{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MBTXUPVPXBVSXCTL5NNY4UTFUY","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":"ab328de9c080efcab435d2969ebc8641f41b63b64ed6f83ab87a5ab5f3ad1fa2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-23T18:06:15Z","title_canon_sha256":"e31d8eb7b78b57a65286b3403d274ecb9c36da6b0239a3ef1eb0ceed2c4f44ed"},"schema_version":"1.0","source":{"id":"2602.20094","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.20094","created_at":"2026-06-29T00:14:02Z"},{"alias_kind":"arxiv_version","alias_value":"2602.20094v2","created_at":"2026-06-29T00:14:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.20094","created_at":"2026-06-29T00:14:02Z"},{"alias_kind":"pith_short_12","alias_value":"MBTXUPVPXBVS","created_at":"2026-06-29T00:14:02Z"},{"alias_kind":"pith_short_16","alias_value":"MBTXUPVPXBVSXCTL","created_at":"2026-06-29T00:14:02Z"},{"alias_kind":"pith_short_8","alias_value":"MBTXUPVP","created_at":"2026-06-29T00:14:02Z"}],"graph_snapshots":[{"event_id":"sha256:30e1ef32266226ab3a2fd4aadc635b62cc78014fa907f74815e6ff5cf28feb7c","target":"graph","created_at":"2026-06-29T00:14:02Z","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/2602.20094/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) witness increasing deployment in complex, high-stakes decision-making scenarios, it becomes imperative to ground their reasoning in causality rather than spurious correlations. However, strong performance on traditional reasoning benchmarks does not guarantee true causal reasoning ability of LLMs, as high accuracy may still arise from memorizing semantic patterns instead of analyzing the underlying true causal structures. To bridge this critical gap, we propose a new causal reasoning benchmark, CausalFlip, designed to encourage the development of new LLM paradig","authors_text":"Jundong Li, Yaochen Zhu, Yuzhe Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-23T18:06:15Z","title":"CausalFlip: A Benchmark for LLM Causal Judgment Beyond Semantic Matching"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.20094","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:f7362de82640801d917df7195eb4700442c6aa30a956c0a08ff8612548662d70","target":"record","created_at":"2026-06-29T00:14:02Z","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":"ab328de9c080efcab435d2969ebc8641f41b63b64ed6f83ab87a5ab5f3ad1fa2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-23T18:06:15Z","title_canon_sha256":"e31d8eb7b78b57a65286b3403d274ecb9c36da6b0239a3ef1eb0ceed2c4f44ed"},"schema_version":"1.0","source":{"id":"2602.20094","kind":"arxiv","version":2}},"canonical_sha256":"60677a3eafb86b2b8a6beb5b8e5265a615a6ec0944ecd8e1d0e06954a02cb0ef","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"60677a3eafb86b2b8a6beb5b8e5265a615a6ec0944ecd8e1d0e06954a02cb0ef","first_computed_at":"2026-06-29T00:14:02.502097Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-29T00:14:02.502097Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iDyRx2LdjPwxHX11Y5yOvXFgO+V0wjbFijraLVbA8mxwIZqW7+xNsMVvwoIvU6iolBoLDBE6rRzC0sAv7+ZuAg==","signature_status":"signed_v1","signed_at":"2026-06-29T00:14:02.502569Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.20094","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f7362de82640801d917df7195eb4700442c6aa30a956c0a08ff8612548662d70","sha256:30e1ef32266226ab3a2fd4aadc635b62cc78014fa907f74815e6ff5cf28feb7c"],"state_sha256":"064b6ef55289f94cc731d90af0e5ffd79f4c23e532936305d115e83d38a9ade9"}