{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:U237FBZ6DJN4MHYCZMHMM2PGZ6","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":"24877a7a1b8d93d7477a5167bde7a1bffff71650f754f1e09f4f2c7745958487","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T17:52:04Z","title_canon_sha256":"34ba87154d58e3ac713a98240c57bea8cd13b3ddc9782f8110213cbf7dc09dde"},"schema_version":"1.0","source":{"id":"2606.06462","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.06462","created_at":"2026-06-05T01:15:46Z"},{"alias_kind":"arxiv_version","alias_value":"2606.06462v1","created_at":"2026-06-05T01:15:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.06462","created_at":"2026-06-05T01:15:46Z"},{"alias_kind":"pith_short_12","alias_value":"U237FBZ6DJN4","created_at":"2026-06-05T01:15:46Z"},{"alias_kind":"pith_short_16","alias_value":"U237FBZ6DJN4MHYC","created_at":"2026-06-05T01:15:46Z"},{"alias_kind":"pith_short_8","alias_value":"U237FBZ6","created_at":"2026-06-05T01:15:46Z"}],"graph_snapshots":[{"event_id":"sha256:bbc591031e722af7643b19ac55ad475142c9c6f74de361cde3db1d44c673989c","target":"graph","created_at":"2026-06-05T01:15:46Z","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/2606.06462/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmarks are fundamental for evaluating and advancing LLMs and MLLMs by providing standardized and explicit measures of performance. However, their construction is labor-intensive and hard to reuse, raising concerns about sustainability and scalability. Moreover, existing benchmarks often quickly reach performance saturation after their release, resulting in insufficient discrimination among state-of-the-art models. To address these challenges, we introduce Benchmark Agent, a fully autonomous agentic system designed for benchmark building. Our framework orchestrates the complete benchmark co","authors_text":"Bokang Yang, Dongming Wu, Peiwen Sun, Shiyun Xiong, Wencheng Han, Xiangyu Yue, Xiao-Hui Li, Yuang Ai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T17:52:04Z","title":"Benchmark Everything Everywhere All at Once"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.06462","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:d9e5146a5ffa52b3334f7fa025db2b673fbb5f5cb0f209fa06e86ff38767c165","target":"record","created_at":"2026-06-05T01:15:46Z","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":"24877a7a1b8d93d7477a5167bde7a1bffff71650f754f1e09f4f2c7745958487","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T17:52:04Z","title_canon_sha256":"34ba87154d58e3ac713a98240c57bea8cd13b3ddc9782f8110213cbf7dc09dde"},"schema_version":"1.0","source":{"id":"2606.06462","kind":"arxiv","version":1}},"canonical_sha256":"a6b7f2873e1a5bc61f02cb0ec669e6cf87ba78e8785af9ebed517c351b1ff362","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a6b7f2873e1a5bc61f02cb0ec669e6cf87ba78e8785af9ebed517c351b1ff362","first_computed_at":"2026-06-05T01:15:46.275576Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-05T01:15:46.275576Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LLq+JB+VhSIhRz9hGjiNtoqZHibb1T0Nr/8SaeyQoWw/oygFMciGuMDN56jVDhwzhwC35AeGzjNbUXaxV+M2Ag==","signature_status":"signed_v1","signed_at":"2026-06-05T01:15:46.276071Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.06462","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9e5146a5ffa52b3334f7fa025db2b673fbb5f5cb0f209fa06e86ff38767c165","sha256:bbc591031e722af7643b19ac55ad475142c9c6f74de361cde3db1d44c673989c"],"state_sha256":"da523cf133d3739a28a0d0d81d2aa10914b9a0dd9fb9d8c37261e682a84f632b"}