959322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.19128","created_at":"2026-07-05T08:13:37.959322+00:00"},{"alias_kind":"pith_short_12","alias_value":"5TYKRYZ5DDBA","created_at":"2026-07-05T08:13:37.959322+00:00"},{"alias_kind":"pith_short_16","alias_value":"5TYKRYZ5DDBA6A7A","created_at":"2026-07-05T08:13:37.959322+00:00"},{"alias_kind":"pith_short_8","alias_value":"5TYKRYZ5","created_at":"2026-07-05T08:13:37.959322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW","json":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW.json","graph_json":"https://pith.science/api/pith-number/5TYKRYZ5DDBA6A7AJQFI4EO4AW/graph.json","events_json":"https://pith.science/api/pith-number/5TYKRYZ5DDBA6A7AJQFI4EO4AW/events.json","paper":"https://pith.science/paper/5TYKRYZ5"},"agent_actions":{"view_html":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW","download_json":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW.json","view_paper":"https://pith.science/paper/5TYKRYZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.19128&json=true","fetch_graph":"https://pith.science/api/pith-number/5TYKRYZ5DDBA6A7AJQFI4EO4AW/graph.json","fetch_events":"https://pith.science/api/pith-number/5TYKRYZ5DDBA6A7AJQFI4EO4AW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW/action/storage_attestation","attest_author":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW/action/author_attestation","sign_citation":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW/action/citation_signature","submit_replication":"https://pith.science/pith/5TYKRYZ5DDBA6A7AJQFI4EO4AW/action/replication_record"}},"created_at":"2026-07-05T08:13:37.959322+00:00","updated_at":"2026-07-05T08:13:37.959322+00:00"}