{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KWXIMKAMBU3F647ZNALJY4RLU6","short_pith_number":"pith:KWXIMKAM","canonical_record":{"source":{"id":"2508.07485","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T21:07:08Z","cross_cats_sorted":["cs.CL","cs.CY","cs.LG"],"title_canon_sha256":"ba1149bb7b9911ffdb78ecdaec748b25e1f168bde49a8c6de4de983355ddd1b6","abstract_canon_sha256":"edef9a2af055dd923795a8ad1d5bdfa4032a19e9b96978859495be8382eb5837"},"schema_version":"1.0"},"canonical_sha256":"55ae86280c0d365f73f968169c722ba7bc484f660951c8a12de68a530579b57a","source":{"kind":"arxiv","id":"2508.07485","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.07485","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"arxiv_version","alias_value":"2508.07485v1","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07485","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"pith_short_12","alias_value":"KWXIMKAMBU3F","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"pith_short_16","alias_value":"KWXIMKAMBU3F647Z","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"pith_short_8","alias_value":"KWXIMKAM","created_at":"2026-07-05T11:51:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KWXIMKAMBU3F647ZNALJY4RLU6","target":"record","payload":{"canonical_record":{"source":{"id":"2508.07485","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T21:07:08Z","cross_cats_sorted":["cs.CL","cs.CY","cs.LG"],"title_canon_sha256":"ba1149bb7b9911ffdb78ecdaec748b25e1f168bde49a8c6de4de983355ddd1b6","abstract_canon_sha256":"edef9a2af055dd923795a8ad1d5bdfa4032a19e9b96978859495be8382eb5837"},"schema_version":"1.0"},"canonical_sha256":"55ae86280c0d365f73f968169c722ba7bc484f660951c8a12de68a530579b57a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:46.898988Z","signature_b64":"IfVTlIJf7hts1p7T93CWadPwvVqvQEB1SfSjSkquwCyzR1n3Njln/FRqEocbNpFI55UNbuoPFAjHOCO2x6lVCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55ae86280c0d365f73f968169c722ba7bc484f660951c8a12de68a530579b57a","last_reissued_at":"2026-07-05T11:51:46.898546Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:46.898546Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.07485","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":"2026-07-05T11:51:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V3KJJUxTGy6uNTDnD6WJHrAC9GQqyqU8+IrV5uRdtbNEPh2cL6a4dOYIV0mXQJjBOx+qMPxbb64emeSC2W/TBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-29T04:31:34.361243Z"},"content_sha256":"6e2b36b00d06190a8c959b38cecb89eff5c268f36b0911a12091b2a260b2cd53","schema_version":"1.0","event_id":"sha256:6e2b36b00d06190a8c959b38cecb89eff5c268f36b0911a12091b2a260b2cd53"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KWXIMKAMBU3F647ZNALJY4RLU6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Democratizing Diplomacy: A Harness for Evaluating Any Large Language Model on Full-Press Diplomacy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.LG"],"primary_cat":"cs.AI","authors_text":"Alexander Duffy, Baptiste Alloui-Cros, Elizabeth Karpinski, Ishana Shastri, Matthew Lyle Olson, Samuel J Paech, Tyler Marques","submitted_at":"2025-08-10T21:07:08Z","abstract_excerpt":"We present the first evaluation harness that enables any out-of-the-box, local, Large Language Models (LLMs) to play full-press Diplomacy without fine-tuning or specialized training. Previous work required frontier LLMs, or fine-tuning, due to the high complexity and information density of Diplomacy's game state. Combined with the high variance of matches, these factors made Diplomacy prohibitive for study. In this work, we used data-driven iteration to optimize a textual game state representation such that a 24B model can reliably complete matches without any fine tuning. We develop tooling t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07485","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/2508.07485/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":"2026-07-05T11:51:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UT0f0ZV+V1pZ6dWqVSS8galLKbyjguPNrS3XdbmsgpoMJvsEnkOSY0gpSGJv7kV4s+QLzruuCfXhlLJuKgUXBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-29T04:31:34.361645Z"},"content_sha256":"8091a68f62502a691d6eb46f6e86d23e8f560763cb620a081adcebc3a7976cbe","schema_version":"1.0","event_id":"sha256:8091a68f62502a691d6eb46f6e86d23e8f560763cb620a081adcebc3a7976cbe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KWXIMKAMBU3F647ZNALJY4RLU6/bundle.json","state_url":"https://pith.science