{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZU4JTVZ3VHXFA2QEDPRVE7F3KQ","short_pith_number":"pith:ZU4JTVZ3","schema_version":"1.0","canonical_sha256":"cd3899d73ba9ee506a041be3527cbb543a023ed4d5641421528a3ef0901b0bd6","source":{"kind":"arxiv","id":"2308.10509","version":2},"attestation_state":"computed","paper":{"title":"An Examination of the Compositionality of Large Generative Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Junwei Liang, Rong Li, Teli Ma","submitted_at":"2023-08-21T06:50:29Z","abstract_excerpt":"With the success of Large Language Models (LLMs), many Generative Vision-Language Models (GVLMs) have been constructed via multimodal instruction tuning. However, the performance of GVLMs in multimodal compositional reasoning remains under-explored. In this paper, we examine both the evaluation metrics (VisualGPTScore, etc.) and current benchmarks for evaluating the compositionality of GVLMs. We identify the syntactical bias in current benchmarks, which is exploited by the linguistic capability of GVLMs. The bias renders VisualGPTScore an insufficient metric for assessing GVLMs. To combat this"},"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":"2308.10509","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-21T06:50:29Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"1bad85c909bbe8a785cd7486437128ebc47a76ad5f692831479c57405f738d9f","abstract_canon_sha256":"565c9335a8ec8c8dcf7ef71d9f665398ebcf7e4643d5a20fe9d120a63126f917"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:21.108918Z","signature_b64":"3l+DKvGttCHNJfxhi0Tm6TyzAALYaexhJ6h4RxOQW1WI5d6IzTkt6zqvj7EVdXmCnG/5+UhTyloQ4xKhnEQbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd3899d73ba9ee506a041be3527cbb543a023ed4d5641421528a3ef0901b0bd6","last_reissued_at":"2026-07-05T08:02:21.108423Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:21.108423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Examination of the Compositionality of Large Generative Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Junwei Liang, Rong Li, Teli Ma","submitted_at":"2023-08-21T06:50:29Z","abstract_excerpt":"With the success of Large Language Models (LLMs), many Generative Vision-Language Models (GVLMs) have been constructed via multimodal instruction tuning. However, the performance of GVLMs in multimodal compositional reasoning remains under-explored. In this paper, we examine both the evaluation metrics (VisualGPTScore, etc.) and current benchmarks for evaluating the compositionality of GVLMs. We identify the syntactical bias in current benchmarks, which is exploited by the linguistic capability of GVLMs. The bias renders VisualGPTScore an insufficient metric for assessing GVLMs. To combat this"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.10509","kind":"arxiv","version":2},"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/2308.10509/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":"2308.10509","created_at":"2026-07-05T08:02:21.108477+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.10509v2","created_at":"2026-07-05T08:02:21.108477+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.10509","created_at":"2026-07-05T08:02:21.108477+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZU4JTVZ3VHXF","created_at":"2026-07-05T08:02:21.108477+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZU4JTVZ3VHXFA2QE","created_at":"2026-07-05T08:02:21.108477+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZU4JTVZ3","created_at":"2026-07-05T08:02:21.108477+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/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ","json":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ.json","graph_json":"https://pith.science/api/pith-number/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/graph.json","events_json":"https://pith.science/api/pith-number/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/events.json","paper":"https://pith.science/paper/ZU4JTVZ3"},"agent_actions":{"view_html":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ","download_json":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ.json","view_paper":"https://pith.science/paper/ZU4JTVZ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.10509&json=true","fetch_graph":"https://pith.science/api/pith-number/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/action/storage_attestation","attest_author":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/action/author_attestation","sign_citation":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/action/citation_signature","submit_replication":"https://pith.science/pith/ZU4JTVZ3VHXFA2QEDPRVE7F3KQ/action/replication_record"}},"created_at":"2026-07-05T08:02:21.108477+00:00","updated_at":"2026-07-05T08:02:21.108477+00:00"}