{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:GXA7JP2NIE23T6FO5RQQNG2UYC","short_pith_number":"pith:GXA7JP2N","schema_version":"1.0","canonical_sha256":"35c1f4bf4d4135b9f8aeec61069b54c09d48acbff9f6eb4ad34d91c86f3c0d52","source":{"kind":"arxiv","id":"2311.17280","version":4},"attestation_state":"computed","paper":{"title":"Does VLN Pretraining Work with Nonsensical or Irrelevant Instructions?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Ishika Singh, Jesse Thomason, Robin Jia, Wang Zhu, Yuan Huang","submitted_at":"2023-11-28T23:40:13Z","abstract_excerpt":"Data augmentation via back-translation is common when pretraining Vision-and-Language Navigation (VLN) models, even though the generated instructions are noisy. But: does that noise matter? We find that nonsensical or irrelevant language instructions during pretraining can have little effect on downstream performance for both HAMT and VLN-BERT on R2R, and is still better than only using clean, human data. To underscore these results, we concoct an efficient augmentation method, Unigram + Object, which generates nonsensical instructions that nonetheless improve downstream performance. Our findi"},"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":"2311.17280","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-28T23:40:13Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"d89dd355ed82eb536df7117863ed00f8a64d7dfa7786c9fad66790d674d3332b","abstract_canon_sha256":"fba1144a889d3e410925ded45b9e2672432cf3accf61f50e6008a13a57d493f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:27:27.583752Z","signature_b64":"dztOUSZIZKgt2rExmPAibsmvt4Pive29JmVu5ZEIDSnjfwSmkGhXl/Q4C+IQaEKcn/TUlJmSCs5D5gwKKlmPCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35c1f4bf4d4135b9f8aeec61069b54c09d48acbff9f6eb4ad34d91c86f3c0d52","last_reissued_at":"2026-07-05T07:27:27.583258Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:27:27.583258Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Does VLN Pretraining Work with Nonsensical or Irrelevant Instructions?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Ishika Singh, Jesse Thomason, Robin Jia, Wang Zhu, Yuan Huang","submitted_at":"2023-11-28T23:40:13Z","abstract_excerpt":"Data augmentation via back-translation is common when pretraining Vision-and-Language Navigation (VLN) models, even though the generated instructions are noisy. But: does that noise matter? We find that nonsensical or irrelevant language instructions during pretraining can have little effect on downstream performance for both HAMT and VLN-BERT on R2R, and is still better than only using clean, human data. To underscore these results, we concoct an efficient augmentation method, Unigram + Object, which generates nonsensical instructions that nonetheless improve downstream performance. Our findi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.17280","kind":"arxiv","version":4},"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/2311.17280/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":"2311.17280","created_at":"2026-07-05T07:27:27.583320+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.17280v4","created_at":"2026-07-05T07:27:27.583320+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.17280","created_at":"2026-07-05T07:27:27.583320+00:00"},{"alias_kind":"pith_short_12","alias_value":"GXA7JP2NIE23","created_at":"2026-07-05T07:27:27.583320+00:00"},{"alias_kind":"pith_short_16","alias_value":"GXA7JP2NIE23T6FO","created_at":"2026-07-05T07:27:27.583320+00:00"},{"alias_kind":"pith_short_8","alias_value":"GXA7JP2N","created_at":"2026-07-05T07:27:27.583320+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC","json":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC.json","graph_json":"https://pith.science/api/pith-number/GXA7JP2NIE23T6FO5RQQNG2UYC/graph.json","events_json":"https://pith.science/api/pith-number/GXA7JP2NIE23T6FO5RQQNG2UYC/events.json","paper":"https://pith.science/paper/GXA7JP2N"},"agent_actions":{"view_html":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC","download_json":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC.json","view_paper":"https://pith.science/paper/GXA7JP2N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.17280&json=true","fetch_graph":"https://pith.science/api/pith-number/GXA7JP2NIE23T6FO5RQQNG2UYC/graph.json","fetch_events":"https://pith.science/api/pith-number/GXA7JP2NIE23T6FO5RQQNG2UYC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC/action/storage_attestation","attest_author":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC/action/author_attestation","sign_citation":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC/action/citation_signature","submit_replication":"https://pith.science/pith/GXA7JP2NIE23T6FO5RQQNG2UYC/action/replication_record"}},"created_at":"2026-07-05T07:27:27.583320+00:00","updated_at":"2026-07-05T07:27:27.583320+00:00"}