{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:E5H2SCZSESW2AU33D6GOQSBKTE","short_pith_number":"pith:E5H2SCZS","canonical_record":{"source":{"id":"2607.29626","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-31T16:58:00Z","cross_cats_sorted":[],"title_canon_sha256":"f8104ac071fad6ac92680bf2dd07ec22eb1b84cad11a63bfe07161239f5589ea","abstract_canon_sha256":"029cf226f011a1298bb4fecb3000ece63f8aa5288b08822c326f3d378e5bf310"},"schema_version":"1.0"},"canonical_sha256":"274fa90b3224ada0537b1f8ce8482a991beef5a89bf5dd8d0a4f4966f415688b","source":{"kind":"arxiv","id":"2607.29626","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.29626","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.29626v1","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.29626","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"pith_short_12","alias_value":"E5H2SCZSESW2","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"pith_short_16","alias_value":"E5H2SCZSESW2AU33","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"pith_short_8","alias_value":"E5H2SCZS","created_at":"2026-08-03T01:40:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:E5H2SCZSESW2AU33D6GOQSBKTE","target":"record","payload":{"canonical_record":{"source":{"id":"2607.29626","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-31T16:58:00Z","cross_cats_sorted":[],"title_canon_sha256":"f8104ac071fad6ac92680bf2dd07ec22eb1b84cad11a63bfe07161239f5589ea","abstract_canon_sha256":"029cf226f011a1298bb4fecb3000ece63f8aa5288b08822c326f3d378e5bf310"},"schema_version":"1.0"},"canonical_sha256":"274fa90b3224ada0537b1f8ce8482a991beef5a89bf5dd8d0a4f4966f415688b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-03T01:40:54.120812Z","signature_b64":"cLotpkaIxSoERUig4zx3USPfyfwoMjaJfqraLx+taWFxleJsYhVePuS/PTSnvo0Rjknr6ZnxTq2CFJUtNME5Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"274fa90b3224ada0537b1f8ce8482a991beef5a89bf5dd8d0a4f4966f415688b","last_reissued_at":"2026-08-03T01:40:54.119301Z","signature_status":"signed_v1","first_computed_at":"2026-08-03T01:40:54.119301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.29626","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":"2026-08-03T01:40:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tj5mlcPZa11+62AdobKh81RJVzbD3DWaiRU6AglP1CEinLOUtGGUhYFEv4qOS0XdcxFx59SWe2i67q28ThycAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T14:08:11.847417Z"},"content_sha256":"c3ebd5a5a2a7620800461e5ea35d4135cdfdbdd93859dd08e377133b43f518ab","schema_version":"1.0","event_id":"sha256:c3ebd5a5a2a7620800461e5ea35d4135cdfdbdd93859dd08e377133b43f518ab"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:E5H2SCZSESW2AU33D6GOQSBKTE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AgentHPOBench: A Benchmark For Evaluating LLM Agents as Sequential Hyperparameter Optimizers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jie Zhou, Shuang Chen, Tianyu Huai, Tingshuo Fan, Xinchi Chen, Xuanjing Huang, Yining Zheng, Yuxin Wang","submitted_at":"2026-07-31T16:58:00Z","abstract_excerpt":"As LLMs evolve from code completion systems into autonomous scientific agents, evaluating their ability to conduct experiments has become increasingly important. Existing benchmarks typically focus on static code generation, paper replication, or final answer correctness, but do not directly assess whether agents can interpret experimental evidence and use it to guide subsequent hyperparameter decisions. To address this gap, we introduce AgentHPOBench, a sequential benchmark comprising 30 executable machine learning tasks across seven research categories. Each task begins with a validated base"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.29626","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/2607.29626/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":"2026-08-03T01:40:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qzW5fP1+Sijw3B6w0OLmVGAU4/BLmMO9W2MBvxCO9x0sj3n5rkqF42SAzuhRb3V8zs4eL2Rr3ShLtgx4dZOaAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T14:08:11.847953Z"},"content_sha256":"16dd5251dbb73de3b0f403c267fc56dc96066569862893202bd582aa40b9b60d","schema_version":"1.0","event_id":"sha256:16dd5251dbb73de3b0f403c267fc56dc96066569862893202bd582aa40b9b60d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/E5H2SCZSESW2AU33D6GOQSBKTE/bundle.json","state_url":"https