{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:LZA5T32BBLG7RI4GAXW4M7KTTW","short_pith_number":"pith:LZA5T32B","canonical_record":{"source":{"id":"2406.14035","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T06:56:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"59e26777a439dda993a8d3067a92154a8b191c38a1aedbc2e78dc4e5c712d2cd","abstract_canon_sha256":"88fa7dd688bcab4bd74b48e7abfb218d5f96a4f13af1addf05db56ebb4134836"},"schema_version":"1.0"},"canonical_sha256":"5e41d9ef410acdf8a38605edc67d539da7a42a82964fdb03b187e157312f9ae5","source":{"kind":"arxiv","id":"2406.14035","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.14035","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"arxiv_version","alias_value":"2406.14035v3","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14035","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"pith_short_12","alias_value":"LZA5T32BBLG7","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"pith_short_16","alias_value":"LZA5T32BBLG7RI4G","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"pith_short_8","alias_value":"LZA5T32B","created_at":"2026-07-05T09:47:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:LZA5T32BBLG7RI4GAXW4M7KTTW","target":"record","payload":{"canonical_record":{"source":{"id":"2406.14035","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T06:56:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"59e26777a439dda993a8d3067a92154a8b191c38a1aedbc2e78dc4e5c712d2cd","abstract_canon_sha256":"88fa7dd688bcab4bd74b48e7abfb218d5f96a4f13af1addf05db56ebb4134836"},"schema_version":"1.0"},"canonical_sha256":"5e41d9ef410acdf8a38605edc67d539da7a42a82964fdb03b187e157312f9ae5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:47:31.301149Z","signature_b64":"JVCrJF0fgmi2BzN+9QacRyYFstaoH3uK/3+vmPkozAiH1NU6ZHBh1P/2MXBO47D/IdbxI3q4eVVj5iAeCCGcBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5e41d9ef410acdf8a38605edc67d539da7a42a82964fdb03b187e157312f9ae5","last_reissued_at":"2026-07-05T09:47:31.300668Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:47:31.300668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.14035","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:47:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VPKOT5i5HWGMcUmNW5xCu6QhoBRD0qVE+kB6H+juiQA4IXNrk/4hak61a8oy+qY2NNg/iMDPJS2g26RfNPHIAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T13:27:35.125876Z"},"content_sha256":"9029abf3ff93e500759fad9ddb81f55dc67ae14c7f2e5e8d0a85e40ce769027b","schema_version":"1.0","event_id":"sha256:9029abf3ff93e500759fad9ddb81f55dc67ae14c7f2e5e8d0a85e40ce769027b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:LZA5T32BBLG7RI4GAXW4M7KTTW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Using Game Play to Investigate Multimodal and Conversational Grounding in Large Multimodal Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anne Beyer, Antonia Schmidt, David Schlangen, Kushal Koshti, Sherzod Hakimov, Yan Weiser, Yerkezhan Abdullayeva","submitted_at":"2024-06-20T06:56:19Z","abstract_excerpt":"While the situation has improved for text-only models, it again seems to be the case currently that multimodal (text and image) models develop faster than ways to evaluate them. In this paper, we bring a recently developed evaluation paradigm from text models to multimodal models, namely evaluation through the goal-oriented game (self) play, complementing reference-based and preference-based evaluation. Specifically, we define games that challenge a model's capability to represent a situation from visual information and align such representations through dialogue. We find that the largest clos"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14035","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/2406.14035/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":"2026-07-05T09:47:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4ZUZz+QqYSyn/9NOZKbb+t8v0hErghtZtpkNUmzbOYXJu2l2xf8acXIFWC7tuwaSlgd/dphsy2ag10+tA2T+Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T13:27:35.126927Z"},"content_sha256":"e90a041be7038cf8f6e8e2ec623995a6a5567787fb3e265b344e250480bcdef3","schema_version":"1.0","event_id":"sha256:e90a041be7038cf8f6e8e2ec623995a6a5567787fb3e265b344e250480bcdef3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LZA5T32BBLG7RI4GAXW4M7KTTW/bundle.json","state_url":"https://pith.science/pith/LZA5T