{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XXAHKAVXCOG65TXOIJTSGMDU45","short_pith_number":"pith:XXAHKAVX","canonical_record":{"source":{"id":"2506.12012","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-13T17:59:10Z","cross_cats_sorted":[],"title_canon_sha256":"a539448c2026b51d339b91b038bbc18b430929aa49c55ebd859334821a2be159","abstract_canon_sha256":"e30cf8c8efb8298bdae50d470debe3a12edd96b66cfd34e8be11b74de4acddc8"},"schema_version":"1.0"},"canonical_sha256":"bdc07502b7138deeceee4267233074e77d7fbe5f7d04d9c272fcaf6ffb6eb133","source":{"kind":"arxiv","id":"2506.12012","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12012","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12012v1","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12012","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"pith_short_12","alias_value":"XXAHKAVXCOG6","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"pith_short_16","alias_value":"XXAHKAVXCOG65TXO","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"pith_short_8","alias_value":"XXAHKAVX","created_at":"2026-07-05T11:21:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XXAHKAVXCOG65TXOIJTSGMDU45","target":"record","payload":{"canonical_record":{"source":{"id":"2506.12012","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-13T17:59:10Z","cross_cats_sorted":[],"title_canon_sha256":"a539448c2026b51d339b91b038bbc18b430929aa49c55ebd859334821a2be159","abstract_canon_sha256":"e30cf8c8efb8298bdae50d470debe3a12edd96b66cfd34e8be11b74de4acddc8"},"schema_version":"1.0"},"canonical_sha256":"bdc07502b7138deeceee4267233074e77d7fbe5f7d04d9c272fcaf6ffb6eb133","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:14.177263Z","signature_b64":"9IxSOfkJa99BEV5kCtpc05pzG5KKFDQad2kGDDY9dRiXM8NRbpuBxhACf2bsGkWRtjUAOQfiqbvQGxCa7gWHCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bdc07502b7138deeceee4267233074e77d7fbe5f7d04d9c272fcaf6ffb6eb133","last_reissued_at":"2026-07-05T11:21:14.176845Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:14.176845Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.12012","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":"2026-07-05T11:21:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gj1jresvBDuXMy3hfN8MiBSXa2Ws3Hv5Gdv+gqh32potx3ppq0DsA3Lat2Q1rXn+MSBimOUUtQl3D99kTCyQCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T17:08:13.748939Z"},"content_sha256":"6fabaaacfd0e7623674607863e87952bd0ca9463c8bb19d13ab04cce8003002f","schema_version":"1.0","event_id":"sha256:6fabaaacfd0e7623674607863e87952bd0ca9463c8bb19d13ab04cce8003002f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XXAHKAVXCOG65TXOIJTSGMDU45","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tracing LLM Reasoning Processes with Strategic Games: A Framework for Planning, Revision, and Resource-Constrained Decision Making","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haohan Wang, Jindong Wang, Ke Xu, Lijun Yu, Xiaopeng Yuan, Xingjian Zhang, Yifan Xu, Yushun Dong","submitted_at":"2025-06-13T17:59:10Z","abstract_excerpt":"Large language models (LLMs) are increasingly used for tasks that require complex reasoning. Most benchmarks focus on final outcomes but overlook the intermediate reasoning steps - such as planning, revision, and decision making under resource constraints. We argue that measuring these internal processes is essential for understanding model behavior and improving reliability. We propose using strategic games as a natural evaluation environment: closed, rule-based systems with clear states, limited resources, and automatic feedback. We introduce a framework that evaluates LLMs along three core "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12012","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/2506.12012/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":"2026-07-05T11:21:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ypYnuMzjMPcBJYy3Vd+4BbHOBPy1IbWAE/+pxMaIG02r62K/nBB9Y3D/9OaVcoZpQBcr/jXJ7ptmLKSObyPLCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T17:08:13.749549Z"},"content_sha256":"ad66287ad51ab0688402277312aeb888642369b990b5a6b25386f322adc1f358","schema_version":"1.0","event_id":"sha256:ad66287ad51ab0688402277312aeb888642369b990b5a6b25386f322adc1f358"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XXAHKAVXCOG65TXOIJTSGMDU45/bundle.json","state_url":