{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EEKNOJNH3NPIO6ZU6RQPGVKXJX","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":"a68599b15e9a0582b3aef49b4ae36f640f5b839bcf4e3d08437117843ec2004a","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T16:20:51Z","title_canon_sha256":"fd2bd41c287548fd92863647a6abcb8f2ff287080bf077adc3083aeb001322a7"},"schema_version":"1.0","source":{"id":"2605.10787","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.10787","created_at":"2026-05-21T01:04:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.10787v2","created_at":"2026-05-21T01:04:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.10787","created_at":"2026-05-21T01:04:27Z"},{"alias_kind":"pith_short_12","alias_value":"EEKNOJNH3NPI","created_at":"2026-05-21T01:04:27Z"},{"alias_kind":"pith_short_16","alias_value":"EEKNOJNH3NPIO6ZU","created_at":"2026-05-21T01:04:27Z"},{"alias_kind":"pith_short_8","alias_value":"EEKNOJNH","created_at":"2026-05-21T01:04:27Z"}],"graph_snapshots":[{"event_id":"sha256:fcdbade56d1bcdf61773e2bf8516bebed4bfe78ea607ab96f7337bba0bb50a84","target":"graph","created_at":"2026-05-21T01:04:27Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"even top-tier models fail to exceed a 60% success rate, far trailing human performance 90%. Granular trajectory analysis identifies three fundamental bottlenecks: (1) tool retrieval saturation as action spaces scale; (2) over-confidence, where agents skip essential environment verifications; and (3) strategic defeatism, a tendency to rationalize failure rather than pursuing recovery."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The 7 stateful sandboxes and 300 tools derived from them sufficiently represent the interdependent, noisy conditions of real-world commercial software automation."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ComplexMCP benchmark shows current LLM agents achieve at most 60% success on interdependent tool tasks versus 90% for humans, due to tool retrieval saturation, over-confidence, and strategic defeatism."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLM agents reach under 60 percent success on interdependent tool tasks where humans hit 90 percent."}],"snapshot_sha256":"8057dbadd909034fb9f5b0fd4661a5d8d4f7110efb64319b3712927e8c22d70f"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T05:22:00.382797Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T14:35:51.574811Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T10:31:17.789485Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:58:52.075793Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.10787/integrity.json","findings":[],"snapshot_sha256":"24642f0bbdb1fe908881af6cae8eeb19b84ccbd618d70c09890213eae89c513e","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current LLM agents are proficient at calling isolated APIs but struggle with the \"last mile\" of commercial software automation. In real-world scenarios, tools are not independent; they are atomic, interdependent, and prone to environmental noise. We introduce $\\textbf{ComplexMCP}$, a benchmark designed to evaluate agents in these rigorous conditions. Built on the Model Context Protocol (MCP), $\\textbf{ComplexMCP}$ provides over 300 meticulously tested tools derived from 7 stateful sandboxes, ranging from office suites to financial systems. Unlike existing datasets, our benchmark utilizes a see","authors_text":"Hongyang Chen, Longyue Wang, Weihua Luo, Xue Yang, Yuanyang Li","cross_cats":["cs.SE"],"headline":"LLM agents reach under 60 percent success on interdependent tool tasks where humans hit 90 percent.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T16:20:51Z","title":"ComplexMCP: Evaluation of LLM Agents in Dynamic, Interdependent, and Large-Scale Tool Sandbox"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.10787","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T04:37:47.107372Z","id":"3b30be21-c64d-419d-8e6d-93bdba03522f","model_set":{"reader":"grok-4.3"},"one_line_summary":"ComplexMCP benchmark shows current LLM agents achieve at most 60% success on interdependent tool tasks versus 90% for humans, due to tool retrieval saturation, over-confidence, and strategic defeatism.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"LLM agents reach under 60 percent success on interdependent tool tasks where humans hit 90 percent.","strongest_claim":"even top-tier models fail to exceed a 60% success rate, far trailing human performance 90%. Granular trajectory analysis identifies three fundamental bottlenecks: (1) tool retrieval saturation as action spaces scale; (2) over-confidence, where agents skip essential environment verifications; and (3) strategic defeatism, a tendency to rationalize failure rather than pursuing recovery.","weakest_assumption":"The 7 stateful sandboxes and 300 tools derived from them sufficiently represent the interdependent, noisy conditions of real-world commercial software automation."}},"verdict_id":"3b30be21-c64d-419d-8e6d-93bdba03522f"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2f1b5999234d0de8970fdb0d4fe93795dcc289f8e0d4d1d07977a859767b68c8","target":"record","created_at":"2026-05-21T01:04:27Z","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":"a68599b15e9a0582b3aef49b4ae36f640f5b839bcf4e3d08437117843ec2004a","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T16:20:51Z","title_canon_sha256":"fd2bd41c287548fd92863647a6abcb8f2ff287080bf077adc3083aeb001322a7"},"schema_version":"1.0","source":{"id":"2605.10787","kind":"arxiv","version":2}},"canonical_sha256":"2114d725a7db5e877b34f460f355574dc44db64e817c18b70a0ec9cb6b6aa131","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2114d725a7db5e877b34f460f355574dc44db64e817c18b70a0ec9cb6b6aa131","first_computed_at":"2026-05-21T01:04:27.539391Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:27.539391Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"R7o3VBG1/wueQqfa56Q4rnpjA4aAtLWyorxJZCnUWFnL/r5JpgFSd39UnLqpvy+dEfp9S+pSFKq4MzaYcBo4Dw==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:27.540192Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.10787","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2f1b5999234d0de8970fdb0d4fe93795dcc289f8e0d4d1d07977a859767b68c8","sha256:fcdbade56d1bcdf61773e2bf8516bebed4bfe78ea607ab96f7337bba0bb50a84"],"state_sha256":"be9ced9353a4990773be74ef85b881f43338a548bc68f1db7a11d5c77f59686d"}