32BBLG7RI4GAXW4M7KTTW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LZA5T32BBLG7RI4GAXW4M7KTTW/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":"2026-08-10T13:27:35Z","links":{"resolver":"https://pith.science/pith/LZA5T32BBLG7RI4GAXW4M7KTTW","bundle":"https://pith.science/pith/LZA5T32BBLG7RI4GAXW4M7KTTW/bundle.json","state":"https://pith.science/pith/LZA5T32BBLG7RI4GAXW4M7KTTW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LZA5T32BBLG7RI4GAXW4M7KTTW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LZA5T32BBLG7RI4GAXW4M7KTTW","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":"88fa7dd688bcab4bd74b48e7abfb218d5f96a4f13af1addf05db56ebb4134836","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T06:56:19Z","title_canon_sha256":"59e26777a439dda993a8d3067a92154a8b191c38a1aedbc2e78dc4e5c712d2cd"},"schema_version":"1.0","source":{"id":"2406.14035","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.14035","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"arxiv_version","alias_value":"2406.14035v3","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14035","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"pith_short_12","alias_value":"LZA5T32BBLG7","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"pith_short_16","alias_value":"LZA5T32BBLG7RI4G","created_at":"2026-07-05T09:47:31Z"},{"alias_kind":"pith_short_8","alias_value":"LZA5T32B","created_at":"2026-07-05T09:47:31Z"}],"graph_snapshots":[{"event_id":"sha256:e90a041be7038cf8f6e8e2ec623995a6a5567787fb3e265b344e250480bcdef3","target":"graph","created_at":"2026-07-05T09:47:31Z","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/2406.14035/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While the situation has improved for text-only models, it again seems to be the case currently that multimodal (text and image) models develop faster than ways to evaluate them. In this paper, we bring a recently developed evaluation paradigm from text models to multimodal models, namely evaluation through the goal-oriented game (self) play, complementing reference-based and preference-based evaluation. Specifically, we define games that challenge a model's capability to represent a situation from visual information and align such representations through dialogue. We find that the largest clos","authors_text":"Anne Beyer, Antonia Schmidt, David Schlangen, Kushal Koshti, Sherzod Hakimov, Yan Weiser, Yerkezhan Abdullayeva","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T06:56:19Z","title":"Using Game Play to Investigate Multimodal and Conversational Grounding in Large Multimodal Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14035","kind":"arxiv","version":3},"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":"sha256:9029abf3ff93e500759fad9ddb81f55dc67ae14c7f2e5e8d0a85e40ce769027b","target":"record","created_at":"2026-07-05T09:47:31Z","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":"88fa7dd688bcab4bd74b48e7abfb218d5f96a4f13af1addf05db56ebb4134836","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T06:56:19Z","title_canon_sha256":"59e26777a439dda993a8d3067a92154a8b191c38a1aedbc2e78dc4e5c712d2cd"},"schema_version":"1.0","source":{"id":"2406.14035","kind":"arxiv","version":3}},"canonical_sha256":"5e41d9ef410acdf8a38605edc67d539da7a42a82964fdb03b187e157312f9ae5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5e41d9ef410acdf8a38605edc67d539da7a42a82964fdb03b187e157312f9ae5","first_computed_at":"2026-07-05T09:47:31.300668Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:47:31.300668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JVCrJF0fgmi2BzN+9QacRyYFstaoH3uK/3+vmPkozAiH1NU6ZHBh1P/2MXBO47D/IdbxI3q4eVVj5iAeCCGcBA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:47:31.301149Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.14035","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9029abf3ff93e500759fad9ddb81f55dc67ae14c7f2e5e8d0a85e40ce769027b","sha256:e90a041be7038cf8f6e8e2ec623995a6a5567787fb3e265b344e250480bcdef3"],"state_sha256":"ed9e38c4ba47c30a18e45e414c73884a48bf39b8a64d42d40df281ed4567e3ff"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JynKOk6fgkgJnsgFlWsK2IXOJFG77bKIQULL2/PmQGJ5Kv+OnGbhYJNfvY51oF8/8ySp1NVvEo9F+MbB6XCzAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T13:27:35.132189Z","bundle_sha256":"f07345f41f0f5db63d4a67a0d52304dec2ec4b63a68471390dde88be305cd944"}}