{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:BCT5MM4JHNHP4Y3FM7H7YR2YJD","short_pith_number":"pith:BCT5MM4J","canonical_record":{"source":{"id":"2607.21217","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T11:31:38Z","cross_cats_sorted":[],"title_canon_sha256":"8d2c62e78af994189bd439f16d1077a9d71b29ec3b90e04ea6a6679bac01bfa5","abstract_canon_sha256":"ce9195cfd93d435e0b6cdd08b7b54e255fc3673f96fd072a110c4c752086c979"},"schema_version":"1.0"},"canonical_sha256":"08a7d633893b4efe636567cffc475848d4f726052f640ac1a38ae9bc195d6fea","source":{"kind":"arxiv","id":"2607.21217","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21217","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21217v1","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21217","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"pith_short_12","alias_value":"BCT5MM4JHNHP","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"pith_short_16","alias_value":"BCT5MM4JHNHP4Y3F","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"pith_short_8","alias_value":"BCT5MM4J","created_at":"2026-07-24T01:24:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:BCT5MM4JHNHP4Y3FM7H7YR2YJD","target":"record","payload":{"canonical_record":{"source":{"id":"2607.21217","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T11:31:38Z","cross_cats_sorted":[],"title_canon_sha256":"8d2c62e78af994189bd439f16d1077a9d71b29ec3b90e04ea6a6679bac01bfa5","abstract_canon_sha256":"ce9195cfd93d435e0b6cdd08b7b54e255fc3673f96fd072a110c4c752086c979"},"schema_version":"1.0"},"canonical_sha256":"08a7d633893b4efe636567cffc475848d4f726052f640ac1a38ae9bc195d6fea","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T01:24:19.592539Z","signature_b64":"EbZWtrcnpLQph4dTb7FuI1tQ+H3S9ddcsrW8QE56Z/oGwUxFcF4LFvf8BJWJ/khJDrDN4YzzXvcROPl+0cdeAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08a7d633893b4efe636567cffc475848d4f726052f640ac1a38ae9bc195d6fea","last_reissued_at":"2026-07-24T01:24:19.591590Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T01:24:19.591590Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.21217","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":"2026-07-24T01:24:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mhtuAqCMogH59+WjZk/mwhLSENcXl1FgO+18pxvYMG90fguAin5u0MLN6BPf+fTkQBjyGh7d1IBqzT9Q5SQ1DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T10:51:17.771235Z"},"content_sha256":"b8f9a919d8a7b6db591d0003f11e0736895bc9f59f7b7a534d3024997e774547","schema_version":"1.0","event_id":"sha256:b8f9a919d8a7b6db591d0003f11e0736895bc9f59f7b7a534d3024997e774547"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:BCT5MM4JHNHP4Y3FM7H7YR2YJD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Caijun Xu, Changyi Xiao, Chuyu Zhang, Dan Huang, David Lo, Jiyuan He, Lin Qiu, Shibo Hong, Xuezhi Cao, Yixin Cao, Zhongyuan Peng","submitted_at":"2026-07-23T11:31:38Z","abstract_excerpt":"The recent emergence of vibe-coding workflows is changing what coding agents are expected to do. Instead of merely completing code under fully specified instructions, agents are increasingly expected to transform incomplete product intent into working software by combining various abilities including planning, requirement clarification, tool use, debugging, and repository-level construction. Yet existing benchmarks have not fully caught up with this shift, evaluating agents on static, fully specified tasks.\n  In this paper, we introduce ICAE-Bench, a benchmark for evaluating coding agents unde"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21217","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/2607.21217/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-24T01:24:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8s9nvCXuKSDY3UtpwDl2n3w4U9HametStPCI9yVXXAfAbjVoWFjEhY+BicazphCVLx7338u0oXf8ieVdBa1kBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T10:51:17.771710Z"},"content_sha256":"2595d01d420fc3edc7725b032ba6af9aa87ea0c7867d8e5d9da0b167b9f0f4ab","schema_version":"1.0","event_id":"sha256:2595d01d420fc3edc7725b032ba6af9aa87ea0c7867d8e5d9da0b167b9f0f4ab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD/bundle.json","state_url":"https://pith.science/