{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TWAZLWRKMKI4P24ALIO7NVZ5VC","short_pith_number":"pith:TWAZLWRK","canonical_record":{"source":{"id":"2507.02083","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-02T18:41:44Z","cross_cats_sorted":[],"title_canon_sha256":"feb0ef878a6c0109182dc6fbcd58b522538afbd099e888bd26c54b1e97933922","abstract_canon_sha256":"2c6de74c5b65405e0f1d1fcf2b0e28c640725e39538018040791ca54cabedb96"},"schema_version":"1.0"},"canonical_sha256":"9d8195da2a6291c7eb805a1df6d73da888c631d7a6a58d3ff51df3940a4595fc","source":{"kind":"arxiv","id":"2507.02083","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.02083","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"arxiv_version","alias_value":"2507.02083v2","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.02083","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"pith_short_12","alias_value":"TWAZLWRKMKI4","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"pith_short_16","alias_value":"TWAZLWRKMKI4P24A","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"pith_short_8","alias_value":"TWAZLWRK","created_at":"2026-07-05T11:36:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TWAZLWRKMKI4P24ALIO7NVZ5VC","target":"record","payload":{"canonical_record":{"source":{"id":"2507.02083","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-02T18:41:44Z","cross_cats_sorted":[],"title_canon_sha256":"feb0ef878a6c0109182dc6fbcd58b522538afbd099e888bd26c54b1e97933922","abstract_canon_sha256":"2c6de74c5b65405e0f1d1fcf2b0e28c640725e39538018040791ca54cabedb96"},"schema_version":"1.0"},"canonical_sha256":"9d8195da2a6291c7eb805a1df6d73da888c631d7a6a58d3ff51df3940a4595fc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:36:34.031002Z","signature_b64":"q56DL4XYGlwLuWjy9ppNAvTyI5oTaBM80qXc+4z2tOj3MoSZddLOXEw3vlDTo6wWubn0vCXOnE9Vqqh9or0HCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d8195da2a6291c7eb805a1df6d73da888c631d7a6a58d3ff51df3940a4595fc","last_reissued_at":"2026-07-05T11:36:34.030516Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:36:34.030516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.02083","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:36:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yXp5U1skD+z/Bs90ODHrLHfAqCfiKDyF4Sub9L4ewHHSE6Su2PhoB4aDwzN47Rk0f0wr0fLXqb18cQgzQOhmBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T13:05:36.361717Z"},"content_sha256":"0acf146160459919612a9a498552f6a70e0a04ee3c8d8d03b4eb9a682b03afb6","schema_version":"1.0","event_id":"sha256:0acf146160459919612a9a498552f6a70e0a04ee3c8d8d03b4eb9a682b03afb6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TWAZLWRKMKI4P24ALIO7NVZ5VC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measuring Scientific Capabilities of Language Models with a Systems Biology Dry Lab","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Caitlin Fiona Harrigan, Chris J. Maddison, Haonan Duan, Jiarui Lu, Leonardo Cotta, Micha{\\l} Koziarski, Nishkrit Desai, Stephen Zhewen Lu","submitted_at":"2025-07-02T18:41:44Z","abstract_excerpt":"Designing experiments and result interpretations are core scientific competencies, particularly in biology, where researchers perturb complex systems to uncover the underlying systems. Recent efforts to evaluate the scientific capabilities of large language models (LLMs) fail to test these competencies because wet-lab experimentation is prohibitively expensive: in expertise, time and equipment. We introduce SciGym, a first-in-class benchmark that assesses LLMs' iterative experiment design and analysis abilities in open-ended scientific discovery tasks. SciGym overcomes the challenge of wet-lab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.02083","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/2507.02083/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":"2026-07-05T11:36:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PZCaWXTz0FJryCCNbmRrlEm61fV3Lbm/q8i7kgCytcoY3x7oF6JfsIM4jGWfFV8Zna35+BuxbYyi/Lg2R0T7Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T13:05:36.362403Z"},"content_sha256":"74ecd30d7c21cc150ff7849ff85ede3d9935e67d05b0a053ffe716f0e8571581","schema_version":"1.0","event_id":"sha256:74ecd30d7c21cc150ff7849ff85ede3d9935e67d05b0a053ffe716f0e8571581"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC/bundle.json","state_url":"https://pith.science/