{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:3SSO3FKMX75FYJ7CK23NWCFAQC","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":"f22545376f88f09c24af8d498290886ac2728684f35c5aa6dfe19cd7dec5cfa3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-11T17:20:29Z","title_canon_sha256":"bdac7bca839265a7c7ab8ab129471cef6ab5692b1fa9c92334712fc20ff6d281"},"schema_version":"1.0","source":{"id":"2606.13602","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.13602","created_at":"2026-06-12T01:10:12Z"},{"alias_kind":"arxiv_version","alias_value":"2606.13602v1","created_at":"2026-06-12T01:10:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.13602","created_at":"2026-06-12T01:10:12Z"},{"alias_kind":"pith_short_12","alias_value":"3SSO3FKMX75F","created_at":"2026-06-12T01:10:12Z"},{"alias_kind":"pith_short_16","alias_value":"3SSO3FKMX75FYJ7C","created_at":"2026-06-12T01:10:12Z"},{"alias_kind":"pith_short_8","alias_value":"3SSO3FKM","created_at":"2026-06-12T01:10:12Z"}],"graph_snapshots":[{"event_id":"sha256:4e747c1525794cbbefb935582d8d9e0f92741ac804abc7de22c0ae5becb9e411","target":"graph","created_at":"2026-06-12T01:10:12Z","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/2606.13602/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce EpiBench, a verifiable benchmark for short-horizon epigenomics analysis. EpiBench evaluates whether agents can make well-defined analysis decisions from realistic workflow states and return deterministically gradable answers. The benchmark includes 106 evaluations across CUT\\&Tag/CUT\\&RUN, ATAC-seq, ChIP-seq, and DNA methylation workflows. Across 5,088 valid trajectories from 16 model-harness pairs, no system passed a majority of attempts: GPT-5.5 / Pi led at 45.0\\% (143/318 attempts; 95\\% confidence interval (CI), 36.3--53.7), followed by GPT-5.5 / OpenAI Codex at 39.9\\% (127/318","authors_text":"Harihara Muralidharan, Kenny Workman, Reema Baskar, Soo Hee Lee, Tim Proctor","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-11T17:20:29Z","title":"EpiBench: Verifiable Evaluation of AI Agents on Epigenomics Analysis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.13602","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":"sha256:13c97aec659892bdd5d2ae3bcd467365d360617e77617f842cd48df7c940c2f1","target":"record","created_at":"2026-06-12T01:10:12Z","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":"f22545376f88f09c24af8d498290886ac2728684f35c5aa6dfe19cd7dec5cfa3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-11T17:20:29Z","title_canon_sha256":"bdac7bca839265a7c7ab8ab129471cef6ab5692b1fa9c92334712fc20ff6d281"},"schema_version":"1.0","source":{"id":"2606.13602","kind":"arxiv","version":1}},"canonical_sha256":"dca4ed954cbffa5c27e256b6db08a080820109713b5d8658b3c6536240e395e5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dca4ed954cbffa5c27e256b6db08a080820109713b5d8658b3c6536240e395e5","first_computed_at":"2026-06-12T01:10:12.519883Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-12T01:10:12.519883Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+ykXbhizFWQgwW60sK4RUerbGNdWwyIVH6mq3Ex6vUdwDXgKM5J4rNW06mJmzrQAZd2Fo9sTsYrK4cNRVmp8Bg==","signature_status":"signed_v1","signed_at":"2026-06-12T01:10:12.520758Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.13602","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:13c97aec659892bdd5d2ae3bcd467365d360617e77617f842cd48df7c940c2f1","sha256:4e747c1525794cbbefb935582d8d9e0f92741ac804abc7de22c0ae5becb9e411"],"state_sha256":"2d9c09d40250aa296bc34b4d2883d1e1e1e70ef355b6e5c074d1028f1634ab1c"}