{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5PXD56BA7H4PJLD3PTT6KUHS3O","short_pith_number":"pith:5PXD56BA","schema_version":"1.0","canonical_sha256":"ebee3ef820f9f8f4ac7b7ce7e550f2dbb87a1d759d479bc04e3f65f5d8616958","source":{"kind":"arxiv","id":"2605.24110","version":1},"attestation_state":"computed","paper":{"title":"EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haiyang Shen, Kuan Li, Liang Chen, Wendong Xu, Xuanzhong Chen, Yun Ma","submitted_at":"2026-05-22T18:17:28Z","abstract_excerpt":"Coding agents are increasingly used as iterative development partners, but most benchmarks still evaluate one specification followed by one final assessment. This leaves out a basic question: can an agent keep its own codebase working as requirements change? We introduce EvoCode-Bench, a benchmark of 26 stateful coding tasks and 227 evaluated rounds. Each task preserves the agent's workspace for 5-15 rounds, states requirements through observable behavior, and uses cumulative executable tests to check new requirements and still-active prior ones. We evaluate 13 coding agents with two metrics: "},"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":"2605.24110","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-22T18:17:28Z","cross_cats_sorted":[],"title_canon_sha256":"1cefa0545aca986aaf4585b61f48484df9427388be9d2c5ecad96e6f5c47fe28","abstract_canon_sha256":"d35f80681186fc930709345a8b9333f237fb24b5471e0ff0488312945ddeb4c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:02:46.719479Z","signature_b64":"dpo7h57Sx7dQJt1fM1gIto/tz2PRtYOWrIFTyRPKxl8XmWFOdvHvv9dTQJf0kJz2U6S8cjNKt/dE2c2grpviAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ebee3ef820f9f8f4ac7b7ce7e550f2dbb87a1d759d479bc04e3f65f5d8616958","last_reissued_at":"2026-05-26T01:02:46.718675Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:02:46.718675Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Haiyang Shen, Kuan Li, Liang Chen, Wendong Xu, Xuanzhong Chen, Yun Ma","submitted_at":"2026-05-22T18:17:28Z","abstract_excerpt":"Coding agents are increasingly used as iterative development partners, but most benchmarks still evaluate one specification followed by one final assessment. This leaves out a basic question: can an agent keep its own codebase working as requirements change? We introduce EvoCode-Bench, a benchmark of 26 stateful coding tasks and 227 evaluated rounds. Each task preserves the agent's workspace for 5-15 rounds, states requirements through observable behavior, and uses cumulative executable tests to check new requirements and still-active prior ones. We evaluate 13 coding agents with two metrics: "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24110","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/2605.24110/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":"2605.24110","created_at":"2026-05-26T01:02:46.718820+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24110v1","created_at":"2026-05-26T01:02:46.718820+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24110","created_at":"2026-05-26T01:02:46.718820+00:00"},{"alias_kind":"pith_short_12","alias_value":"5PXD56BA7H4P","created_at":"2026-05-26T01:02:46.718820+00:00"},{"alias_kind":"pith_short_16","alias_value":"5PXD56BA7H4PJLD3","created_at":"2026-05-26T01:02:46.718820+00:00"},{"alias_kind":"pith_short_8","alias_value":"5PXD56BA","created_at":"2026-05-26T01:02:46.718820+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O","json":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O.json","graph_json":"https://pith.science/api/pith-number/5PXD56BA7H4PJLD3PTT6KUHS3O/graph.json","events_json":"https://pith.science/api/pith-number/5PXD56BA7H4PJLD3PTT6KUHS3O/events.json","paper":"https://pith.science/paper/5PXD56BA"},"agent_actions":{"view_html":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O","download_json":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O.json","view_paper":"https://pith.science/paper/5PXD56BA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24110&json=true","fetch_graph":"https://pith.science/api/pith-number/5PXD56BA7H4PJLD3PTT6KUHS3O/graph.json","fetch_events":"https://pith.science/api/pith-number/5PXD56BA7H4PJLD3PTT6KUHS3O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O/action/storage_attestation","attest_author":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O/action/author_attestation","sign_citation":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O/action/citation_signature","submit_replication":"https://pith.science/pith/5PXD56BA7H4PJLD3PTT6KUHS3O/action/replication_record"}},"created_at":"2026-05-26T01:02:46.718820+00:00","updated_at":"2026-05-26T01:02:46.718820+00:00"}