{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:APFOEU6I5FVLE2JLCVD2DCWM3K","short_pith_number":"pith:APFOEU6I","canonical_record":{"source":{"id":"2508.11944","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:10:26Z","cross_cats_sorted":["cs.CL","cs.HC"],"title_canon_sha256":"c7e035d8b069a84827974e23e4eaddb123ce60c9273bfc8d19f36464f70b7e3f","abstract_canon_sha256":"92c1ffc4037804314dd4944aa59eda6decca8aa9edb15ef7f9dbb678d2092243"},"schema_version":"1.0"},"canonical_sha256":"03cae253c8e96ab2692b1547a18accda8ea8f2c3cef7bd9887031d1adb530b12","source":{"kind":"arxiv","id":"2508.11944","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.11944","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"arxiv_version","alias_value":"2508.11944v1","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11944","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_12","alias_value":"APFOEU6I5FVL","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_16","alias_value":"APFOEU6I5FVLE2JL","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_8","alias_value":"APFOEU6I","created_at":"2026-07-05T11:54:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:APFOEU6I5FVLE2JLCVD2DCWM3K","target":"record","payload":{"canonical_record":{"source":{"id":"2508.11944","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:10:26Z","cross_cats_sorted":["cs.CL","cs.HC"],"title_canon_sha256":"c7e035d8b069a84827974e23e4eaddb123ce60c9273bfc8d19f36464f70b7e3f","abstract_canon_sha256":"92c1ffc4037804314dd4944aa59eda6decca8aa9edb15ef7f9dbb678d2092243"},"schema_version":"1.0"},"canonical_sha256":"03cae253c8e96ab2692b1547a18accda8ea8f2c3cef7bd9887031d1adb530b12","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:57.604848Z","signature_b64":"z87Q04elF3hghgI2nZWe/q/Z9tI9uSXOEeNWJ0BSiNTwQf+0dc2ZgAj6h1Y1Ci6/43W+b+VbCkFUyvmXks50DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03cae253c8e96ab2692b1547a18accda8ea8f2c3cef7bd9887031d1adb530b12","last_reissued_at":"2026-07-05T11:54:57.604339Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:57.604339Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.11944","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":"2026-07-05T11:54:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ODBPMyOTGAbQ6M9S4c/sbNhsgTv06Tf7ejut/DNROwgipzC12zOmdM54nyqFliJi+/rhvhp5Cg1xMiPIwGbqAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:09:19.458941Z"},"content_sha256":"4437e32ea8ce3d62b1651d2f4f2c12c5f7fec319588087a581a4254d6c5aa859","schema_version":"1.0","event_id":"sha256:4437e32ea8ce3d62b1651d2f4f2c12c5f7fec319588087a581a4254d6c5aa859"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:APFOEU6I5FVLE2JLCVD2DCWM3K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CHBench: A Cognitive Hierarchy Benchmark for Evaluating Strategic Reasoning Capability of LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Hongtao Liu, Weiran Shen, Zhicheng Du, Zihe Wang","submitted_at":"2025-08-16T07:10:26Z","abstract_excerpt":"Game-playing ability serves as an indicator for evaluating the strategic reasoning capability of large language models (LLMs). While most existing studies rely on utility performance metrics, which are not robust enough due to variations in opponent behavior and game structure. To address this limitation, we propose \\textbf{Cognitive Hierarchy Benchmark (CHBench)}, a novel evaluation framework inspired by the cognitive hierarchy models from behavioral economics. We hypothesize that agents have bounded rationality -- different agents behave at varying reasoning depths/levels. We evaluate LLMs' "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11944","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/2508.11944/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":"2026-07-05T11:54:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S7ke83nSmTHwjdn4cUzvIS+SFKR0h+cohGfXQgcPt6IatmqsWw4J8SHzCm5yZd6YUo22VV6XqXnfZwKKf5ydBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:09:19.459437Z"},"content_sha256":"5a0caf10a56e5c2b6b060fe8bd605d6e636e4e8311af0d9a75ff9df2654556fb","schema_version":"1.0","event_id":"sha256:5a0caf10a56e5c2b6b060fe8bd605d6e636e4e8311af0d9a75ff9df2654556fb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/APFOEU6I5FVLE2JLCVD2DCWM3K/bundle.json","state_url":"https://pith.science/pith/APFOEU6I5FVLE2JLCVD2DCWM3