{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FVSU5YOL6Y343LXHHDEBK4B5QV","short_pith_number":"pith:FVSU5YOL","schema_version":"1.0","canonical_sha256":"2d654ee1cbf637cdaee738c815703d85407d55bcec74bb7250d55ecae7aefc50","source":{"kind":"arxiv","id":"2305.14196","version":3},"attestation_state":"computed","paper":{"title":"ZeroSCROLLS: A Zero-Shot Benchmark for Long Text Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Avia Efrat, Jonathan Berant, Maor Ivgi, Omer Levy, Uri Shaham","submitted_at":"2023-05-23T16:15:31Z","abstract_excerpt":"We introduce ZeroSCROLLS, a zero-shot benchmark for natural language understanding over long texts, which contains only test and small validation sets, without training data. We adapt six tasks from the SCROLLS benchmark, and add four new datasets, including two novel information fusing tasks, such as aggregating the percentage of positive reviews. Using ZeroSCROLLS, we conduct a comprehensive evaluation of both open-source and closed large language models, finding that Claude outperforms ChatGPT, and that GPT-4 achieves the highest average score. However, there is still room for improvement o"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2305.14196","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T16:15:31Z","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"title_canon_sha256":"a878cb18bbcbd9880f4b165cc76c09e88659b249762f7c9efcf6e4a459b684fd","abstract_canon_sha256":"97f33acbdde1ab3b3e7097cc3a6e9d14c7b1002a6d06c4e7ad79afcbf7685a4c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:24:50.281954Z","signature_b64":"Hq4l/0530VMruUthkj5WPvkgNRDKUJRpsyGIXHe4vTcIOY5SbPQaKJjWRkrCBeGQ38P4zaTb8YGzkjvhVx5pCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d654ee1cbf637cdaee738c815703d85407d55bcec74bb7250d55ecae7aefc50","last_reissued_at":"2026-07-05T07:24:50.281492Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:24:50.281492Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ZeroSCROLLS: A Zero-Shot Benchmark for Long Text Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Avia Efrat, Jonathan Berant, Maor Ivgi, Omer Levy, Uri Shaham","submitted_at":"2023-05-23T16:15:31Z","abstract_excerpt":"We introduce ZeroSCROLLS, a zero-shot benchmark for natural language understanding over long texts, which contains only test and small validation sets, without training data. We adapt six tasks from the SCROLLS benchmark, and add four new datasets, including two novel information fusing tasks, such as aggregating the percentage of positive reviews. Using ZeroSCROLLS, we conduct a comprehensive evaluation of both open-source and closed large language models, finding that Claude outperforms ChatGPT, and that GPT-4 achieves the highest average score. However, there is still room for improvement o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14196","kind":"arxiv","version":3},"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/2305.14196/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2305.14196","created_at":"2026-07-05T07:24:50.281551+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14196v3","created_at":"2026-07-05T07:24:50.281551+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14196","created_at":"2026-07-05T07:24:50.281551+00:00"},{"alias_kind":"pith_short_12","alias_value":"FVSU5YOL6Y34","created_at":"2026-07-05T07:24:50.281551+00:00"},{"alias_kind":"pith_short_16","alias_value":"FVSU5YOL6Y343LXH","created_at":"2026-07-05T07:24:50.281551+00:00"},{"alias_kind":"pith_short_8","alias_value":"FVSU5YOL","created_at":"2026-07-05T07:24:50.281551+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01101","citing_title":"Soft-NBCE: Entropy-Weighted Chunk Fusion for Long-Context","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27705","citing_title":"Mitigating Position Bias in Transformers via Layer-Specific Positional Embedding Scaling","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23170","citing_title":"Positional Failures in Long-Context LLMs: A Blind Spot in Reasoning Benchmarks","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2506.04565","citing_title":"From Standalone LLMs to Integrated Intelligence: A Survey of Compound Al Systems","ref_index":153,"is_internal_anchor":false},{"citing_arxiv_id":"2404.13501","citing_title":"A Survey on the Memory Mechanism of Large Language Model based Agents","ref_index":138,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06464","citing_title":"Gated Delta Networks: Improving Mamba2 with Delta Rule","ref_index":166,"is_internal_anchor":false},{"citing_arxiv_id":"2308.14508","citing_title":"LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding","ref_index":114,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV","json":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV.json","graph_json":"https://pith.science/api/pith-number/FVSU5YOL6Y343LXHHDEBK4B5QV/graph.json","events_json":"https://pith.science/api/pith-number/FVSU5YOL6Y343LXHHDEBK4B5QV/events.json","paper":"https://pith.science/paper/FVSU5YOL"},"agent_actions":{"view_html":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV","download_json":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV.json","view_paper":"https://pith.science/paper/FVSU5YOL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14196&json=true","fetch_graph":"https://pith.science/api/pith-number/FVSU5YOL6Y343LXHHDEBK4B5QV/graph.json","fetch_events":"https://pith.science/api/pith-number/FVSU5YOL6Y343LXHHDEBK4B5QV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV/action/storage_attestation","attest_author":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV/action/author_attestation","sign_citation":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV/action/citation_signature","submit_replication":"https://pith.science/pith/FVSU5YOL6Y343LXHHDEBK4B5QV/action/replication_record"}},"created_at":"2026-07-05T07:24:50.281551+00:00","updated_at":"2026-07-05T07:24:50.281551+00:00"}