{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:I5F23DTSSUO3OIJMK6U2EAF43M","short_pith_number":"pith:I5F23DTS","canonical_record":{"source":{"id":"2410.13204","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-17T04:12:17Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"107b1d67cd283573f578409cd954b5f75b1d77ef199146f1c414bba99c336356","abstract_canon_sha256":"66057bf40652e92dc20d59ee653b64cc51fb45c5d3b50ce2b9c372e7c0a8d7db"},"schema_version":"1.0"},"canonical_sha256":"474bad8e72951db7212c57a9a200bcdb1de8f72afbcffd91cab8cc93b6dad0fe","source":{"kind":"arxiv","id":"2410.13204","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.13204","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"arxiv_version","alias_value":"2410.13204v1","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13204","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"pith_short_12","alias_value":"I5F23DTSSUO3","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"pith_short_16","alias_value":"I5F23DTSSUO3OIJM","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"pith_short_8","alias_value":"I5F23DTS","created_at":"2026-07-05T09:23:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:I5F23DTSSUO3OIJMK6U2EAF43M","target":"record","payload":{"canonical_record":{"source":{"id":"2410.13204","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-17T04:12:17Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"107b1d67cd283573f578409cd954b5f75b1d77ef199146f1c414bba99c336356","abstract_canon_sha256":"66057bf40652e92dc20d59ee653b64cc51fb45c5d3b50ce2b9c372e7c0a8d7db"},"schema_version":"1.0"},"canonical_sha256":"474bad8e72951db7212c57a9a200bcdb1de8f72afbcffd91cab8cc93b6dad0fe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:23:15.329117Z","signature_b64":"gfQutgoSWEiQVLcTzMQrwc3SFBK30hu/rIJBp/JzUedASnuSdQXkl81qpcvc7Qaa24eTS8j048iWAkimmkEoBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"474bad8e72951db7212c57a9a200bcdb1de8f72afbcffd91cab8cc93b6dad0fe","last_reissued_at":"2026-07-05T09:23:15.327629Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:23:15.327629Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.13204","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":"2026-07-05T09:23:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7bpe0rcQw5AZRQ6hucaryaVoGDI25/WzoD0WmgW54Zrea23DPpBJshjcczqBlRiUliE9Rkj1/0fK9iNpCOZJBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T11:25:47.395272Z"},"content_sha256":"f2535c7c4fee11ff2cd08fd8e2bb1dcac844f4c8a3c441bd785c1e18b48f09c2","schema_version":"1.0","event_id":"sha256:f2535c7c4fee11ff2cd08fd8e2bb1dcac844f4c8a3c441bd785c1e18b48f09c2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:I5F23DTSSUO3OIJMK6U2EAF43M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measuring Free-Form Decision-Making Inconsistency of Language Models in Military Crisis Simulations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CL","authors_text":"Aryan Shrivastava, Jessica Hullman, Max Lamparth","submitted_at":"2024-10-17T04:12:17Z","abstract_excerpt":"There is an increasing interest in using language models (LMs) for automated decision-making, with multiple countries actively testing LMs to aid in military crisis decision-making. To scrutinize relying on LM decision-making in high-stakes settings, we examine the inconsistency of responses in a crisis simulation (\"wargame\"), similar to reported tests conducted by the US military. Prior work illustrated escalatory tendencies and varying levels of aggression among LMs but were constrained to simulations with pre-defined actions. This was due to the challenges associated with quantitatively mea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13204","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/2410.13204/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":"2026-07-05T09:23:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CtxGSVzxNogydxS0cQLYWClv1USnLAWjfmIFzLyqXP6GuLdwKqkwADrBqoRdBOdYK1riENaae3mBekXVAc+SAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T11:25:47.396250Z"},"content_sha256":"5fffb71eca279c8e53625377cf6410e27f0b93ddb07c97b005015304baeb7508","schema_version":"1.0","event_id":"sha256:5fffb71eca279c8e53625377cf6410e27f0b93ddb07c97b005015304baeb7508"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/I5F23DTSSUO3OIJMK6U2EAF43M/bundle.json","state_url":"https://