{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Z7ZAMPUDHXH2W44JUCFVABSOTI","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":"ce6836f29b542bd52962abf62a7d303301d46051e739f76ccb7a619e4f3877c0","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-18T17:09:50Z","title_canon_sha256":"0e51878ed65ffea8ce52f9ddbeb87e9bb16ffa53f88d5cb8987073df8d7c062b"},"schema_version":"1.0","source":{"id":"2605.18663","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18663","created_at":"2026-05-20T00:06:13Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18663v1","created_at":"2026-05-20T00:06:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18663","created_at":"2026-05-20T00:06:13Z"},{"alias_kind":"pith_short_12","alias_value":"Z7ZAMPUDHXH2","created_at":"2026-05-20T00:06:13Z"},{"alias_kind":"pith_short_16","alias_value":"Z7ZAMPUDHXH2W44J","created_at":"2026-05-20T00:06:13Z"},{"alias_kind":"pith_short_8","alias_value":"Z7ZAMPUD","created_at":"2026-05-20T00:06:13Z"}],"graph_snapshots":[{"event_id":"sha256:5168d9564ecb5a3ae6ac1893cd5ce98f98c2f4fdb9067bc71f9963f52783661e","target":"graph","created_at":"2026-05-20T00:06:13Z","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":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.147771Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.18663/integrity.json","findings":[],"snapshot_sha256":"6d1e06b3aeeb9bad3c640681ad2e2b5589ff660034681b77cb0f0bbc0844f979","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As LLM benchmarks saturate, the evaluation community has pursued two strategies to increase difficulty: escalating knowledge demands (GPQA, HLE) or removing knowledge entirely in favor of abstract reasoning (ARC-AGI). The first conflates memorization with capability; the second divorces reasoning from the practical contexts in which it matters. We take a different approach. The Grounded Integration Measure (GIM) is a benchmark of 820 original problems (615 public, 205 private) where difficulty comes from integration; individual problems require coordinating multiple cognitive operations (const","authors_text":"Alexandre Rezende, Rohit Patel, Steven McClain","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-18T17:09:50Z","title":"GIM: Evaluating models via tasks that integrate multiple cognitive domains"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18663","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:d50c45618f02154902833f1083a0e0025cbe3953e82b2865e092659577f0b0f5","target":"record","created_at":"2026-05-20T00:06:13Z","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":"ce6836f29b542bd52962abf62a7d303301d46051e739f76ccb7a619e4f3877c0","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-18T17:09:50Z","title_canon_sha256":"0e51878ed65ffea8ce52f9ddbeb87e9bb16ffa53f88d5cb8987073df8d7c062b"},"schema_version":"1.0","source":{"id":"2605.18663","kind":"arxiv","version":1}},"canonical_sha256":"cff2063e833dcfab7389a08b50064e9a1e79b66902643571f9efa6edb1946581","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cff2063e833dcfab7389a08b50064e9a1e79b66902643571f9efa6edb1946581","first_computed_at":"2026-05-20T00:06:13.504596Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:06:13.504596Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Z/AB2cgh1gdhvzjhqtNjvQ31FjbC078gSBvSmOP87oMYHUVF/Z+GhiPQiNcMXMRmyPDUNWA2DlWd/D1xN5e5BQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:06:13.505311Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.18663","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d50c45618f02154902833f1083a0e0025cbe3953e82b2865e092659577f0b0f5","sha256:5168d9564ecb5a3ae6ac1893cd5ce98f98c2f4fdb9067bc71f9963f52783661e"],"state_sha256":"eb20a5eddb146e6c1c83ad007682cc653c3418276e1a6bf91770887979e65f74"}