://pith.science/pith/E5H2SCZSESW2AU33D6GOQSBKTE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/E5H2SCZSESW2AU33D6GOQSBKTE/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":"2026-08-04T14:08:11Z","links":{"resolver":"https://pith.science/pith/E5H2SCZSESW2AU33D6GOQSBKTE","bundle":"https://pith.science/pith/E5H2SCZSESW2AU33D6GOQSBKTE/bundle.json","state":"https://pith.science/pith/E5H2SCZSESW2AU33D6GOQSBKTE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/E5H2SCZSESW2AU33D6GOQSBKTE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:E5H2SCZSESW2AU33D6GOQSBKTE","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":"029cf226f011a1298bb4fecb3000ece63f8aa5288b08822c326f3d378e5bf310","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-31T16:58:00Z","title_canon_sha256":"f8104ac071fad6ac92680bf2dd07ec22eb1b84cad11a63bfe07161239f5589ea"},"schema_version":"1.0","source":{"id":"2607.29626","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.29626","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.29626v1","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.29626","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"pith_short_12","alias_value":"E5H2SCZSESW2","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"pith_short_16","alias_value":"E5H2SCZSESW2AU33","created_at":"2026-08-03T01:40:54Z"},{"alias_kind":"pith_short_8","alias_value":"E5H2SCZS","created_at":"2026-08-03T01:40:54Z"}],"graph_snapshots":[{"event_id":"sha256:16dd5251dbb73de3b0f403c267fc56dc96066569862893202bd582aa40b9b60d","target":"graph","created_at":"2026-08-03T01:40:54Z","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/2607.29626/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As LLMs evolve from code completion systems into autonomous scientific agents, evaluating their ability to conduct experiments has become increasingly important. Existing benchmarks typically focus on static code generation, paper replication, or final answer correctness, but do not directly assess whether agents can interpret experimental evidence and use it to guide subsequent hyperparameter decisions. To address this gap, we introduce AgentHPOBench, a sequential benchmark comprising 30 executable machine learning tasks across seven research categories. Each task begins with a validated base","authors_text":"Jie Zhou, Shuang Chen, Tianyu Huai, Tingshuo Fan, Xinchi Chen, Xuanjing Huang, Yining Zheng, Yuxin Wang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-31T16:58:00Z","title":"AgentHPOBench: A Benchmark For Evaluating LLM Agents as Sequential Hyperparameter Optimizers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.29626","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:c3ebd5a5a2a7620800461e5ea35d4135cdfdbdd93859dd08e377133b43f518ab","target":"record","created_at":"2026-08-03T01:40:54Z","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":"029cf226f011a1298bb4fecb3000ece63f8aa5288b08822c326f3d378e5bf310","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-31T16:58:00Z","title_canon_sha256":"f8104ac071fad6ac92680bf2dd07ec22eb1b84cad11a63bfe07161239f5589ea"},"schema_version":"1.0","source":{"id":"2607.29626","kind":"arxiv","version":1}},"canonical_sha256":"274fa90b3224ada0537b1f8ce8482a991beef5a89bf5dd8d0a4f4966f415688b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"274fa90b3224ada0537b1f8ce8482a991beef5a89bf5dd8d0a4f4966f415688b","first_computed_at":"2026-08-03T01:40:54.119301Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-03T01:40:54.119301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cLotpkaIxSoERUig4zx3USPfyfwoMjaJfqraLx+taWFxleJsYhVePuS/PTSnvo0Rjknr6ZnxTq2CFJUtNME5Cg==","signature_status":"signed_v1","signed_at":"2026-08-03T01:40:54.120812Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.29626","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c3ebd5a5a2a7620800461e5ea35d4135cdfdbdd93859dd08e377133b43f518ab","sha256:16dd5251dbb73de3b0f403c267fc56dc96066569862893202bd582aa40b9b60d"],"state_sha256":"8acce1f979d8455117612d92fcaa5c631123eaf0099d1ed41a5cba6a2f5966d9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PQPimxn0Ng7vdaoB10rOI7ZkrtuR0bokchGd+nviEyKKxb1qZ1bdcz1GKrY6peA+SEpuYfnVkl1qsOYfXxk+CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T14:08:11.897142Z","bundle_sha256":"efa833ede4021806c9e7ca4a726ed57b42ac9e8f7fa9516ac78691fdc967fe41"}}