pith.science/pith/I5F23DTSSUO3OIJMK6U2EAF43M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/I5F23DTSSUO3OIJMK6U2EAF43M/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":"2026-08-10T11:25:47Z","links":{"resolver":"https://pith.science/pith/I5F23DTSSUO3OIJMK6U2EAF43M","bundle":"https://pith.science/pith/I5F23DTSSUO3OIJMK6U2EAF43M/bundle.json","state":"https://pith.science/pith/I5F23DTSSUO3OIJMK6U2EAF43M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/I5F23DTSSUO3OIJMK6U2EAF43M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:I5F23DTSSUO3OIJMK6U2EAF43M","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":"66057bf40652e92dc20d59ee653b64cc51fb45c5d3b50ce2b9c372e7c0a8d7db","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-17T04:12:17Z","title_canon_sha256":"107b1d67cd283573f578409cd954b5f75b1d77ef199146f1c414bba99c336356"},"schema_version":"1.0","source":{"id":"2410.13204","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.13204","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"arxiv_version","alias_value":"2410.13204v1","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13204","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"pith_short_12","alias_value":"I5F23DTSSUO3","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"pith_short_16","alias_value":"I5F23DTSSUO3OIJM","created_at":"2026-07-05T09:23:15Z"},{"alias_kind":"pith_short_8","alias_value":"I5F23DTS","created_at":"2026-07-05T09:23:15Z"}],"graph_snapshots":[{"event_id":"sha256:5fffb71eca279c8e53625377cf6410e27f0b93ddb07c97b005015304baeb7508","target":"graph","created_at":"2026-07-05T09:23:15Z","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/2410.13204/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"There is an increasing interest in using language models (LMs) for automated decision-making, with multiple countries actively testing LMs to aid in military crisis decision-making. To scrutinize relying on LM decision-making in high-stakes settings, we examine the inconsistency of responses in a crisis simulation (\"wargame\"), similar to reported tests conducted by the US military. Prior work illustrated escalatory tendencies and varying levels of aggression among LMs but were constrained to simulations with pre-defined actions. This was due to the challenges associated with quantitatively mea","authors_text":"Aryan Shrivastava, Jessica Hullman, Max Lamparth","cross_cats":["cs.AI","cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-17T04:12:17Z","title":"Measuring Free-Form Decision-Making Inconsistency of Language Models in Military Crisis Simulations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13204","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:f2535c7c4fee11ff2cd08fd8e2bb1dcac844f4c8a3c441bd785c1e18b48f09c2","target":"record","created_at":"2026-07-05T09:23:15Z","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":"66057bf40652e92dc20d59ee653b64cc51fb45c5d3b50ce2b9c372e7c0a8d7db","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-17T04:12:17Z","title_canon_sha256":"107b1d67cd283573f578409cd954b5f75b1d77ef199146f1c414bba99c336356"},"schema_version":"1.0","source":{"id":"2410.13204","kind":"arxiv","version":1}},"canonical_sha256":"474bad8e72951db7212c57a9a200bcdb1de8f72afbcffd91cab8cc93b6dad0fe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"474bad8e72951db7212c57a9a200bcdb1de8f72afbcffd91cab8cc93b6dad0fe","first_computed_at":"2026-07-05T09:23:15.327629Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:23:15.327629Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gfQutgoSWEiQVLcTzMQrwc3SFBK30hu/rIJBp/JzUedASnuSdQXkl81qpcvc7Qaa24eTS8j048iWAkimmkEoBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:23:15.329117Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.13204","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f2535c7c4fee11ff2cd08fd8e2bb1dcac844f4c8a3c441bd785c1e18b48f09c2","sha256:5fffb71eca279c8e53625377cf6410e27f0b93ddb07c97b005015304baeb7508"],"state_sha256":"de208c21deac430c6e93214b1d879f0028bc9f6180a29bd10ba15d4e23c24b83"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sCKTDUCf+uG/RDouJkHbOrmudp/3TDE79ayRDqwGS28vcLVEwQ5WWoOqpTol+C2xhaa+z8ipRl+JrX11xpf0Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T11:25:47.403105Z","bundle_sha256":"2db87d16ffad536f526b52bdaa1c0805cb606e23f1a7833b07b2ef9b04d3188d"}}