/pith/KWXIMKAMBU3F647ZNALJY4RLU6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KWXIMKAMBU3F647ZNALJY4RLU6/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":"2026-07-29T04:31:34Z","links":{"resolver":"https://pith.science/pith/KWXIMKAMBU3F647ZNALJY4RLU6","bundle":"https://pith.science/pith/KWXIMKAMBU3F647ZNALJY4RLU6/bundle.json","state":"https://pith.science/pith/KWXIMKAMBU3F647ZNALJY4RLU6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KWXIMKAMBU3F647ZNALJY4RLU6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KWXIMKAMBU3F647ZNALJY4RLU6","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":"edef9a2af055dd923795a8ad1d5bdfa4032a19e9b96978859495be8382eb5837","cross_cats_sorted":["cs.CL","cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T21:07:08Z","title_canon_sha256":"ba1149bb7b9911ffdb78ecdaec748b25e1f168bde49a8c6de4de983355ddd1b6"},"schema_version":"1.0","source":{"id":"2508.07485","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.07485","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"arxiv_version","alias_value":"2508.07485v1","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07485","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"pith_short_12","alias_value":"KWXIMKAMBU3F","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"pith_short_16","alias_value":"KWXIMKAMBU3F647Z","created_at":"2026-07-05T11:51:46Z"},{"alias_kind":"pith_short_8","alias_value":"KWXIMKAM","created_at":"2026-07-05T11:51:46Z"}],"graph_snapshots":[{"event_id":"sha256:8091a68f62502a691d6eb46f6e86d23e8f560763cb620a081adcebc3a7976cbe","target":"graph","created_at":"2026-07-05T11:51:46Z","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/2508.07485/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present the first evaluation harness that enables any out-of-the-box, local, Large Language Models (LLMs) to play full-press Diplomacy without fine-tuning or specialized training. Previous work required frontier LLMs, or fine-tuning, due to the high complexity and information density of Diplomacy's game state. Combined with the high variance of matches, these factors made Diplomacy prohibitive for study. In this work, we used data-driven iteration to optimize a textual game state representation such that a 24B model can reliably complete matches without any fine tuning. We develop tooling t","authors_text":"Alexander Duffy, Baptiste Alloui-Cros, Elizabeth Karpinski, Ishana Shastri, Matthew Lyle Olson, Samuel J Paech, Tyler Marques","cross_cats":["cs.CL","cs.CY","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T21:07:08Z","title":"Democratizing Diplomacy: A Harness for Evaluating Any Large Language Model on Full-Press Diplomacy"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07485","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":"sha256:6e2b36b00d06190a8c959b38cecb89eff5c268f36b0911a12091b2a260b2cd53","target":"record","created_at":"2026-07-05T11:51:46Z","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":"edef9a2af055dd923795a8ad1d5bdfa4032a19e9b96978859495be8382eb5837","cross_cats_sorted":["cs.CL","cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-10T21:07:08Z","title_canon_sha256":"ba1149bb7b9911ffdb78ecdaec748b25e1f168bde49a8c6de4de983355ddd1b6"},"schema_version":"1.0","source":{"id":"2508.07485","kind":"arxiv","version":1}},"canonical_sha256":"55ae86280c0d365f73f968169c722ba7bc484f660951c8a12de68a530579b57a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"55ae86280c0d365f73f968169c722ba7bc484f660951c8a12de68a530579b57a","first_computed_at":"2026-07-05T11:51:46.898546Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:51:46.898546Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IfVTlIJf7hts1p7T93CWadPwvVqvQEB1SfSjSkquwCyzR1n3Njln/FRqEocbNpFI55UNbuoPFAjHOCO2x6lVCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:51:46.898988Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.07485","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6e2b36b00d06190a8c959b38cecb89eff5c268f36b0911a12091b2a260b2cd53","sha256:8091a68f62502a691d6eb46f6e86d23e8f560763cb620a081adcebc3a7976cbe"],"state_sha256":"d32446a4e433c6465bb4277e669f358ddf5915991aeeee122618efff67bd78df"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VBZ3KyB6o11hxFKUXIKhoYYunj9mQfUMtlTYZyczjt9O1z8bNnS9QCtGmdkeStn4rK/kuUj7kzs9hdikLtUVCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-29T04:31:34.364214Z","bundle_sha256":"e338759d350b47a82dcbaf842eb7288b4c81b4563c9584e9416ca5ec27717178"}}