{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:76PVNWT6SMYBLYDYYPOLQ5KQXI","short_pith_number":"pith:76PVNWT6","canonical_record":{"source":{"id":"2505.13291","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-19T16:11:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"eedccf3d6b555b28f7e823bae864063a7950decf3006482f23c02604f8e95431","abstract_canon_sha256":"2817d57a94ba6622b2e6236b97f87642bb3d1cbb0f1db929cb009d5566a4ebbe"},"schema_version":"1.0"},"canonical_sha256":"ff9f56da7e933015e078c3dcb87550ba33a567111caad18cf35a2bc3b20896ed","source":{"kind":"arxiv","id":"2505.13291","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.13291","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"arxiv_version","alias_value":"2505.13291v1","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13291","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"pith_short_12","alias_value":"76PVNWT6SMYB","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"pith_short_16","alias_value":"76PVNWT6SMYBLYDY","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"pith_short_8","alias_value":"76PVNWT6","created_at":"2026-07-05T11:05:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:76PVNWT6SMYBLYDYYPOLQ5KQXI","target":"record","payload":{"canonical_record":{"source":{"id":"2505.13291","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-19T16:11:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"eedccf3d6b555b28f7e823bae864063a7950decf3006482f23c02604f8e95431","abstract_canon_sha256":"2817d57a94ba6622b2e6236b97f87642bb3d1cbb0f1db929cb009d5566a4ebbe"},"schema_version":"1.0"},"canonical_sha256":"ff9f56da7e933015e078c3dcb87550ba33a567111caad18cf35a2bc3b20896ed","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:27.844116Z","signature_b64":"FSdVa2QYCgCN5leoasKUAdqR8k/U9yE0/JHngy00lvb1QYswCAtUEWtfMNt4jBED2KztCdxeyjhTnCIK9UcFDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff9f56da7e933015e078c3dcb87550ba33a567111caad18cf35a2bc3b20896ed","last_reissued_at":"2026-07-05T11:05:27.843652Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:27.843652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.13291","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":"2026-07-05T11:05:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wRkdfZ8m0qlLDrp/NpixZp88hSS9iOMLsc+c3HB5UZR+3nmxCh2ujxouob3utnVVh3jDXVqPplzXdHFAlb1gAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T04:08:50.000093Z"},"content_sha256":"669f5773fc1f943f3057ecb8f5dd9133e47a7e1563af8b01a042ceb6b0cc81c0","schema_version":"1.0","event_id":"sha256:669f5773fc1f943f3057ecb8f5dd9133e47a7e1563af8b01a042ceb6b0cc81c0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:76PVNWT6SMYBLYDYYPOLQ5KQXI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TimeSeriesGym: A Scalable Benchmark for (Time Series) Machine Learning Engineering Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Artur Dubrawski, Gus Welter, Micha{\\l} Wili\\'nski, Mononito Goswami, Xinyu Li, Yifu Cai","submitted_at":"2025-05-19T16:11:23Z","abstract_excerpt":"We introduce TimeSeriesGym, a scalable benchmarking framework for evaluating Artificial Intelligence (AI) agents on time series machine learning engineering challenges. Existing benchmarks lack scalability, focus narrowly on model building in well-defined settings, and evaluate only a limited set of research artifacts (e.g., CSV submission files). To make AI agent benchmarking more relevant to the practice of machine learning engineering, our framework scales along two critical dimensions. First, recognizing that effective ML engineering requires a range of diverse skills, TimeSeriesGym incorp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13291","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/2505.13291/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":"2026-07-05T11:05:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9/fP1uR66BK841IibnAmeht1UgRLgyZCZBbEGCIt/6C+L6K1my5vIbOqTLI66UvgHzj3pJRZY/69odK95XErDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T04:08:50.000916Z"},"content_sha256":"a92ca7f4feca23f9692a4e01437bcd2af628930f6ccba8862ab7d5896e708526","schema_version":"1.0","event_id":"sha256:a92ca7f4feca23f9692a4e01437bcd2af628930f6ccba8862ab7d5896e708526"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI/bundle.json","state_