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC/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":"2026-07-12T13:05:36Z","links":{"resolver":"https://pith.science/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC","bundle":"https://pith.science/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC/bundle.json","state":"https://pith.science/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TWAZLWRKMKI4P24ALIO7NVZ5VC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TWAZLWRKMKI4P24ALIO7NVZ5VC","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":"2c6de74c5b65405e0f1d1fcf2b0e28c640725e39538018040791ca54cabedb96","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-02T18:41:44Z","title_canon_sha256":"feb0ef878a6c0109182dc6fbcd58b522538afbd099e888bd26c54b1e97933922"},"schema_version":"1.0","source":{"id":"2507.02083","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.02083","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"arxiv_version","alias_value":"2507.02083v2","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.02083","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"pith_short_12","alias_value":"TWAZLWRKMKI4","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"pith_short_16","alias_value":"TWAZLWRKMKI4P24A","created_at":"2026-07-05T11:36:34Z"},{"alias_kind":"pith_short_8","alias_value":"TWAZLWRK","created_at":"2026-07-05T11:36:34Z"}],"graph_snapshots":[{"event_id":"sha256:74ecd30d7c21cc150ff7849ff85ede3d9935e67d05b0a053ffe716f0e8571581","target":"graph","created_at":"2026-07-05T11:36:34Z","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/2507.02083/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Designing experiments and result interpretations are core scientific competencies, particularly in biology, where researchers perturb complex systems to uncover the underlying systems. Recent efforts to evaluate the scientific capabilities of large language models (LLMs) fail to test these competencies because wet-lab experimentation is prohibitively expensive: in expertise, time and equipment. We introduce SciGym, a first-in-class benchmark that assesses LLMs' iterative experiment design and analysis abilities in open-ended scientific discovery tasks. SciGym overcomes the challenge of wet-lab","authors_text":"Caitlin Fiona Harrigan, Chris J. Maddison, Haonan Duan, Jiarui Lu, Leonardo Cotta, Micha{\\l} Koziarski, Nishkrit Desai, Stephen Zhewen Lu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-02T18:41:44Z","title":"Measuring Scientific Capabilities of Language Models with a Systems Biology Dry Lab"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.02083","kind":"arxiv","version":2},"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":"sha256:0acf146160459919612a9a498552f6a70e0a04ee3c8d8d03b4eb9a682b03afb6","target":"record","created_at":"2026-07-05T11:36:34Z","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":"2c6de74c5b65405e0f1d1fcf2b0e28c640725e39538018040791ca54cabedb96","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-02T18:41:44Z","title_canon_sha256":"feb0ef878a6c0109182dc6fbcd58b522538afbd099e888bd26c54b1e97933922"},"schema_version":"1.0","source":{"id":"2507.02083","kind":"arxiv","version":2}},"canonical_sha256":"9d8195da2a6291c7eb805a1df6d73da888c631d7a6a58d3ff51df3940a4595fc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9d8195da2a6291c7eb805a1df6d73da888c631d7a6a58d3ff51df3940a4595fc","first_computed_at":"2026-07-05T11:36:34.030516Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:36:34.030516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"q56DL4XYGlwLuWjy9ppNAvTyI5oTaBM80qXc+4z2tOj3MoSZddLOXEw3vlDTo6wWubn0vCXOnE9Vqqh9or0HCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:36:34.031002Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.02083","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0acf146160459919612a9a498552f6a70e0a04ee3c8d8d03b4eb9a682b03afb6","sha256:74ecd30d7c21cc150ff7849ff85ede3d9935e67d05b0a053ffe716f0e8571581"],"state_sha256":"6051b6a6ddd96497ce99f38a96b20aa9a3a4332c6848e07d407b9ee9c31fbffc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+opsxs+2G8C/EOtEYOUPBIzMRm5uDMD+/dVqEdWF5HK744Nn3kA31mWWEbAOrWQCXNPp/TKzAbUc3QRef3QeCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T13:05:36.366922Z","bundle_sha256":"a029677e4b7132bb9283bd96250bf0c900d5eedcecb8444f3682b08a82f49078"}}