{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:E5BYB6YGHUJBZPVTVQZW3XEVUD","short_pith_number":"pith:E5BYB6YG","canonical_record":{"source":{"id":"2406.06769","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2024-06-10T20:08:44Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"60a8343065fdcfcdace5850af0e6a7e5afa323ccbfa51bb9396899fa87375f95","abstract_canon_sha256":"2609544642f75f47f54760016d8b67bdb8146a72fac7aeb104a73a8adf8e3369"},"schema_version":"1.0"},"canonical_sha256":"274380fb063d121cbeb3ac336ddc95a0cace5d30c29f63383a2046e3705f6caf","source":{"kind":"arxiv","id":"2406.06769","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.06769","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"arxiv_version","alias_value":"2406.06769v2","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06769","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"pith_short_12","alias_value":"E5BYB6YGHUJB","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"pith_short_16","alias_value":"E5BYB6YGHUJBZPVT","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"pith_short_8","alias_value":"E5BYB6YG","created_at":"2026-07-05T09:17:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:E5BYB6YGHUJBZPVTVQZW3XEVUD","target":"record","payload":{"canonical_record":{"source":{"id":"2406.06769","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2024-06-10T20:08:44Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"60a8343065fdcfcdace5850af0e6a7e5afa323ccbfa51bb9396899fa87375f95","abstract_canon_sha256":"2609544642f75f47f54760016d8b67bdb8146a72fac7aeb104a73a8adf8e3369"},"schema_version":"1.0"},"canonical_sha256":"274380fb063d121cbeb3ac336ddc95a0cace5d30c29f63383a2046e3705f6caf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:17:13.709683Z","signature_b64":"Hjrcqcx091IeaJWxhUiLB1WvrZc3LQUVExvPKNcsgnd6MLZZG8wtXDY4rsDp4q5gLivPTytTKTSLcVThYX9wDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"274380fb063d121cbeb3ac336ddc95a0cace5d30c29f63383a2046e3705f6caf","last_reissued_at":"2026-07-05T09:17:13.709196Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:17:13.709196Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.06769","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":"2026-07-05T09:17:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"88jsDcVA1uZyG/+I4N3MRv5eWZfSzwMsHJLI1ifsB2hEOOBAP1nwvbVlClu4pQw7tLlMK9BDVnHlFniIvh6fCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:46:56.358413Z"},"content_sha256":"d14ddd68862405d0ac4347fa0cebc9e9b847ed74d2eac6acfc8ba3345f0da1fe","schema_version":"1.0","event_id":"sha256:d14ddd68862405d0ac4347fa0cebc9e9b847ed74d2eac6acfc8ba3345f0da1fe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:E5BYB6YGHUJBZPVTVQZW3XEVUD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Bhavana Dalvi Mishra, Bodhisattwa Prasad Majumder, Erin Bransom, Marc-Alexandre C\\^ot\\'e, Oyvind Tafjord, Peter Clark, Peter Jansen, Tushar Khot","submitted_at":"2024-06-10T20:08:44Z","abstract_excerpt":"Automated scientific discovery promises to accelerate progress across scientific domains. However, developing and evaluating an AI agent's capacity for end-to-end scientific reasoning is challenging as running real-world experiments is often prohibitively expensive or infeasible. In this work we introduce DISCOVERYWORLD, the first virtual environment for developing and benchmarking an agent's ability to perform complete cycles of novel scientific discovery. DISCOVERYWORLD contains a variety of different challenges, covering topics as diverse as radioisotope dating, rocket science, and proteomi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06769","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/2406.06769/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":"2026-07-05T09:17:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ez8HkYT/G16oAYyQ13zL8ST/w7I98Rh5AL/tN1gc33I0HpCMVkco5a7+x3y+NJLSxTJBdrP2/MgnNUVJOZAeAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:46:56.358920Z"},"content_sha256":"efe9a8398084c2a26eab1088361c0dc64788c1631ed5c6a47972397e4f43c228","schema_version":"1.0","event_id":"sha256:efe9a8398084c2a26eab1088361c0dc64788c1631ed5c6a47972397e4f43c228"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD/bundle.json","state_url":"https://pith.science/