"https://pith.science/pith/XXAHKAVXCOG65TXOIJTSGMDU45/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XXAHKAVXCOG65TXOIJTSGMDU45/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":"2026-08-07T17:08:13Z","links":{"resolver":"https://pith.science/pith/XXAHKAVXCOG65TXOIJTSGMDU45","bundle":"https://pith.science/pith/XXAHKAVXCOG65TXOIJTSGMDU45/bundle.json","state":"https://pith.science/pith/XXAHKAVXCOG65TXOIJTSGMDU45/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XXAHKAVXCOG65TXOIJTSGMDU45/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XXAHKAVXCOG65TXOIJTSGMDU45","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":"e30cf8c8efb8298bdae50d470debe3a12edd96b66cfd34e8be11b74de4acddc8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-13T17:59:10Z","title_canon_sha256":"a539448c2026b51d339b91b038bbc18b430929aa49c55ebd859334821a2be159"},"schema_version":"1.0","source":{"id":"2506.12012","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12012","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12012v1","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12012","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"pith_short_12","alias_value":"XXAHKAVXCOG6","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"pith_short_16","alias_value":"XXAHKAVXCOG65TXO","created_at":"2026-07-05T11:21:14Z"},{"alias_kind":"pith_short_8","alias_value":"XXAHKAVX","created_at":"2026-07-05T11:21:14Z"}],"graph_snapshots":[{"event_id":"sha256:ad66287ad51ab0688402277312aeb888642369b990b5a6b25386f322adc1f358","target":"graph","created_at":"2026-07-05T11:21:14Z","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/2506.12012/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are increasingly used for tasks that require complex reasoning. Most benchmarks focus on final outcomes but overlook the intermediate reasoning steps - such as planning, revision, and decision making under resource constraints. We argue that measuring these internal processes is essential for understanding model behavior and improving reliability. We propose using strategic games as a natural evaluation environment: closed, rule-based systems with clear states, limited resources, and automatic feedback. We introduce a framework that evaluates LLMs along three core ","authors_text":"Haohan Wang, Jindong Wang, Ke Xu, Lijun Yu, Xiaopeng Yuan, Xingjian Zhang, Yifan Xu, Yushun Dong","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-13T17:59:10Z","title":"Tracing LLM Reasoning Processes with Strategic Games: A Framework for Planning, Revision, and Resource-Constrained Decision Making"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12012","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":"sha256:6fabaaacfd0e7623674607863e87952bd0ca9463c8bb19d13ab04cce8003002f","target":"record","created_at":"2026-07-05T11:21:14Z","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":"e30cf8c8efb8298bdae50d470debe3a12edd96b66cfd34e8be11b74de4acddc8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-13T17:59:10Z","title_canon_sha256":"a539448c2026b51d339b91b038bbc18b430929aa49c55ebd859334821a2be159"},"schema_version":"1.0","source":{"id":"2506.12012","kind":"arxiv","version":1}},"canonical_sha256":"bdc07502b7138deeceee4267233074e77d7fbe5f7d04d9c272fcaf6ffb6eb133","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bdc07502b7138deeceee4267233074e77d7fbe5f7d04d9c272fcaf6ffb6eb133","first_computed_at":"2026-07-05T11:21:14.176845Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:21:14.176845Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9IxSOfkJa99BEV5kCtpc05pzG5KKFDQad2kGDDY9dRiXM8NRbpuBxhACf2bsGkWRtjUAOQfiqbvQGxCa7gWHCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:21:14.177263Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.12012","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6fabaaacfd0e7623674607863e87952bd0ca9463c8bb19d13ab04cce8003002f","sha256:ad66287ad51ab0688402277312aeb888642369b990b5a6b25386f322adc1f358"],"state_sha256":"be83ebeb9cab94aaf03905d18e72f3d9e10c18f817650bcf7e6cf8e3fbef467f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/1xRF99sqLghQ6nnXQlHxqNqEbsfUj7IhAtAr9gV+8AXk8SXPvctaYObBpKTCA7YsRO9MgthU9XoKnBt0s8YAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T17:08:13.753953Z","bundle_sha256":"8ca77d0d34356be500b2079b02fd95864cf0ae3586a637098cd47b4c866ee508"}}