{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IZFZ4VAG2MMMATKVRB6CXUUGVQ","short_pith_number":"pith:IZFZ4VAG","schema_version":"1.0","canonical_sha256":"464b9e5406d318c04d55887c2bd286ac3fd70442d0d94ffade7f82a7c4fe9c42","source":{"kind":"arxiv","id":"2504.16060","version":3},"attestation_state":"computed","paper":{"title":"Vision-Language Models Are Not Pragmatically Competent in Referring Expression Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dezhi Luo, Jiahe Ding, Jing Ding, Joyce Chai, Run Peng, Sihan Xu, Xuejun Zhang, Yuchen Huang, Ziqiao Ma","submitted_at":"2025-04-22T17:37:16Z","abstract_excerpt":"Referring Expression Generation (REG) is a core task for evaluating the pragmatic competence of vision-language systems, requiring not only accurate semantic grounding but also adherence to principles of cooperative communication (Grice, 1975). However, current evaluations of vision-language models (VLMs) often overlook the pragmatic dimension, reducing REG to a region-based captioning task and neglecting Gricean maxims. In this work, we revisit REG from a pragmatic perspective, introducing a new dataset (RefOI) of 1.5k images annotated with both written and spoken referring expressions. Throu"},"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":"2504.16060","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-22T17:37:16Z","cross_cats_sorted":[],"title_canon_sha256":"6ce6eae299b76638142480802efd25f2f1a5f0186e3444167753f86b7b9b9ee0","abstract_canon_sha256":"10857d8885838aa315788b97971aca45dec76199148e6bbde9bffc5821b20c21"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:57.583120Z","signature_b64":"2+cj+sUdeuqbXAHVWBKrYjftWtP9A4yI24RHAI1IQIy9Hd0GFTVzHA92kplQupyf2TBv9Kwchw3lph99/yP8Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"464b9e5406d318c04d55887c2bd286ac3fd70442d0d94ffade7f82a7c4fe9c42","last_reissued_at":"2026-07-05T11:45:57.582508Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:57.582508Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vision-Language Models Are Not Pragmatically Competent in Referring Expression Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dezhi Luo, Jiahe Ding, Jing Ding, Joyce Chai, Run Peng, Sihan Xu, Xuejun Zhang, Yuchen Huang, Ziqiao Ma","submitted_at":"2025-04-22T17:37:16Z","abstract_excerpt":"Referring Expression Generation (REG) is a core task for evaluating the pragmatic competence of vision-language systems, requiring not only accurate semantic grounding but also adherence to principles of cooperative communication (Grice, 1975). However, current evaluations of vision-language models (VLMs) often overlook the pragmatic dimension, reducing REG to a region-based captioning task and neglecting Gricean maxims. In this work, we revisit REG from a pragmatic perspective, introducing a new dataset (RefOI) of 1.5k images annotated with both written and spoken referring expressions. Throu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.16060","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/2504.16060/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":"2504.16060","created_at":"2026-07-05T11:45:57.582586+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.16060v3","created_at":"2026-07-05T11:45:57.582586+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.16060","created_at":"2026-07-05T11:45:57.582586+00:00"},{"alias_kind":"pith_short_12","alias_value":"IZFZ4VAG2MMM","created_at":"2026-07-05T11:45:57.582586+00:00"},{"alias_kind":"pith_short_16","alias_value":"IZFZ4VAG2MMMATKV","created_at":"2026-07-05T11:45:57.582586+00:00"},{"alias_kind":"pith_short_8","alias_value":"IZFZ4VAG","created_at":"2026-07-05T11:45:57.582586+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2510.23340","citing_title":"Planning Ahead with RSA: Efficient Signalling in Dynamic Environments by Projecting User Awareness across Future Timesteps","ref_index":31,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ","json":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ.json","graph_json":"https://pith.science/api/pith-number/IZFZ4VAG2MMMATKVRB6CXUUGVQ/graph.json","events_json":"https://pith.science/api/pith-number/IZFZ4VAG2MMMATKVRB6CXUUGVQ/events.json","paper":"https://pith.science/paper/IZFZ4VAG"},"agent_actions":{"view_html":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ","download_json":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ.json","view_paper":"https://pith.science/paper/IZFZ4VAG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.16060&json=true","fetch_graph":"https://pith.science/api/pith-number/IZFZ4VAG2MMMATKVRB6CXUUGVQ/graph.json","fetch_events":"https://pith.science/api/pith-number/IZFZ4VAG2MMMATKVRB6CXUUGVQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ/action/storage_attestation","attest_author":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ/action/author_attestation","sign_citation":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ/action/citation_signature","submit_replication":"https://pith.science/pith/IZFZ4VAG2MMMATKVRB6CXUUGVQ/action/replication_record"}},"created_at":"2026-07-05T11:45:57.582586+00:00","updated_at":"2026-07-05T11:45:57.582586+00:00"}