{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6BYBBB4ULVLJERDALVZSQ6WWC2","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":"b0297fdb50aacd876f36ae9384dc11ea9f88b245524541775008b36f7199529e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T07:11:56Z","title_canon_sha256":"bf713b7908e1e49fdc2dd3f230f97f6a7498d5750f84a9c7c7a83e7039ec2ab1"},"schema_version":"1.0","source":{"id":"2606.09118","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09118","created_at":"2026-06-09T02:08:00Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09118v1","created_at":"2026-06-09T02:08:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09118","created_at":"2026-06-09T02:08:00Z"},{"alias_kind":"pith_short_12","alias_value":"6BYBBB4ULVLJ","created_at":"2026-06-09T02:08:00Z"},{"alias_kind":"pith_short_16","alias_value":"6BYBBB4ULVLJERDA","created_at":"2026-06-09T02:08:00Z"},{"alias_kind":"pith_short_8","alias_value":"6BYBBB4U","created_at":"2026-06-09T02:08:00Z"}],"graph_snapshots":[{"event_id":"sha256:1734f1d0f1f7f943061451d8d76336104d4e00476a30c5b5bb6f54628d845518","target":"graph","created_at":"2026-06-09T02:08:00Z","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/2606.09118/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As LLM capabilities advance rapidly, the evaluation methods used to assess them increasingly lag behind. Traditional benchmarks relied on programmatic verification of narrow, surface-level constraints, but real-world instruction following and agentic tasks demand assessment of nuanced, context-dependent behaviors that resist simple scripted checks. We present a systematic analysis of expert-curated rubric-based evaluation as an alternative paradigm, drawing on empirical evidence from two domains: complex instruction following and enterprise agentic tasks. We first articulate five design princi","authors_text":"Edwin Chen, Liudas Panavas, Sushant Mehta","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T07:11:56Z","title":"ComplexConstraints and Beyond: Expert Rubrics for RLVR"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09118","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:fe1d34c4276351ecf4dee787a9e13fa61060860124de522d8e41b4c632301cab","target":"record","created_at":"2026-06-09T02:08:00Z","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":"b0297fdb50aacd876f36ae9384dc11ea9f88b245524541775008b36f7199529e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-08T07:11:56Z","title_canon_sha256":"bf713b7908e1e49fdc2dd3f230f97f6a7498d5750f84a9c7c7a83e7039ec2ab1"},"schema_version":"1.0","source":{"id":"2606.09118","kind":"arxiv","version":1}},"canonical_sha256":"f0701087945d569244605d73287ad6168f86db2f77a49fe004d651144ec1eb43","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f0701087945d569244605d73287ad6168f86db2f77a49fe004d651144ec1eb43","first_computed_at":"2026-06-09T02:08:00.251115Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:08:00.251115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UCNULKrbzsvz7zGWzevwW72/CEK2roP//jvAyQuRc6jnNgjxQn9GfEVMtYOLD5ZlGTRWCBYLEMSjUvWU6DT2DQ==","signature_status":"signed_v1","signed_at":"2026-06-09T02:08:00.251925Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.09118","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fe1d34c4276351ecf4dee787a9e13fa61060860124de522d8e41b4c632301cab","sha256:1734f1d0f1f7f943061451d8d76336104d4e00476a30c5b5bb6f54628d845518"],"state_sha256":"31f4d0c26cdb7bd809d1afcee51874474e811268f2311df5f73ebdc3cfb37b6b"}