{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WGZU5ZZRFNWOMBS7SUDALDO5DG","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":"b083ad7040951d2947e1a5eba1ab92aa06dd37c9ec904fab03f834776cac2447","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T14:52:42Z","title_canon_sha256":"23a75f73fe909aec46515f85458b42ee8ba2d56e9f3eb371f09fdafc389d9111"},"schema_version":"1.0","source":{"id":"2607.12835","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.12835","created_at":"2026-07-15T01:22:20Z"},{"alias_kind":"arxiv_version","alias_value":"2607.12835v1","created_at":"2026-07-15T01:22:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.12835","created_at":"2026-07-15T01:22:20Z"},{"alias_kind":"pith_short_12","alias_value":"WGZU5ZZRFNWO","created_at":"2026-07-15T01:22:20Z"},{"alias_kind":"pith_short_16","alias_value":"WGZU5ZZRFNWOMBS7","created_at":"2026-07-15T01:22:20Z"},{"alias_kind":"pith_short_8","alias_value":"WGZU5ZZR","created_at":"2026-07-15T01:22:20Z"}],"graph_snapshots":[{"event_id":"sha256:fc172db3bda7eda2b7dfe84b0a6baab29551fd61821e4a52d04522fc6fde3a83","target":"graph","created_at":"2026-07-15T01:22:20Z","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/2607.12835/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Rubric-based evaluation is a promising approach for assessing open-ended outputs from LLM-based research agents, particularly in paper reproduction, where direct paper-to-repository comparison is prone to hallucination. However, constructing paper-specific rubrics requires substantial expert effort, limiting the scalability of benchmarks such as PaperBench. In this work, we present, to our knowledge, the first systematic meta-evaluation of LLM-generated rubrics for paper reproduction. We reformulate rubrics into a checklist-style format and evaluate four generation settings across two backbone","authors_text":"Chenghua Lin, Hanhua Hong, Jiaoyan Chen, Jung-jae Kim, Luu Gia Huy, Sophia Ananiadou, Yizhi Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T14:52:42Z","title":"Can LLMs Write Reliable Rubrics? A Meta-Evaluation for Experiment Reproduction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.12835","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":"sha256:55d900f20b94d5dd244600666020278557abc7ed7f9bd423ccde9c82d77b220c","target":"record","created_at":"2026-07-15T01:22:20Z","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":"b083ad7040951d2947e1a5eba1ab92aa06dd37c9ec904fab03f834776cac2447","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T14:52:42Z","title_canon_sha256":"23a75f73fe909aec46515f85458b42ee8ba2d56e9f3eb371f09fdafc389d9111"},"schema_version":"1.0","source":{"id":"2607.12835","kind":"arxiv","version":1}},"canonical_sha256":"b1b34ee7312b6ce6065f9506058ddd19b3b17aa61195af91233eea6394f6027b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b1b34ee7312b6ce6065f9506058ddd19b3b17aa61195af91233eea6394f6027b","first_computed_at":"2026-07-15T01:22:20.922754Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-15T01:22:20.922754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DEVi1NuyhU6xo8J4nPzTFLDzS898eOu6ABDpgUe71Nmpn2Llw/Coe4X+gMDhYjC0XL4OoFotxhnNfTmg5NIoCQ==","signature_status":"signed_v1","signed_at":"2026-07-15T01:22:20.923826Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.12835","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:55d900f20b94d5dd244600666020278557abc7ed7f9bd423ccde9c82d77b220c","sha256:fc172db3bda7eda2b7dfe84b0a6baab29551fd61821e4a52d04522fc6fde3a83"],"state_sha256":"7be086588c558b2fb181003cb033499cb243e9b83ee7cbf2c70129881e90eecf"}