{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EPDCRWPQTHVH4LXCCV3ZDBGXNC","short_pith_number":"pith:EPDCRWPQ","schema_version":"1.0","canonical_sha256":"23c628d9f099ea7e2ee215779184d76881abd37290a72eda5c3bd77ec7cdf30a","source":{"kind":"arxiv","id":"2604.00594","version":2},"attestation_state":"computed","paper":{"title":"Agent psychometrics: Task-level performance prediction in agentic coding benchmarks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chris Ge, Daniel Fried, Daria Kryvosheieva, Kaivalya Hariharan, Uzay Girit","submitted_at":"2026-04-01T07:59:59Z","abstract_excerpt":"As the focus in LLM-based coding shifts from static single-step code generation to multi-step agentic interaction with tools and environments, understanding which tasks will challenge agents and why becomes increasingly difficult. This is compounded by current practice: agent performance is typically measured by aggregate pass rates on benchmarks, but single-number metrics obscure the diversity of tasks within a benchmark. We present a framework for predicting success or failure on individual tasks tailored to the agentic coding regime. Our approach augments Item Response Theory (IRT) with ric"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2604.00594","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-01T07:59:59Z","cross_cats_sorted":[],"title_canon_sha256":"3a0ff1fcd1a0b893f86c67e8801785ee4e77626319bdb5f379c351c321b2325e","abstract_canon_sha256":"0b9fdc285353d5b68c3079dd81bd1460a2e677f9e25a2107901f8a5803b0528b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T02:21:34.774834Z","signature_b64":"Gv3e93XQejMUnsuuA7Ag3du/sbAUKXQohFczxRS20B5YofGxg2iOq472EYEcscrwOTiwTxMiBNm3u6qlkhaVDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23c628d9f099ea7e2ee215779184d76881abd37290a72eda5c3bd77ec7cdf30a","last_reissued_at":"2026-07-21T02:21:34.773921Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T02:21:34.773921Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agent psychometrics: Task-level performance prediction in agentic coding benchmarks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chris Ge, Daniel Fried, Daria Kryvosheieva, Kaivalya Hariharan, Uzay Girit","submitted_at":"2026-04-01T07:59:59Z","abstract_excerpt":"As the focus in LLM-based coding shifts from static single-step code generation to multi-step agentic interaction with tools and environments, understanding which tasks will challenge agents and why becomes increasingly difficult. This is compounded by current practice: agent performance is typically measured by aggregate pass rates on benchmarks, but single-number metrics obscure the diversity of tasks within a benchmark. We present a framework for predicting success or failure on individual tasks tailored to the agentic coding regime. Our approach augments Item Response Theory (IRT) with ric"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.00594","kind":"arxiv","version":2},"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/2604.00594/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2604.00594","created_at":"2026-07-21T02:21:34.774387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.00594v2","created_at":"2026-07-21T02:21:34.774387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.00594","created_at":"2026-07-21T02:21:34.774387+00:00"},{"alias_kind":"pith_short_12","alias_value":"EPDCRWPQTHVH","created_at":"2026-07-21T02:21:34.774387+00:00"},{"alias_kind":"pith_short_16","alias_value":"EPDCRWPQTHVH4LXC","created_at":"2026-07-21T02:21:34.774387+00:00"},{"alias_kind":"pith_short_8","alias_value":"EPDCRWPQ","created_at":"2026-07-21T02:21:34.774387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.23950","citing_title":"Stop Comparing LLM Agents Without Disclosing the Harness","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC","json":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC.json","graph_json":"https://pith.science/api/pith-number/EPDCRWPQTHVH4LXCCV3ZDBGXNC/graph.json","events_json":"https://pith.science/api/pith-number/EPDCRWPQTHVH4LXCCV3ZDBGXNC/events.json","paper":"https://pith.science/paper/EPDCRWPQ"},"agent_actions":{"view_html":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC","download_json":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC.json","view_paper":"https://pith.science/paper/EPDCRWPQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.00594&json=true","fetch_graph":"https://pith.science/api/pith-number/EPDCRWPQTHVH4LXCCV3ZDBGXNC/graph.json","fetch_events":"https://pith.science/api/pith-number/EPDCRWPQTHVH4LXCCV3ZDBGXNC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC/action/storage_attestation","attest_author":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC/action/author_attestation","sign_citation":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC/action/citation_signature","submit_replication":"https://pith.science/pith/EPDCRWPQTHVH4LXCCV3ZDBGXNC/action/replication_record"}},"created_at":"2026-07-21T02:21:34.774387+00:00","updated_at":"2026-07-21T02:21:34.774387+00:00"}