{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:EPJQOI2H2KZW2FVGQINFJUHX35","short_pith_number":"pith:EPJQOI2H","canonical_record":{"source":{"id":"2509.03805","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T01:43:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cab05a47c1fc4e4656d6695a3a234b476ee184d82c10263300e2a2c961f74bd1","abstract_canon_sha256":"e35404ddb07e7c56ccd6b90b4cd3a5ab14516a171ee3bebdca9f1c3d8340fdd7"},"schema_version":"1.0"},"canonical_sha256":"23d3072347d2b36d16a6821a54d0f7df730544d137b62619586d1394bd030fe2","source":{"kind":"arxiv","id":"2509.03805","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"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:EPJQOI2H2KZW2FVGQINFJUHX35","target":"record","payload":{"canonical_record":{"source":{"id":"2509.03805","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-04T01:43:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cab05a47c1fc4e4656d6695a3a234b476ee184d82c10263300e2a2c961f74bd1","abstract_canon_sha256":"e35404ddb07e7c56ccd6b90b4cd3a5ab14516a171ee3bebdca9f1c3d8340fdd7"},"schema_version":"1.0"},"canonical_sha256":"23d3072347d2b36d16a6821a54d0f7df730544d137b62619586d1394bd030fe2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:46.760499Z","signature_b64":"yJzp19AtL866CBOFVQ0S6hOz6pzJoCTP3kc87TvgvjvZMkBW5J1MqVnFwgcNzZsP8A6AKv8YtyYMx/HHvEM6Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23d3072347d2b36d16a6821a54d0f7df730544d137b62619586d1394bd030fe2","last_reissued_at":"2026-07-05T12:04:46.759994Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:46.759994Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.03805","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:04:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DBOvqOYeAb9dLjqV7eYeShi3iR04ZwwnOxbEExfemQAgOvzzQqzQ+mCyR/Di86kQDywMf1QuAmkJuHXx5Ln6Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:24:24.401076Z"},"content_sha256":"16172c3ffaf680d4de7bf4a5689d3ff85aed9e1efbddf56fbbfeaa9173c8a1ae","schema_version":"1.0","event_id":"sha256:16172c3ffaf680d4de7bf4a5689d3ff85aed9e1efbddf56fbbfeaa9173c8a1ae"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:EPJQOI2H2KZW2FVGQINFJUHX35","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measuring How (Not Just Whether) VLMs Build Common Ground","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anthony Sicilia, Malihe Alikhani, Mert \\.Inan, Saki Imai","submitted_at":"2025-09-04T01:43:49Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.03805","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/2509.03805/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:04:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/TymVAfvhVTviYkI3JEYtEPdI9gFA30ZHs81gXY0FO4bkXXW6yx3P5466Sh46UVHgdFEyd2yHAx2g2EOhwfxBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:24:24.401569Z"},"content_sha256":"58406117b494561f06be55794eba153f93bcec77578103551ba06cd56334b044","schema_version":"1.0","event_id":"sha256:58406117b494561f06be55794eba153f93bcec77578103551ba06cd56334b044"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EPJQOI2H2KZW2FVGQINFJUHX35/bundle.json","state_url":"https://pith.science/pith/EPJQOI2H2KZW2FVGQINFJUHX35/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EPJQOI2H2KZW2FVGQINFJUHX35/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T16:24:24Z","links":{"resolver":"https://pith.science/pith/EPJQOI2H2KZW2FVGQINFJUHX35","bundle":"https://pith.science/pith/EPJQOI2H2KZW2FVGQINFJUHX35/bundle.json","state":"https://pith.science/pith/EPJQOI2H2KZW2FVGQINFJUHX35/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EPJQOI2H2KZW2FVGQINFJUHX35/bundle.json"},"state":{"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"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ylhxMjBXCto8o+Z09gMp8a2Gkn86y0ltHlSxvkxv4oQ5AWvfWeB/LyJFkKc+Re7bVObQMFppdXSGZLYcqlq1BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T16:24:24.404884Z","bundle_sha256":"f7b0b8bc208b3f13516d5ccb48eeff384b19ac836b11abe126ddcc81b32c86cd"}}