url":"https://pith.science/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI/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":"2026-08-16T04:08:50Z","links":{"resolver":"https://pith.science/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI","bundle":"https://pith.science/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI/bundle.json","state":"https://pith.science/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/76PVNWT6SMYBLYDYYPOLQ5KQXI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:76PVNWT6SMYBLYDYYPOLQ5KQXI","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":"2817d57a94ba6622b2e6236b97f87642bb3d1cbb0f1db929cb009d5566a4ebbe","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-19T16:11:23Z","title_canon_sha256":"eedccf3d6b555b28f7e823bae864063a7950decf3006482f23c02604f8e95431"},"schema_version":"1.0","source":{"id":"2505.13291","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.13291","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"arxiv_version","alias_value":"2505.13291v1","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13291","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"pith_short_12","alias_value":"76PVNWT6SMYB","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"pith_short_16","alias_value":"76PVNWT6SMYBLYDY","created_at":"2026-07-05T11:05:27Z"},{"alias_kind":"pith_short_8","alias_value":"76PVNWT6","created_at":"2026-07-05T11:05:27Z"}],"graph_snapshots":[{"event_id":"sha256:a92ca7f4feca23f9692a4e01437bcd2af628930f6ccba8862ab7d5896e708526","target":"graph","created_at":"2026-07-05T11:05:27Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.13291/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce TimeSeriesGym, a scalable benchmarking framework for evaluating Artificial Intelligence (AI) agents on time series machine learning engineering challenges. Existing benchmarks lack scalability, focus narrowly on model building in well-defined settings, and evaluate only a limited set of research artifacts (e.g., CSV submission files). To make AI agent benchmarking more relevant to the practice of machine learning engineering, our framework scales along two critical dimensions. First, recognizing that effective ML engineering requires a range of diverse skills, TimeSeriesGym incorp","authors_text":"Artur Dubrawski, Gus Welter, Micha{\\l} Wili\\'nski, Mononito Goswami, Xinyu Li, Yifu Cai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-19T16:11:23Z","title":"TimeSeriesGym: A Scalable Benchmark for (Time Series) Machine Learning Engineering Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13291","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":"sha256:669f5773fc1f943f3057ecb8f5dd9133e47a7e1563af8b01a042ceb6b0cc81c0","target":"record","created_at":"2026-07-05T11:05:27Z","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":"2817d57a94ba6622b2e6236b97f87642bb3d1cbb0f1db929cb009d5566a4ebbe","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-19T16:11:23Z","title_canon_sha256":"eedccf3d6b555b28f7e823bae864063a7950decf3006482f23c02604f8e95431"},"schema_version":"1.0","source":{"id":"2505.13291","kind":"arxiv","version":1}},"canonical_sha256":"ff9f56da7e933015e078c3dcb87550ba33a567111caad18cf35a2bc3b20896ed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ff9f56da7e933015e078c3dcb87550ba33a567111caad18cf35a2bc3b20896ed","first_computed_at":"2026-07-05T11:05:27.843652Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:05:27.843652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FSdVa2QYCgCN5leoasKUAdqR8k/U9yE0/JHngy00lvb1QYswCAtUEWtfMNt4jBED2KztCdxeyjhTnCIK9UcFDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:05:27.844116Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.13291","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:669f5773fc1f943f3057ecb8f5dd9133e47a7e1563af8b01a042ceb6b0cc81c0","sha256:a92ca7f4feca23f9692a4e01437bcd2af628930f6ccba8862ab7d5896e708526"],"state_sha256":"5833b20452208c15c9180f3db5482cfe79d98ad07be438c9816c42339c16f931"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rl5HL5iJ/XXnDWXB15ZeY2roFHpqc43O+rqrfyzSUozv/yg4J/q5YG74C1eUd/MjihCy9e6B03cf3QE5WVvvBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T04:08:50.006793Z","bundle_sha256":"65e97fbd821428020ca9534e7c63448aabf7fabad047f25ba248e2f367a00b7c"}}