{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:GS467ITWDDXN36K4C6BXWWL4TW","short_pith_number":"pith:GS467ITW","canonical_record":{"source":{"id":"2203.07540","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-14T22:52:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e23af4b8e860e7886ee50d0757a09b68cdbdc7cbc6f20f79cce6163bf560af72","abstract_canon_sha256":"b7baf3b8a31069f667583a36b366e44104ada49c87bf128d875df04018f4edc8"},"schema_version":"1.0"},"canonical_sha256":"34b9efa27618eeddf95c17837b597c9daa85d39948d9447e13c7626ed1dab163","source":{"kind":"arxiv","id":"2203.07540","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.07540","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"arxiv_version","alias_value":"2203.07540v2","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.07540","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"pith_short_12","alias_value":"GS467ITWDDXN","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"pith_short_16","alias_value":"GS467ITWDDXN36K4","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"pith_short_8","alias_value":"GS467ITW","created_at":"2026-07-05T05:15:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:GS467ITWDDXN36K4C6BXWWL4TW","target":"record","payload":{"canonical_record":{"source":{"id":"2203.07540","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-14T22:52:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e23af4b8e860e7886ee50d0757a09b68cdbdc7cbc6f20f79cce6163bf560af72","abstract_canon_sha256":"b7baf3b8a31069f667583a36b366e44104ada49c87bf128d875df04018f4edc8"},"schema_version":"1.0"},"canonical_sha256":"34b9efa27618eeddf95c17837b597c9daa85d39948d9447e13c7626ed1dab163","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:15:37.763461Z","signature_b64":"01AfjVRfharo4uU6SyfGiLKeCsZmJYzN1aAomVNbjGTOixKEDBxlRZ3xLqpNxroQ3ygG60I3ZLn9wrmbHQdgAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"34b9efa27618eeddf95c17837b597c9daa85d39948d9447e13c7626ed1dab163","last_reissued_at":"2026-07-05T05:15:37.762948Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:15:37.762948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2203.07540","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":"2026-07-05T05:15:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VKGhQVP9bZoB/kW7ZOvc7zz1jKIfFhfdVoPugUXTdl717GE4JfWe4jeMrXChOptVlZkM9VJkIN+4dTUHq/sKAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T01:22:02.801307Z"},"content_sha256":"6be3b5dadcefb846d10c6448ae125a9d51574174161eb5347fc439ac6568971e","schema_version":"1.0","event_id":"sha256:6be3b5dadcefb846d10c6448ae125a9d51574174161eb5347fc439ac6568971e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:GS467ITWDDXN36K4C6BXWWL4TW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ScienceWorld: Is your Agent Smarter than a 5th Grader?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Marc-Alexandre C\\^ot\\'e, Peter Jansen, Prithviraj Ammanabrolu, Ruoyao Wang","submitted_at":"2022-03-14T22:52:34Z","abstract_excerpt":"We present ScienceWorld, a benchmark to test agents' scientific reasoning abilities in a new interactive text environment at the level of a standard elementary school science curriculum. Despite the transformer-based progress seen in question-answering and scientific text processing, we find that current models cannot reason about or explain learned science concepts in novel contexts. For instance, models can easily answer what the conductivity of a known material is but struggle when asked how they would conduct an experiment in a grounded environment to find the conductivity of an unknown ma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.07540","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/2203.07540/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":"2026-07-05T05:15:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FH1gJ6FKXIaSXq1GnphCfSbi0EtuG0u9djXkGAkt4FsEQ6FcXlpl4LhZgBuVgXuToMRMaA61+FzBKAdumJLPAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T01:22:02.802188Z"},"content_sha256":"862780385aa69a96efb2392f1670c874e3fbe42d353c466cfb3a1780b1a605ea","schema_version":"1.0","event_id":"sha256:862780385aa69a96efb2392f1670c874e3fbe42d353c466cfb3a1780b1a605ea"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GS467ITWDDXN36K4C6BXWWL4TW/bundle.json","state_url":"https://pith.science/pith/GS467ITWDDXN36K4C6BXWWL4TW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GS467ITWDDXN36K4C6BXWWL4TW/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":"2026-08-13T01:22:02Z","links":{"resolver":"https://pith.science/pith/GS467ITWDDXN36K4C6BXWWL4TW","bundle":"https://pith.science/pith/GS467ITWDDXN36K4C6BXWWL4TW/bundle.json","state":"https://pith.science/pith/GS467ITWDDXN36K4C6BXWWL4TW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GS467ITWDDXN36K4C6BXWWL4TW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:GS467ITWDDXN36K4C6BXWWL4TW","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":"b7baf3b8a31069f667583a36b366e44104ada49c87bf128d875df04018f4edc8","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-14T22:52:34Z","title_canon_sha256":"e23af4b8e860e7886ee50d0757a09b68cdbdc7cbc6f20f79cce6163bf560af72"},"schema_version":"1.0","source":{"id":"2203.07540","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.07540","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"arxiv_version","alias_value":"2203.07540v2","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.07540","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"pith_short_12","alias_value":"GS467ITWDDXN","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"pith_short_16","alias_value":"GS467ITWDDXN36K4","created_at":"2026-07-05T05:15:37Z"},{"alias_kind":"pith_short_8","alias_value":"GS467ITW","created_at":"2026-07-05T05:15:37Z"}],"graph_snapshots":[{"event_id":"sha256:862780385aa69a96efb2392f1670c874e3fbe42d353c466cfb3a1780b1a605ea","target":"graph","created_at":"2026-07-05T05:15:37Z","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/2203.07540/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present ScienceWorld, a benchmark to test agents' scientific reasoning abilities in a new interactive text environment at the level of a standard elementary school science curriculum. Despite the transformer-based progress seen in question-answering and scientific text processing, we find that current models cannot reason about or explain learned science concepts in novel contexts. For instance, models can easily answer what the conductivity of a known material is but struggle when asked how they would conduct an experiment in a grounded environment to find the conductivity of an unknown ma","authors_text":"Marc-Alexandre C\\^ot\\'e, Peter Jansen, Prithviraj Ammanabrolu, Ruoyao Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-14T22:52:34Z","title":"ScienceWorld: Is your Agent Smarter than a 5th Grader?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.07540","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":"sha256:6be3b5dadcefb846d10c6448ae125a9d51574174161eb5347fc439ac6568971e","target":"record","created_at":"2026-07-05T05:15:37Z","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":"b7baf3b8a31069f667583a36b366e44104ada49c87bf128d875df04018f4edc8","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-14T22:52:34Z","title_canon_sha256":"e23af4b8e860e7886ee50d0757a09b68cdbdc7cbc6f20f79cce6163bf560af72"},"schema_version":"1.0","source":{"id":"2203.07540","kind":"arxiv","version":2}},"canonical_sha256":"34b9efa27618eeddf95c17837b597c9daa85d39948d9447e13c7626ed1dab163","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"34b9efa27618eeddf95c17837b597c9daa85d39948d9447e13c7626ed1dab163","first_computed_at":"2026-07-05T05:15:37.762948Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:15:37.762948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"01AfjVRfharo4uU6SyfGiLKeCsZmJYzN1aAomVNbjGTOixKEDBxlRZ3xLqpNxroQ3ygG60I3ZLn9wrmbHQdgAg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:15:37.763461Z","signed_message":"canonical_sha256_bytes"},"source_id":"2203.07540","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6be3b5dadcefb846d10c6448ae125a9d51574174161eb5347fc439ac6568971e","sha256:862780385aa69a96efb2392f1670c874e3fbe42d353c466cfb3a1780b1a605ea"],"state_sha256":"e95155374da1e8d50a29a7d05e8fd8e4a57d883d9727c1f1b6068a048f0a585c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Td9W89p9DK0SpvzlgkWi9hjF2qB883gX4Zzh5pjLgSm9Z0dDqOuWsJeDptG/KwSEs67skIlmrmJidxCyj253DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T01:22:02.809388Z","bundle_sha256":"a002667a63e7b83ff082abb5d8b1636b7e751539bc25e52260aabb3a1b21e22d"}}