{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:BXJCIXOWCCMLZK6UL5P4WZJBEX","short_pith_number":"pith:BXJCIXOW","schema_version":"1.0","canonical_sha256":"0dd2245dd61098bcabd45f5fcb652125c7673863b1af6203ec2723778e24112b","source":{"kind":"arxiv","id":"2201.11766","version":2},"attestation_state":"computed","paper":{"title":"Recursive Decoding: A Situated Cognition Approach to Compositional Generation in Grounded Language Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Lauren Phillips, Matthew Setzler, Scott Howland","submitted_at":"2022-01-27T19:13:42Z","abstract_excerpt":"Compositional generalization is a troubling blind spot for neural language models. Recent efforts have presented techniques for improving a model's ability to encode novel combinations of known inputs, but less work has focused on generating novel combinations of known outputs. Here we focus on this latter \"decode-side\" form of generalization in the context of gSCAN, a synthetic benchmark for compositional generalization in grounded language understanding. We present Recursive Decoding (RD), a novel procedure for training and using seq2seq models, targeted towards decode-side generalization. R"},"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":"2201.11766","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-01-27T19:13:42Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a08920711b95af6503fae0f9ab8d80cff97f963e503d9c4a1941b7bd9f0f51d3","abstract_canon_sha256":"81d0ca90a29f76b72e7321ad8b8a2b7b94d82b8c8b36f5870d869d8b61c4de6a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:58:06.193332Z","signature_b64":"/ithLWUziIhqzHrGQe0YNjciIi5mfeTEgA/EjPFMFHn3BIoptJfTzSIvpofNQrRrlWm7Tf1baPgehPesDyDADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0dd2245dd61098bcabd45f5fcb652125c7673863b1af6203ec2723778e24112b","last_reissued_at":"2026-07-05T03:58:06.192857Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:58:06.192857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Recursive Decoding: A Situated Cognition Approach to Compositional Generation in Grounded Language Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Lauren Phillips, Matthew Setzler, Scott Howland","submitted_at":"2022-01-27T19:13:42Z","abstract_excerpt":"Compositional generalization is a troubling blind spot for neural language models. Recent efforts have presented techniques for improving a model's ability to encode novel combinations of known inputs, but less work has focused on generating novel combinations of known outputs. Here we focus on this latter \"decode-side\" form of generalization in the context of gSCAN, a synthetic benchmark for compositional generalization in grounded language understanding. We present Recursive Decoding (RD), a novel procedure for training and using seq2seq models, targeted towards decode-side generalization. R"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.11766","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/2201.11766/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":"2201.11766","created_at":"2026-07-05T03:58:06.192915+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.11766v2","created_at":"2026-07-05T03:58:06.192915+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.11766","created_at":"2026-07-05T03:58:06.192915+00:00"},{"alias_kind":"pith_short_12","alias_value":"BXJCIXOWCCML","created_at":"2026-07-05T03:58:06.192915+00:00"},{"alias_kind":"pith_short_16","alias_value":"BXJCIXOWCCMLZK6U","created_at":"2026-07-05T03:58:06.192915+00:00"},{"alias_kind":"pith_short_8","alias_value":"BXJCIXOW","created_at":"2026-07-05T03:58:06.192915+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.19372","citing_title":"Learning neuro-symbolic convergent term rewriting systems","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX","json":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX.json","graph_json":"https://pith.science/api/pith-number/BXJCIXOWCCMLZK6UL5P4WZJBEX/graph.json","events_json":"https://pith.science/api/pith-number/BXJCIXOWCCMLZK6UL5P4WZJBEX/events.json","paper":"https://pith.science/paper/BXJCIXOW"},"agent_actions":{"view_html":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX","download_json":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX.json","view_paper":"https://pith.science/paper/BXJCIXOW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.11766&json=true","fetch_graph":"https://pith.science/api/pith-number/BXJCIXOWCCMLZK6UL5P4WZJBEX/graph.json","fetch_events":"https://pith.science/api/pith-number/BXJCIXOWCCMLZK6UL5P4WZJBEX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX/action/storage_attestation","attest_author":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX/action/author_attestation","sign_citation":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX/action/citation_signature","submit_replication":"https://pith.science/pith/BXJCIXOWCCMLZK6UL5P4WZJBEX/action/replication_record"}},"created_at":"2026-07-05T03:58:06.192915+00:00","updated_at":"2026-07-05T03:58:06.192915+00:00"}