{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:XCRNB3ECFPLNRK35TKN2OKGPAE","short_pith_number":"pith:XCRNB3EC","canonical_record":{"source":{"id":"2409.06820","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-10T19:00:44Z","cross_cats_sorted":[],"title_canon_sha256":"88a853b5ce94698e61b954aba73a4749f8f21c4579b01b365bc94bebec4ab1cb","abstract_canon_sha256":"3478769e00ad161652c817c8f61e5d97266ab5234f13dad17b8c9021e21ed835"},"schema_version":"1.0"},"canonical_sha256":"b8a2d0ec822bd6d8ab7d9a9ba728cf013c0f2de7a6ca398d2d47e35b2bcd1ea0","source":{"kind":"arxiv","id":"2409.06820","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.06820","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"arxiv_version","alias_value":"2409.06820v4","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.06820","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"pith_short_12","alias_value":"XCRNB3ECFPLN","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"pith_short_16","alias_value":"XCRNB3ECFPLNRK35","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"pith_short_8","alias_value":"XCRNB3EC","created_at":"2026-07-05T10:46:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:XCRNB3ECFPLNRK35TKN2OKGPAE","target":"record","payload":{"canonical_record":{"source":{"id":"2409.06820","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-10T19:00:44Z","cross_cats_sorted":[],"title_canon_sha256":"88a853b5ce94698e61b954aba73a4749f8f21c4579b01b365bc94bebec4ab1cb","abstract_canon_sha256":"3478769e00ad161652c817c8f61e5d97266ab5234f13dad17b8c9021e21ed835"},"schema_version":"1.0"},"canonical_sha256":"b8a2d0ec822bd6d8ab7d9a9ba728cf013c0f2de7a6ca398d2d47e35b2bcd1ea0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:38.289849Z","signature_b64":"/fQ1iwhhIJc6G+8brhZ9mPtVS8FW8eR29/7z2NWwYGnAHVTmLKP0StW9fI1d7K3//akzB3Nx83/Wo3Im9oLNAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b8a2d0ec822bd6d8ab7d9a9ba728cf013c0f2de7a6ca398d2d47e35b2bcd1ea0","last_reissued_at":"2026-07-05T10:46:38.289335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:38.289335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.06820","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:46:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kGZ7Azeh6JhM9HGsVd1P8ErIdT4bZZ1H7Aljvbb/pL7AY/mnNkIccSxaP6rXPdBEu2AoO3o8LMzDAQedWRFSCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T07:40:41.102595Z"},"content_sha256":"e4a1e4abf3b2d000c82665422e66ed06698c96b7346aa482d8ab7adfe3fd2e1b","schema_version":"1.0","event_id":"sha256:e4a1e4abf3b2d000c82665422e66ed06698c96b7346aa482d8ab7adfe3fd2e1b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:XCRNB3ECFPLNRK35TKN2OKGPAE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PingPong: A Benchmark for Role-Playing Language Models with User Emulation and Multi-Model Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ilya Gusev","submitted_at":"2024-09-10T19:00:44Z","abstract_excerpt":"We introduce a benchmark for evaluating the role-playing capabilities of language models. Our approach leverages different language models to simulate users in dynamic, multi-turn conversations and assess the resulting dialogues. Our methodology involves three main components: a player model that adopts a specific character role, an interrogator model that simulates user behavior in a specific situation, and a judge model ensemble that evaluates conversation quality with 3 metrics: character consistency, entertainment value, and language fluency. We evaluated more than 40 models in both Englis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.06820","kind":"arxiv","version":4},"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/2409.06820/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":"2026-07-05T10:46:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kmSh+9x8HQQ5hTYaSh/9X6G+y9Pm5U9KRL1Zv3PiIJ+kzbp/F71KAkRkTaT/bR3MohU6sCDZkp0apeiqVT8oAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T07:40:41.103638Z"},"content_sha256":"69d8ac87a46dbbfd107ddc5da07b4d0dad1c7fe8930e97f43d9e0ac9662e9d7e","schema_version":"1.0","event_id":"sha256:69d8ac87a46dbbfd107ddc5da07b4d0dad1c7fe8930e97f43d9e0ac9662e9d7e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XCRNB3ECFPLNRK35TKN2OKGPAE/bundle.json","state_url":"https://pith.science