{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MXAUWQ6IRVAVGWMVUE5EP45WPM","short_pith_number":"pith:MXAUWQ6I","canonical_record":{"source":{"id":"2605.17554","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-17T17:32:52Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"af72164c678330ecf1bc9e56063a94ef6958b05859230e91c81097f42a569bd8","abstract_canon_sha256":"e22f6c995cfc738dabc25581f0bd34a1665d2646bcb969706211a3682267362c"},"schema_version":"1.0"},"canonical_sha256":"65c14b43c88d41535995a13a47f3b67b34bd63b10bd5f464bb16e476d6a6318b","source":{"kind":"arxiv","id":"2605.17554","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.17554","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"arxiv_version","alias_value":"2605.17554v1","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17554","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"pith_short_12","alias_value":"MXAUWQ6IRVAV","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"pith_short_16","alias_value":"MXAUWQ6IRVAVGWMV","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"pith_short_8","alias_value":"MXAUWQ6I","created_at":"2026-05-20T00:04:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MXAUWQ6IRVAVGWMVUE5EP45WPM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.17554","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-17T17:32:52Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"af72164c678330ecf1bc9e56063a94ef6958b05859230e91c81097f42a569bd8","abstract_canon_sha256":"e22f6c995cfc738dabc25581f0bd34a1665d2646bcb969706211a3682267362c"},"schema_version":"1.0"},"canonical_sha256":"65c14b43c88d41535995a13a47f3b67b34bd63b10bd5f464bb16e476d6a6318b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:45.461828Z","signature_b64":"6yhcDF8j81AJLfwnV2x74NIX6jJAjIZmGVOI/IvEYV+i9vJBPpBuw4t90xBzeRFkv1sNGJCKzqJwSM7qyaa3CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"65c14b43c88d41535995a13a47f3b67b34bd63b10bd5f464bb16e476d6a6318b","last_reissued_at":"2026-05-20T00:04:45.460846Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:45.460846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.17554","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2lhfo2KYjs/REQvZvVGSu+1AZYftvODz7QJ2vxbdJvkzwjfWU5SfRoTsJX9UVLlYiUCPe7WDbMgiQ0X+qmDhDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T15:39:07.869991Z"},"content_sha256":"d65ded350982ad3d8e3da8b7a6a538d57ce4fa5fd2c9105fb48fbfb5943a2fd0","schema_version":"1.0","event_id":"sha256:d65ded350982ad3d8e3da8b7a6a538d57ce4fa5fd2c9105fb48fbfb5943a2fd0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MXAUWQ6IRVAVGWMVUE5EP45WPM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating Deep Research Agents on Expert Consulting Work: A Benchmark with Verifiers, Rubrics, and Cognitive Traps","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Aman Saksena, Divyansh Sahu, Tanmay Asthana","submitted_at":"2026-05-17T17:32:52Z","abstract_excerpt":"Frontier deep research agents (DRAs) plan a research task, synthesize across documents, and return a structured deliverable on demand. They are being deployed in enterprise workflows faster than they are being evaluated. Existing benchmarks measure factual recall, single-hop QA, or generic agentic skill, missing the multi-document, decision-grade work DRAs are deployed to produce. We introduce a benchmark targeting the structured analytical deliverables that fill a management consultant's typical week.\n  We grade three frontier agents, namely Claude Opus 4.6 with web search, OpenAI o3-deep-res"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17554","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.17554/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.605579Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T21:21:57.539251Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"492ff49601fd25e8b0440ef023112fb9acf2d56371e65d2dc0402f1f8c0528c5"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0C5GdV9bwTmmWCi1Xyl6gJoyxDlJ3SYYNapiiaIPn/hdwhEAli7zYhJjZ4ktBDB1QgirBZNDqbdNmGMZJBMWDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T15:39:07.870680Z"},"content_sha256":"5e153fa2838e7038f6f19429231fbd28343aa1e3c05de9822d0ac9e82f00c1dc","schema_version":"1.0","event_id":"sha256:5e153fa2838e7038f6f19429231fbd28343aa1e3c05de9822d0ac9e82f00c1dc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM/bundle.json","state_url":"https://pith.science/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-27T15:39