K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/APFOEU6I5FVLE2JLCVD2DCWM3K/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":"2026-08-06T06:09:19Z","links":{"resolver":"https://pith.science/pith/APFOEU6I5FVLE2JLCVD2DCWM3K","bundle":"https://pith.science/pith/APFOEU6I5FVLE2JLCVD2DCWM3K/bundle.json","state":"https://pith.science/pith/APFOEU6I5FVLE2JLCVD2DCWM3K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/APFOEU6I5FVLE2JLCVD2DCWM3K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:APFOEU6I5FVLE2JLCVD2DCWM3K","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":"92c1ffc4037804314dd4944aa59eda6decca8aa9edb15ef7f9dbb678d2092243","cross_cats_sorted":["cs.CL","cs.HC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:10:26Z","title_canon_sha256":"c7e035d8b069a84827974e23e4eaddb123ce60c9273bfc8d19f36464f70b7e3f"},"schema_version":"1.0","source":{"id":"2508.11944","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.11944","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"arxiv_version","alias_value":"2508.11944v1","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11944","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_12","alias_value":"APFOEU6I5FVL","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_16","alias_value":"APFOEU6I5FVLE2JL","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_8","alias_value":"APFOEU6I","created_at":"2026-07-05T11:54:57Z"}],"graph_snapshots":[{"event_id":"sha256:5a0caf10a56e5c2b6b060fe8bd605d6e636e4e8311af0d9a75ff9df2654556fb","target":"graph","created_at":"2026-07-05T11:54:57Z","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/2508.11944/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Game-playing ability serves as an indicator for evaluating the strategic reasoning capability of large language models (LLMs). While most existing studies rely on utility performance metrics, which are not robust enough due to variations in opponent behavior and game structure. To address this limitation, we propose \\textbf{Cognitive Hierarchy Benchmark (CHBench)}, a novel evaluation framework inspired by the cognitive hierarchy models from behavioral economics. We hypothesize that agents have bounded rationality -- different agents behave at varying reasoning depths/levels. We evaluate LLMs' ","authors_text":"Hongtao Liu, Weiran Shen, Zhicheng Du, Zihe Wang","cross_cats":["cs.CL","cs.HC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:10:26Z","title":"CHBench: A Cognitive Hierarchy Benchmark for Evaluating Strategic Reasoning Capability of LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11944","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":"sha256:4437e32ea8ce3d62b1651d2f4f2c12c5f7fec319588087a581a4254d6c5aa859","target":"record","created_at":"2026-07-05T11:54:57Z","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":"92c1ffc4037804314dd4944aa59eda6decca8aa9edb15ef7f9dbb678d2092243","cross_cats_sorted":["cs.CL","cs.HC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:10:26Z","title_canon_sha256":"c7e035d8b069a84827974e23e4eaddb123ce60c9273bfc8d19f36464f70b7e3f"},"schema_version":"1.0","source":{"id":"2508.11944","kind":"arxiv","version":1}},"canonical_sha256":"03cae253c8e96ab2692b1547a18accda8ea8f2c3cef7bd9887031d1adb530b12","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"03cae253c8e96ab2692b1547a18accda8ea8f2c3cef7bd9887031d1adb530b12","first_computed_at":"2026-07-05T11:54:57.604339Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:54:57.604339Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z87Q04elF3hghgI2nZWe/q/Z9tI9uSXOEeNWJ0BSiNTwQf+0dc2ZgAj6h1Y1Ci6/43W+b+VbCkFUyvmXks50DA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:54:57.604848Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.11944","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4437e32ea8ce3d62b1651d2f4f2c12c5f7fec319588087a581a4254d6c5aa859","sha256:5a0caf10a56e5c2b6b060fe8bd605d6e636e4e8311af0d9a75ff9df2654556fb"],"state_sha256":"7d42d824b1e5d5cc2e8a61dfc51c498f8a99ed37af63fd866efe49cbcca68fa3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t9VAZ4/TQfJYARGphKvSLw8iknv3VsZK7fMdlADCbe7k37AvvGEJbBOJn++WLvpobutX79JSan7oU2zMA1ygBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T06:09:19.463059Z","bundle_sha256":"4f52ed9db82389246d4c32f7ec8e1144ee8540ec76ca26ed55443df083bb9995"}}