/pith/XCRNB3ECFPLNRK35TKN2OKGPAE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XCRNB3ECFPLNRK35TKN2OKGPAE/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":"2026-08-05T07:40:41Z","links":{"resolver":"https://pith.science/pith/XCRNB3ECFPLNRK35TKN2OKGPAE","bundle":"https://pith.science/pith/XCRNB3ECFPLNRK35TKN2OKGPAE/bundle.json","state":"https://pith.science/pith/XCRNB3ECFPLNRK35TKN2OKGPAE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XCRNB3ECFPLNRK35TKN2OKGPAE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XCRNB3ECFPLNRK35TKN2OKGPAE","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":"3478769e00ad161652c817c8f61e5d97266ab5234f13dad17b8c9021e21ed835","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-10T19:00:44Z","title_canon_sha256":"88a853b5ce94698e61b954aba73a4749f8f21c4579b01b365bc94bebec4ab1cb"},"schema_version":"1.0","source":{"id":"2409.06820","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.06820","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"arxiv_version","alias_value":"2409.06820v4","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.06820","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"pith_short_12","alias_value":"XCRNB3ECFPLN","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"pith_short_16","alias_value":"XCRNB3ECFPLNRK35","created_at":"2026-07-05T10:46:38Z"},{"alias_kind":"pith_short_8","alias_value":"XCRNB3EC","created_at":"2026-07-05T10:46:38Z"}],"graph_snapshots":[{"event_id":"sha256:69d8ac87a46dbbfd107ddc5da07b4d0dad1c7fe8930e97f43d9e0ac9662e9d7e","target":"graph","created_at":"2026-07-05T10:46:38Z","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/2409.06820/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce a benchmark for evaluating the role-playing capabilities of language models. Our approach leverages different language models to simulate users in dynamic, multi-turn conversations and assess the resulting dialogues. Our methodology involves three main components: a player model that adopts a specific character role, an interrogator model that simulates user behavior in a specific situation, and a judge model ensemble that evaluates conversation quality with 3 metrics: character consistency, entertainment value, and language fluency. We evaluated more than 40 models in both Englis","authors_text":"Ilya Gusev","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-10T19:00:44Z","title":"PingPong: A Benchmark for Role-Playing Language Models with User Emulation and Multi-Model Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.06820","kind":"arxiv","version":4},"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:e4a1e4abf3b2d000c82665422e66ed06698c96b7346aa482d8ab7adfe3fd2e1b","target":"record","created_at":"2026-07-05T10:46:38Z","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":"3478769e00ad161652c817c8f61e5d97266ab5234f13dad17b8c9021e21ed835","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-10T19:00:44Z","title_canon_sha256":"88a853b5ce94698e61b954aba73a4749f8f21c4579b01b365bc94bebec4ab1cb"},"schema_version":"1.0","source":{"id":"2409.06820","kind":"arxiv","version":4}},"canonical_sha256":"b8a2d0ec822bd6d8ab7d9a9ba728cf013c0f2de7a6ca398d2d47e35b2bcd1ea0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b8a2d0ec822bd6d8ab7d9a9ba728cf013c0f2de7a6ca398d2d47e35b2bcd1ea0","first_computed_at":"2026-07-05T10:46:38.289335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:46:38.289335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/fQ1iwhhIJc6G+8brhZ9mPtVS8FW8eR29/7z2NWwYGnAHVTmLKP0StW9fI1d7K3//akzB3Nx83/Wo3Im9oLNAg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:46:38.289849Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.06820","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e4a1e4abf3b2d000c82665422e66ed06698c96b7346aa482d8ab7adfe3fd2e1b","sha256:69d8ac87a46dbbfd107ddc5da07b4d0dad1c7fe8930e97f43d9e0ac9662e9d7e"],"state_sha256":"463811d2de3ff2161d89c20061eec91a94b3873e3b18a9c8538ebb1283d22fc3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aBxCc3Tokuhl70MWtxXZ4xJSioucZAT9u01Y2J+i1dP4Wt8D9sCXa58/AIStcoy/XpeGLFnOmRVhiNeQHozBDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T07:40:41.111465Z","bundle_sha256":"36936cd31a8edaa80f5a548725737ef41c6817f7d4a11ba2111226bdd15f9e2c"}}