:07Z","links":{"resolver":"https://pith.science/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM","bundle":"https://pith.science/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM/bundle.json","state":"https://pith.science/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MXAUWQ6IRVAVGWMVUE5EP45WPM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MXAUWQ6IRVAVGWMVUE5EP45WPM","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":"e22f6c995cfc738dabc25581f0bd34a1665d2646bcb969706211a3682267362c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-17T17:32:52Z","title_canon_sha256":"af72164c678330ecf1bc9e56063a94ef6958b05859230e91c81097f42a569bd8"},"schema_version":"1.0","source":{"id":"2605.17554","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.17554","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"arxiv_version","alias_value":"2605.17554v1","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17554","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"pith_short_12","alias_value":"MXAUWQ6IRVAV","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"pith_short_16","alias_value":"MXAUWQ6IRVAVGWMV","created_at":"2026-05-20T00:04:45Z"},{"alias_kind":"pith_short_8","alias_value":"MXAUWQ6I","created_at":"2026-05-20T00:04:45Z"}],"graph_snapshots":[{"event_id":"sha256:5e153fa2838e7038f6f19429231fbd28343aa1e3c05de9822d0ac9e82f00c1dc","target":"graph","created_at":"2026-05-20T00:04:45Z","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":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.605579Z","status":"skipped","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T21:21:57.539251Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.17554/integrity.json","findings":[],"snapshot_sha256":"492ff49601fd25e8b0440ef023112fb9acf2d56371e65d2dc0402f1f8c0528c5","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Frontier deep research agents (DRAs) plan a research task, synthesize across documents, and return a structured deliverable on demand. They are being deployed in enterprise workflows faster than they are being evaluated. Existing benchmarks measure factual recall, single-hop QA, or generic agentic skill, missing the multi-document, decision-grade work DRAs are deployed to produce. We introduce a benchmark targeting the structured analytical deliverables that fill a management consultant's typical week.\n  We grade three frontier agents, namely Claude Opus 4.6 with web search, OpenAI o3-deep-res","authors_text":"Aman Saksena, Divyansh Sahu, Tanmay Asthana","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-17T17:32:52Z","title":"Evaluating Deep Research Agents on Expert Consulting Work: A Benchmark with Verifiers, Rubrics, and Cognitive Traps"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17554","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:d65ded350982ad3d8e3da8b7a6a538d57ce4fa5fd2c9105fb48fbfb5943a2fd0","target":"record","created_at":"2026-05-20T00:04:45Z","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":"e22f6c995cfc738dabc25581f0bd34a1665d2646bcb969706211a3682267362c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-17T17:32:52Z","title_canon_sha256":"af72164c678330ecf1bc9e56063a94ef6958b05859230e91c81097f42a569bd8"},"schema_version":"1.0","source":{"id":"2605.17554","kind":"arxiv","version":1}},"canonical_sha256":"65c14b43c88d41535995a13a47f3b67b34bd63b10bd5f464bb16e476d6a6318b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"65c14b43c88d41535995a13a47f3b67b34bd63b10bd5f464bb16e476d6a6318b","first_computed_at":"2026-05-20T00:04:45.460846Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:45.460846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6yhcDF8j81AJLfwnV2x74NIX6jJAjIZmGVOI/IvEYV+i9vJBPpBuw4t90xBzeRFkv1sNGJCKzqJwSM7qyaa3CQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:45.461828Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.17554","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d65ded350982ad3d8e3da8b7a6a538d57ce4fa5fd2c9105fb48fbfb5943a2fd0","sha256:5e153fa2838e7038f6f19429231fbd28343aa1e3c05de9822d0ac9e82f00c1dc"],"state_sha256":"16350bf901644feed323e8b8264cd16b41d514587993f94dc753f9d5b76fca98"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XD34KzOn12HGhR0cPM9dAiWf4O2BNZKii8ldZLa1/ylhxqk7FbLweCNjL59qfzi+AdRcP6HYvlTQmVYp8ftyBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T15:39:07.874219Z","bundle_sha256":"0d0919eb545ba0d9bb5476e8bd8d229f2bf299054d7de1c076736495d29e7cc2"}}