{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:CA2TI6JP6YLKMD2CRVKQZY2QOO","short_pith_number":"pith:CA2TI6JP","canonical_record":{"source":{"id":"2302.06729","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-13T22:34:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b660155e92618028b44de974e4a95ebb974426040fa5fafad0ddf04de60d60c6","abstract_canon_sha256":"da5f66cbbd02716ff0e6165c53932c0444d29aab8fcc53765ec44add7d5188f3"},"schema_version":"1.0"},"canonical_sha256":"103534792ff616a60f428d550ce35073a586cc26bea42b0900a78661873280de","source":{"kind":"arxiv","id":"2302.06729","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.06729","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"arxiv_version","alias_value":"2302.06729v1","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.06729","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"pith_short_12","alias_value":"CA2TI6JP6YLK","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"pith_short_16","alias_value":"CA2TI6JP6YLKMD2C","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"pith_short_8","alias_value":"CA2TI6JP","created_at":"2026-07-05T05:41:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:CA2TI6JP6YLKMD2CRVKQZY2QOO","target":"record","payload":{"canonical_record":{"source":{"id":"2302.06729","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-13T22:34:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b660155e92618028b44de974e4a95ebb974426040fa5fafad0ddf04de60d60c6","abstract_canon_sha256":"da5f66cbbd02716ff0e6165c53932c0444d29aab8fcc53765ec44add7d5188f3"},"schema_version":"1.0"},"canonical_sha256":"103534792ff616a60f428d550ce35073a586cc26bea42b0900a78661873280de","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:41:43.316957Z","signature_b64":"HozXkfIb7QvvfeqndQ7xLEHCd4pVMmkcEi8DCKSeulsM7zcawJMqguD2RgZduDAml3ExYlhAOtcftooOO8JxCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"103534792ff616a60f428d550ce35073a586cc26bea42b0900a78661873280de","last_reissued_at":"2026-07-05T05:41:43.316434Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:41:43.316434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2302.06729","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":"2026-07-05T05:41:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cYRCfyEOtus9yehRugSr2PcACFwxqxndos1l0DEQz02oGQTNLyIFe6u/gT40JB3fI98jvUKFzTZQ5MzCwXLGCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T21:48:31.359770Z"},"content_sha256":"6afc3a9f30deee1f648362bb05d06e593073128674d3a666ad4b1ca84f51d31d","schema_version":"1.0","event_id":"sha256:6afc3a9f30deee1f648362bb05d06e593073128674d3a666ad4b1ca84f51d31d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:CA2TI6JP6YLKMD2CRVKQZY2QOO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"STREET: A Multi-Task Structured Reasoning and Explanation Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anjelica Ramos, Bing Xiang, Danilo Ribeiro, Dan Roth, Deguang Kong, George Karypis, Henry Zhu, Juliette Burger, Rui Dong, Shen Wang, William Wang, Xiaofei Ma, Zhiheng Huang","submitted_at":"2023-02-13T22:34:02Z","abstract_excerpt":"We introduce STREET, a unified multi-task and multi-domain natural language reasoning and explanation benchmark. Unlike most existing question-answering (QA) datasets, we expect models to not only answer questions, but also produce step-by-step structured explanations describing how premises in the question are used to produce intermediate conclusions that can prove the correctness of a certain answer. We perform extensive evaluation with popular language models such as few-shot prompting GPT-3 and fine-tuned T5. We find that these models still lag behind human performance when producing such "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.06729","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/2302.06729/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":"2026-07-05T05:41:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"px3w4R2MY05LmaaYezE8dk/GJInahb3H9PdTkVXMrgr6GhHAIyuINysvl0T61sXAtxbcB9ubsPwaMicZWrlcAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T21:48:31.360776Z"},"content_sha256":"1cd10e80bf567eaa309691e67d4cb891606a4e980b06a659aa9d005d554f54a8","schema_version":"1.0","event_id":"sha256:1cd10e80bf567eaa309691e67d4cb891606a4e980b06a659aa9d005d554f54a8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO/bundle.json","state