{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:WPJVUJEGMD26U3JZC7FUN3IWZR","short_pith_number":"pith:WPJVUJEG","schema_version":"1.0","canonical_sha256":"b3d35a248660f5ea6d3917cb46ed16cc6ebbd8d63bf667be33a137347af049b1","source":{"kind":"arxiv","id":"2111.14034","version":1},"attestation_state":"computed","paper":{"title":"ORCHARD: A Benchmark For Measuring Systematic Generalization of Multi-Hierarchical Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alvin Chan, Bill Tuck Weng Pung","submitted_at":"2021-11-28T03:11:37Z","abstract_excerpt":"The ability to reason with multiple hierarchical structures is an attractive and desirable property of sequential inductive biases for natural language processing. Do the state-of-the-art Transformers and LSTM architectures implicitly encode for these biases? To answer this, we propose ORCHARD, a diagnostic dataset for systematically evaluating hierarchical reasoning in state-of-the-art neural sequence models. While there have been prior evaluation frameworks such as ListOps or Logical Inference, our work presents a novel and more natural setting where our models learn to reason with multiple "},"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":"2111.14034","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-11-28T03:11:37Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9370cd2929bcee9cb7cd5adfd7bbdde4c195d2b1227dbb9819a8a8a71da5662a","abstract_canon_sha256":"cfa191c9459dab1db390fca84ed8a9b24023c61643123626f0d456dcb7230166"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:35:38.465995Z","signature_b64":"cc7OchDNKIjZ1eMznMqJbB78mvtnyKpViWnPPupWWN8bpPO76SIyktE/lGiLkS5PN2CyhyPgpVfaE/G+247gDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b3d35a248660f5ea6d3917cb46ed16cc6ebbd8d63bf667be33a137347af049b1","last_reissued_at":"2026-07-05T03:35:38.465558Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:35:38.465558Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ORCHARD: A Benchmark For Measuring Systematic Generalization of Multi-Hierarchical Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alvin Chan, Bill Tuck Weng Pung","submitted_at":"2021-11-28T03:11:37Z","abstract_excerpt":"The ability to reason with multiple hierarchical structures is an attractive and desirable property of sequential inductive biases for natural language processing. Do the state-of-the-art Transformers and LSTM architectures implicitly encode for these biases? To answer this, we propose ORCHARD, a diagnostic dataset for systematically evaluating hierarchical reasoning in state-of-the-art neural sequence models. While there have been prior evaluation frameworks such as ListOps or Logical Inference, our work presents a novel and more natural setting where our models learn to reason with multiple "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.14034","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/2111.14034/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":"2111.14034","created_at":"2026-07-05T03:35:38.465617+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.14034v1","created_at":"2026-07-05T03:35:38.465617+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.14034","created_at":"2026-07-05T03:35:38.465617+00:00"},{"alias_kind":"pith_short_12","alias_value":"WPJVUJEGMD26","created_at":"2026-07-05T03:35:38.465617+00:00"},{"alias_kind":"pith_short_16","alias_value":"WPJVUJEGMD26U3JZ","created_at":"2026-07-05T03:35:38.465617+00:00"},{"alias_kind":"pith_short_8","alias_value":"WPJVUJEG","created_at":"2026-07-05T03:35:38.465617+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.00847","citing_title":"H-Probes: Extracting Hierarchical Structures From Latent Representations of Language Models","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR","json":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR.json","graph_json":"https://pith.science/api/pith-number/WPJVUJEGMD26U3JZC7FUN3IWZR/graph.json","events_json":"https://pith.science/api/pith-number/WPJVUJEGMD26U3JZC7FUN3IWZR/events.json","paper":"https://pith.science/paper/WPJVUJEG"},"agent_actions":{"view_html":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR","download_json":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR.json","view_paper":"https://pith.science/paper/WPJVUJEG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.14034&json=true","fetch_graph":"https://pith.science/api/pith-number/WPJVUJEGMD26U3JZC7FUN3IWZR/graph.json","fetch_events":"https://pith.science/api/pith-number/WPJVUJEGMD26U3JZC7FUN3IWZR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR/action/storage_attestation","attest_author":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR/action/author_attestation","sign_citation":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR/action/citation_signature","submit_replication":"https://pith.science/pith/WPJVUJEGMD26U3JZC7FUN3IWZR/action/replication_record"}},"created_at":"2026-07-05T03:35:38.465617+00:00","updated_at":"2026-07-05T03:35:38.465617+00:00"}