{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:EPJQOI2H2KZW2FVGQINFJUHX35","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":"e35404ddb07e7c56ccd6b90b4cd3a5ab14516a171ee3bebdca9f1c3d8340fdd7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T01:43:49Z","title_canon_sha256":"cab05a47c1fc4e4656d6695a3a234b476ee184d82c10263300e2a2c961f74bd1"},"schema_version":"1.0","source":{"id":"2509.03805","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.03805","created_at":"2026-07-05T12:04:46Z"},{"alias_kind":"arxiv_version","alias_value":"2509.03805v1","created_at":"2026-07-05T12:04:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.03805","created_at":"2026-07-05T12:04:46Z"},{"alias_kind":"pith_short_12","alias_value":"EPJQOI2H2KZW","created_at":"2026-07-05T12:04:46Z"},{"alias_kind":"pith_short_16","alias_value":"EPJQOI2H2KZW2FVG","created_at":"2026-07-05T12:04:46Z"},{"alias_kind":"pith_short_8","alias_value":"EPJQOI2H","created_at":"2026-07-05T12:04:46Z"}],"graph_snapshots":[{"event_id":"sha256:58406117b494561f06be55794eba153f93bcec77578103551ba06cd56334b044","target":"graph","created_at":"2026-07-05T12:04:46Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2509.03805/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large vision language models (VLMs) increasingly claim reasoning skills, yet current benchmarks evaluate them in single-turn or question answering settings. However, grounding is an interactive process in which people gradually develop shared understanding through ongoing communication. We introduce a four-metric suite (grounding efficiency, content alignment, lexical adaptation, and human-likeness) to systematically evaluate VLM performance in interactive grounding contexts. We deploy the suite on 150 self-play sessions of interactive referential games between three proprietary VLMs and compa","authors_text":"Anthony Sicilia, Malihe Alikhani, Mert \\.Inan, Saki Imai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T01:43:49Z","title":"Measuring How (Not Just Whether) VLMs Build Common Ground"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.03805","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:16172c3ffaf680d4de7bf4a5689d3ff85aed9e1efbddf56fbbfeaa9173c8a1ae","target":"record","created_at":"2026-07-05T12:04:46Z","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":"e35404ddb07e7c56ccd6b90b4cd3a5ab14516a171ee3bebdca9f1c3d8340fdd7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T01:43:49Z","title_canon_sha256":"cab05a47c1fc4e4656d6695a3a234b476ee184d82c10263300e2a2c961f74bd1"},"schema_version":"1.0","source":{"id":"2509.03805","kind":"arxiv","version":1}},"canonical_sha256":"23d3072347d2b36d16a6821a54d0f7df730544d137b62619586d1394bd030fe2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"23d3072347d2b36d16a6821a54d0f7df730544d137b62619586d1394bd030fe2","first_computed_at":"2026-07-05T12:04:46.759994Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:04:46.759994Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yJzp19AtL866CBOFVQ0S6hOz6pzJoCTP3kc87TvgvjvZMkBW5J1MqVnFwgcNzZsP8A6AKv8YtyYMx/HHvEM6Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:04:46.760499Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.03805","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:16172c3ffaf680d4de7bf4a5689d3ff85aed9e1efbddf56fbbfeaa9173c8a1ae","sha256:58406117b494561f06be55794eba153f93bcec77578103551ba06cd56334b044"],"state_sha256":"531916d6aa2e0a8f94a1218daa4387c97947e7661227b7d52ba3a7b99a60fb5c"}