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD/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":"2026-08-03T19:46:56Z","links":{"resolver":"https://pith.science/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD","bundle":"https://pith.science/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD/bundle.json","state":"https://pith.science/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/E5BYB6YGHUJBZPVTVQZW3XEVUD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:E5BYB6YGHUJBZPVTVQZW3XEVUD","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":"2609544642f75f47f54760016d8b67bdb8146a72fac7aeb104a73a8adf8e3369","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2024-06-10T20:08:44Z","title_canon_sha256":"60a8343065fdcfcdace5850af0e6a7e5afa323ccbfa51bb9396899fa87375f95"},"schema_version":"1.0","source":{"id":"2406.06769","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.06769","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"arxiv_version","alias_value":"2406.06769v2","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.06769","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"pith_short_12","alias_value":"E5BYB6YGHUJB","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"pith_short_16","alias_value":"E5BYB6YGHUJBZPVT","created_at":"2026-07-05T09:17:13Z"},{"alias_kind":"pith_short_8","alias_value":"E5BYB6YG","created_at":"2026-07-05T09:17:13Z"}],"graph_snapshots":[{"event_id":"sha256:efe9a8398084c2a26eab1088361c0dc64788c1631ed5c6a47972397e4f43c228","target":"graph","created_at":"2026-07-05T09:17:13Z","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/2406.06769/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Automated scientific discovery promises to accelerate progress across scientific domains. However, developing and evaluating an AI agent's capacity for end-to-end scientific reasoning is challenging as running real-world experiments is often prohibitively expensive or infeasible. In this work we introduce DISCOVERYWORLD, the first virtual environment for developing and benchmarking an agent's ability to perform complete cycles of novel scientific discovery. DISCOVERYWORLD contains a variety of different challenges, covering topics as diverse as radioisotope dating, rocket science, and proteomi","authors_text":"Bhavana Dalvi Mishra, Bodhisattwa Prasad Majumder, Erin Bransom, Marc-Alexandre C\\^ot\\'e, Oyvind Tafjord, Peter Clark, Peter Jansen, Tushar Khot","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2024-06-10T20:08:44Z","title":"DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.06769","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:d14ddd68862405d0ac4347fa0cebc9e9b847ed74d2eac6acfc8ba3345f0da1fe","target":"record","created_at":"2026-07-05T09:17:13Z","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":"2609544642f75f47f54760016d8b67bdb8146a72fac7aeb104a73a8adf8e3369","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2024-06-10T20:08:44Z","title_canon_sha256":"60a8343065fdcfcdace5850af0e6a7e5afa323ccbfa51bb9396899fa87375f95"},"schema_version":"1.0","source":{"id":"2406.06769","kind":"arxiv","version":2}},"canonical_sha256":"274380fb063d121cbeb3ac336ddc95a0cace5d30c29f63383a2046e3705f6caf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"274380fb063d121cbeb3ac336ddc95a0cace5d30c29f63383a2046e3705f6caf","first_computed_at":"2026-07-05T09:17:13.709196Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:17:13.709196Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Hjrcqcx091IeaJWxhUiLB1WvrZc3LQUVExvPKNcsgnd6MLZZG8wtXDY4rsDp4q5gLivPTytTKTSLcVThYX9wDg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:17:13.709683Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.06769","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d14ddd68862405d0ac4347fa0cebc9e9b847ed74d2eac6acfc8ba3345f0da1fe","sha256:efe9a8398084c2a26eab1088361c0dc64788c1631ed5c6a47972397e4f43c228"],"state_sha256":"445790ef478c560268c37ab577ac694db497db3930a31470fd2d0fe67283c732"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SqLtjjVfYzy1UC3zu6n+V1gbA1RVi6+h+z7JjV7N0lgfpBUQNR5QkWiBZEIkAZFIMREWYSxm2qPXsnjNQD+hBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T19:46:56.369824Z","bundle_sha256":"ca3ac51585ea3ffb48ed9a17a07af353d50adccf9426bc04afb525c191dc0473"}}