{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Q7ZK5ZEXLUDWYMDULOBJW4Z6TM","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":"f7c97090b2b7fd9f92c593c9f326f80da443f3faa96739f33408d4fae4eaced0","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-23T15:37:40Z","title_canon_sha256":"cab4fbbb3899e32c8f49ef927d74b33da7e6eaefb25007fb70969dfca357c8d2"},"schema_version":"1.0","source":{"id":"2505.18034","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.18034","created_at":"2026-07-05T11:10:09Z"},{"alias_kind":"arxiv_version","alias_value":"2505.18034v2","created_at":"2026-07-05T11:10:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18034","created_at":"2026-07-05T11:10:09Z"},{"alias_kind":"pith_short_12","alias_value":"Q7ZK5ZEXLUDW","created_at":"2026-07-05T11:10:09Z"},{"alias_kind":"pith_short_16","alias_value":"Q7ZK5ZEXLUDWYMDU","created_at":"2026-07-05T11:10:09Z"},{"alias_kind":"pith_short_8","alias_value":"Q7ZK5ZEX","created_at":"2026-07-05T11:10:09Z"}],"graph_snapshots":[{"event_id":"sha256:b8858b560c0062dd2ac1ab5921ea23f485dce6a2df90ea114f6de0a7f01c8e6e","target":"graph","created_at":"2026-07-05T11:10:09Z","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/2505.18034/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite remarkable advances in the field, LLMs remain unreliable in distinguishing causation from correlation. Recent results from the Corr2Cause dataset benchmark reveal that state-of-the-art LLMs -- such as GPT-4 (F1 score: 29.08) -- only marginally outperform random baselines (Random Uniform, F1 score: 20.38), indicating limited capacity of generalization. To tackle this limitation, we propose a novel structured approach: rather than directly answering causal queries, we provide the model with the capability to structure its thinking by guiding the model to build a structured knowledge grap","authors_text":"Alonso Silva, Jo\\~ao Paulo Nogueira, Wentao Sun","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-23T15:37:40Z","title":"Structured Thinking Matters: Improving LLMs Generalization in Causal Inference Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18034","kind":"arxiv","version":2},"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:b03b8c2c761ad9d1559aba78068826206928d5968dc7c3a8f9d73cbc185baca3","target":"record","created_at":"2026-07-05T11:10:09Z","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":"f7c97090b2b7fd9f92c593c9f326f80da443f3faa96739f33408d4fae4eaced0","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-23T15:37:40Z","title_canon_sha256":"cab4fbbb3899e32c8f49ef927d74b33da7e6eaefb25007fb70969dfca357c8d2"},"schema_version":"1.0","source":{"id":"2505.18034","kind":"arxiv","version":2}},"canonical_sha256":"87f2aee4975d076c30745b829b733e9b29a5b46d54d6989d7860e03f783bdff1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"87f2aee4975d076c30745b829b733e9b29a5b46d54d6989d7860e03f783bdff1","first_computed_at":"2026-07-05T11:10:09.636821Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:09.636821Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0moUIV31svh8tPsyPaunVCe+LDYSSJfh5Xb4dA8Wxb9gP85drV73zccWkMucFIa5SaYXdvoUTYS/nraV6FnWDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:09.637306Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.18034","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b03b8c2c761ad9d1559aba78068826206928d5968dc7c3a8f9d73cbc185baca3","sha256:b8858b560c0062dd2ac1ab5921ea23f485dce6a2df90ea114f6de0a7f01c8e6e"],"state_sha256":"dfe037b1d4b5232516829f36080abc57a3cb6bdfca4602ca75aa48db91775e40"}