_url":"https://pith.science/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO/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":"2026-08-16T21:48:31Z","links":{"resolver":"https://pith.science/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO","bundle":"https://pith.science/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO/bundle.json","state":"https://pith.science/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CA2TI6JP6YLKMD2CRVKQZY2QOO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:CA2TI6JP6YLKMD2CRVKQZY2QOO","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":"da5f66cbbd02716ff0e6165c53932c0444d29aab8fcc53765ec44add7d5188f3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-13T22:34:02Z","title_canon_sha256":"b660155e92618028b44de974e4a95ebb974426040fa5fafad0ddf04de60d60c6"},"schema_version":"1.0","source":{"id":"2302.06729","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.06729","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"arxiv_version","alias_value":"2302.06729v1","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.06729","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"pith_short_12","alias_value":"CA2TI6JP6YLK","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"pith_short_16","alias_value":"CA2TI6JP6YLKMD2C","created_at":"2026-07-05T05:41:43Z"},{"alias_kind":"pith_short_8","alias_value":"CA2TI6JP","created_at":"2026-07-05T05:41:43Z"}],"graph_snapshots":[{"event_id":"sha256:1cd10e80bf567eaa309691e67d4cb891606a4e980b06a659aa9d005d554f54a8","target":"graph","created_at":"2026-07-05T05:41:43Z","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/2302.06729/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce STREET, a unified multi-task and multi-domain natural language reasoning and explanation benchmark. Unlike most existing question-answering (QA) datasets, we expect models to not only answer questions, but also produce step-by-step structured explanations describing how premises in the question are used to produce intermediate conclusions that can prove the correctness of a certain answer. We perform extensive evaluation with popular language models such as few-shot prompting GPT-3 and fine-tuned T5. We find that these models still lag behind human performance when producing such ","authors_text":"Anjelica Ramos, Bing Xiang, Danilo Ribeiro, Dan Roth, Deguang Kong, George Karypis, Henry Zhu, Juliette Burger, Rui Dong, Shen Wang, William Wang, Xiaofei Ma, Zhiheng Huang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-13T22:34:02Z","title":"STREET: A Multi-Task Structured Reasoning and Explanation Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.06729","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":"sha256:6afc3a9f30deee1f648362bb05d06e593073128674d3a666ad4b1ca84f51d31d","target":"record","created_at":"2026-07-05T05:41:43Z","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":"da5f66cbbd02716ff0e6165c53932c0444d29aab8fcc53765ec44add7d5188f3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-13T22:34:02Z","title_canon_sha256":"b660155e92618028b44de974e4a95ebb974426040fa5fafad0ddf04de60d60c6"},"schema_version":"1.0","source":{"id":"2302.06729","kind":"arxiv","version":1}},"canonical_sha256":"103534792ff616a60f428d550ce35073a586cc26bea42b0900a78661873280de","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"103534792ff616a60f428d550ce35073a586cc26bea42b0900a78661873280de","first_computed_at":"2026-07-05T05:41:43.316434Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:41:43.316434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HozXkfIb7QvvfeqndQ7xLEHCd4pVMmkcEi8DCKSeulsM7zcawJMqguD2RgZduDAml3ExYlhAOtcftooOO8JxCA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:41:43.316957Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.06729","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6afc3a9f30deee1f648362bb05d06e593073128674d3a666ad4b1ca84f51d31d","sha256:1cd10e80bf567eaa309691e67d4cb891606a4e980b06a659aa9d005d554f54a8"],"state_sha256":"849a0ca79d9a7c918da6dc5113ba4b33674ec30ac9b154ca488c3f4c644c3602"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9mCUerbdco8RuzOCuIDrUbvRT+c1CzHJQ91XDTid4eQxGYAaRUYLANFUs1AzEm35B7fLu+qQ0tW8ONFZ7dJsAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T21:48:31.366088Z","bundle_sha256":"a315365706667fcba50f554fc16124447bf1027395da095289b1010fa48fa5d6"}}