{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:GB3LHC6SVBYMDM6VQJGDVVGPMB","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":"3318b2b2d3574e1640b0e2181928f9c0f0d8c46353c6bb76494a18ab4cefc242","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-26T05:24:22Z","title_canon_sha256":"e3e3d4e132dda2144c93e0933794da45b0b0755c01f7fb2a67e1308021120fba"},"schema_version":"1.0","source":{"id":"2502.18836","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.18836","created_at":"2026-07-05T11:48:47Z"},{"alias_kind":"arxiv_version","alias_value":"2502.18836v2","created_at":"2026-07-05T11:48:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.18836","created_at":"2026-07-05T11:48:47Z"},{"alias_kind":"pith_short_12","alias_value":"GB3LHC6SVBYM","created_at":"2026-07-05T11:48:47Z"},{"alias_kind":"pith_short_16","alias_value":"GB3LHC6SVBYMDM6V","created_at":"2026-07-05T11:48:47Z"},{"alias_kind":"pith_short_8","alias_value":"GB3LHC6S","created_at":"2026-07-05T11:48:47Z"}],"graph_snapshots":[{"event_id":"sha256:8c7370f2304a696562f06e0a49d68e33102c8a0aaa184d76f0f436e6fbe7f374","target":"graph","created_at":"2026-07-05T11:48:47Z","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/2502.18836/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This benchmark suite provides a comprehensive evaluation framework for assessing both individual LLMs and multi-agent systems in Real-world planning and scheduling scenarios. The suite encompasses 14 designed planning and scheduling problems that progress from basic to highly complex, incorporating key aspects such as multi-agent coordination, inter-agent dependencies, and dynamic environmental disruptions. Each problem can be scaled along three dimensions: the number of parallel planning threads, the complexity of inter-dependencies, and the frequency of unexpected disruptions requiring Real-","authors_text":"Edward Y. Chang, Longling Geng","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-26T05:24:22Z","title":"REALM-Bench: A Benchmark for Evaluating Multi-Agent Systems on Real-world, Dynamic Planning and Scheduling Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.18836","kind":"arxiv","version":2},"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":"sha256:65ff5f725aca1ab92d6d5cea56ec042a6309ee2fae6b08fad6beaeeefb651f18","target":"record","created_at":"2026-07-05T11:48:47Z","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":"3318b2b2d3574e1640b0e2181928f9c0f0d8c46353c6bb76494a18ab4cefc242","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-02-26T05:24:22Z","title_canon_sha256":"e3e3d4e132dda2144c93e0933794da45b0b0755c01f7fb2a67e1308021120fba"},"schema_version":"1.0","source":{"id":"2502.18836","kind":"arxiv","version":2}},"canonical_sha256":"3076b38bd2a870c1b3d5824c3ad4cf60658f9f02109654432ae65b143ab1f7d9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3076b38bd2a870c1b3d5824c3ad4cf60658f9f02109654432ae65b143ab1f7d9","first_computed_at":"2026-07-05T11:48:47.761809Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:48:47.761809Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YdcDkoxc4FoGiCaU2TyMfAWySp6CxofkjmReBFBmPOAFYzrR91CZrJS4JTNUjnFvNxFGIc4J7qYfFQ/6FCkRBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:48:47.762373Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.18836","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:65ff5f725aca1ab92d6d5cea56ec042a6309ee2fae6b08fad6beaeeefb651f18","sha256:8c7370f2304a696562f06e0a49d68e33102c8a0aaa184d76f0f436e6fbe7f374"],"state_sha256":"de4a98771666b6546d7fb313773c34ef1243ffcee33773efa1a1f2843c2d5bc0"}