{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:J5Z7TJHM5KIX5WRPBMORWQGOIH","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":"c08badc7933d2977c4be666392ee74cc2279cc6d3d51ee8366bd1de4c2fcad38","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T10:01:01Z","title_canon_sha256":"9785614c733db9b39f3562149d4d5b4a56442df162a08fc3d15298fdee1cfa80"},"schema_version":"1.0","source":{"id":"2605.26789","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.26789","created_at":"2026-05-27T01:06:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.26789v1","created_at":"2026-05-27T01:06:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26789","created_at":"2026-05-27T01:06:12Z"},{"alias_kind":"pith_short_12","alias_value":"J5Z7TJHM5KIX","created_at":"2026-05-27T01:06:12Z"},{"alias_kind":"pith_short_16","alias_value":"J5Z7TJHM5KIX5WRP","created_at":"2026-05-27T01:06:12Z"},{"alias_kind":"pith_short_8","alias_value":"J5Z7TJHM","created_at":"2026-05-27T01:06:12Z"}],"graph_snapshots":[{"event_id":"sha256:d036f71b348cb6d94cc986fc57b1f39550a2ba7f300464dc8ab83369d4939adb","target":"graph","created_at":"2026-05-27T01:06:12Z","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/2605.26789/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Post-training is routinely evaluated through aggregate benchmark scores that treat multi-hop reasoning as a single capability -- as if a model that answers more questions correctly must be better at assembling facts. We show that this assumption can be misleading: recipes with statistically indistinguishable atomic knowledge produce composition behaviour separated by over 40 percentage points, a phenomenon we call composition collapse: the systematic failure to assemble stably-known facts into chains, invisible to aggregate metrics. We introduce a double-gate protocol that changes the estimand","authors_text":"Hongzhi Wang, Jie Chen, Meng Han, Wenpeng Xing, Xuyang Teng, Yunzhao Wei, Zhe Yu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T10:01:01Z","title":"Composition Collapse: Stable Factual Knowledge Does Not Imply Compositional Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26789","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":"sha256:101f202b5756073101e94818eec2276cb1731a47161c6e4f4efdf2023a727b46","target":"record","created_at":"2026-05-27T01:06:12Z","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":"c08badc7933d2977c4be666392ee74cc2279cc6d3d51ee8366bd1de4c2fcad38","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T10:01:01Z","title_canon_sha256":"9785614c733db9b39f3562149d4d5b4a56442df162a08fc3d15298fdee1cfa80"},"schema_version":"1.0","source":{"id":"2605.26789","kind":"arxiv","version":1}},"canonical_sha256":"4f73f9a4ecea917eda2f0b1d1b40ce41d295d63b197c7cdf0d61730a40f57898","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4f73f9a4ecea917eda2f0b1d1b40ce41d295d63b197c7cdf0d61730a40f57898","first_computed_at":"2026-05-27T01:06:12.772137Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:06:12.772137Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZM27+GnSbbKomApab6399eaLFmeMoL1LVLiLVW4Sl5wvDiID/iaQsxlMKMtZC7C71TIj6CgpTfCp2rSYqnFhBA==","signature_status":"signed_v1","signed_at":"2026-05-27T01:06:12.773199Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.26789","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:101f202b5756073101e94818eec2276cb1731a47161c6e4f4efdf2023a727b46","sha256:d036f71b348cb6d94cc986fc57b1f39550a2ba7f300464dc8ab83369d4939adb"],"state_sha256":"dac8c265169633176c95a35bd6d43b483dea6772ed1d3453ac8fd7e067362824"}