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD/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":"2026-08-04T10:51:17Z","links":{"resolver":"https://pith.science/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD","bundle":"https://pith.science/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD/bundle.json","state":"https://pith.science/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BCT5MM4JHNHP4Y3FM7H7YR2YJD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BCT5MM4JHNHP4Y3FM7H7YR2YJD","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":"ce9195cfd93d435e0b6cdd08b7b54e255fc3673f96fd072a110c4c752086c979","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T11:31:38Z","title_canon_sha256":"8d2c62e78af994189bd439f16d1077a9d71b29ec3b90e04ea6a6679bac01bfa5"},"schema_version":"1.0","source":{"id":"2607.21217","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21217","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21217v1","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21217","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"pith_short_12","alias_value":"BCT5MM4JHNHP","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"pith_short_16","alias_value":"BCT5MM4JHNHP4Y3F","created_at":"2026-07-24T01:24:19Z"},{"alias_kind":"pith_short_8","alias_value":"BCT5MM4J","created_at":"2026-07-24T01:24:19Z"}],"graph_snapshots":[{"event_id":"sha256:2595d01d420fc3edc7725b032ba6af9aa87ea0c7867d8e5d9da0b167b9f0f4ab","target":"graph","created_at":"2026-07-24T01:24:19Z","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/2607.21217/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The recent emergence of vibe-coding workflows is changing what coding agents are expected to do. Instead of merely completing code under fully specified instructions, agents are increasingly expected to transform incomplete product intent into working software by combining various abilities including planning, requirement clarification, tool use, debugging, and repository-level construction. Yet existing benchmarks have not fully caught up with this shift, evaluating agents on static, fully specified tasks.\n  In this paper, we introduce ICAE-Bench, a benchmark for evaluating coding agents unde","authors_text":"Caijun Xu, Changyi Xiao, Chuyu Zhang, Dan Huang, David Lo, Jiyuan He, Lin Qiu, Shibo Hong, Xuezhi Cao, Yixin Cao, Zhongyuan Peng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T11:31:38Z","title":"ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21217","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:b8f9a919d8a7b6db591d0003f11e0736895bc9f59f7b7a534d3024997e774547","target":"record","created_at":"2026-07-24T01:24:19Z","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":"ce9195cfd93d435e0b6cdd08b7b54e255fc3673f96fd072a110c4c752086c979","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T11:31:38Z","title_canon_sha256":"8d2c62e78af994189bd439f16d1077a9d71b29ec3b90e04ea6a6679bac01bfa5"},"schema_version":"1.0","source":{"id":"2607.21217","kind":"arxiv","version":1}},"canonical_sha256":"08a7d633893b4efe636567cffc475848d4f726052f640ac1a38ae9bc195d6fea","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"08a7d633893b4efe636567cffc475848d4f726052f640ac1a38ae9bc195d6fea","first_computed_at":"2026-07-24T01:24:19.591590Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-24T01:24:19.591590Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EbZWtrcnpLQph4dTb7FuI1tQ+H3S9ddcsrW8QE56Z/oGwUxFcF4LFvf8BJWJ/khJDrDN4YzzXvcROPl+0cdeAw==","signature_status":"signed_v1","signed_at":"2026-07-24T01:24:19.592539Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.21217","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b8f9a919d8a7b6db591d0003f11e0736895bc9f59f7b7a534d3024997e774547","sha256:2595d01d420fc3edc7725b032ba6af9aa87ea0c7867d8e5d9da0b167b9f0f4ab"],"state_sha256":"f073f72932e34b113419d554df0c6a1aa368549d826af567ab95b6d86f511bbd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9L6AX0qHu37dF6IaIh5v7pBABVAgjaO/FGoAAcit5xk8SRfBJyDdICIdI2FLzw5Wp2Nc/vtFj0fRv0oYpzlNDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T10:51:17.775208Z","bundle_sha256":"1a29fbe0c9043602abfda46404c89dc3600a4989f126a7629f3456c9c986fb14"}}