{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:B2AVNQO5X6BX2QLWNAE2CY2JQH","short_pith_number":"pith:B2AVNQO5","schema_version":"1.0","canonical_sha256":"0e8156c1ddbf837d41766809a1634981d4f105d5722c549247c7a779ec31263d","source":{"kind":"arxiv","id":"2310.13486","version":1},"attestation_state":"computed","paper":{"title":"Mind the instructions: a holistic evaluation of consistency and interactions in prompt-based learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dieuwke Hupkes, Elia Bruni, Lucas Weber","submitted_at":"2023-10-20T13:25:24Z","abstract_excerpt":"Finding the best way of adapting pre-trained language models to a task is a big challenge in current NLP. Just like the previous generation of task-tuned models (TT), models that are adapted to tasks via in-context-learning (ICL) are robust in some setups but not in others. Here, we present a detailed analysis of which design choices cause instabilities and inconsistencies in LLM predictions. First, we show how spurious correlations between input distributions and labels -- a known issue in TT models -- form only a minor problem for prompted models. Then, we engage in a systematic, holistic ev"},"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":"2310.13486","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-20T13:25:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"45ff96eb0e039c1dd6268ab4e8b5edcd2be9d98837c5678379cb309063b158c6","abstract_canon_sha256":"31d912097a70ba972ade922f04ab55350f04cda106f3629b5a114d22413640dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:04.361934Z","signature_b64":"iPuLzrxcYmGRzT8HtJU0l8ktEyJX+26JgC6zuRh1iP5e+oo4k/L2bqdum1L3PBu75ajX/DHNlF2a8+twKXD0Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e8156c1ddbf837d41766809a1634981d4f105d5722c549247c7a779ec31263d","last_reissued_at":"2026-07-05T07:03:04.361469Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:04.361469Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mind the instructions: a holistic evaluation of consistency and interactions in prompt-based learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dieuwke Hupkes, Elia Bruni, Lucas Weber","submitted_at":"2023-10-20T13:25:24Z","abstract_excerpt":"Finding the best way of adapting pre-trained language models to a task is a big challenge in current NLP. Just like the previous generation of task-tuned models (TT), models that are adapted to tasks via in-context-learning (ICL) are robust in some setups but not in others. Here, we present a detailed analysis of which design choices cause instabilities and inconsistencies in LLM predictions. First, we show how spurious correlations between input distributions and labels -- a known issue in TT models -- form only a minor problem for prompted models. Then, we engage in a systematic, holistic ev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13486","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/2310.13486/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":"2310.13486","created_at":"2026-07-05T07:03:04.361525+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13486v1","created_at":"2026-07-05T07:03:04.361525+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13486","created_at":"2026-07-05T07:03:04.361525+00:00"},{"alias_kind":"pith_short_12","alias_value":"B2AVNQO5X6BX","created_at":"2026-07-05T07:03:04.361525+00:00"},{"alias_kind":"pith_short_16","alias_value":"B2AVNQO5X6BX2QLW","created_at":"2026-07-05T07:03:04.361525+00:00"},{"alias_kind":"pith_short_8","alias_value":"B2AVNQO5","created_at":"2026-07-05T07:03:04.361525+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/B2AVNQO5X6BX2QLWNAE2CY2JQH","json":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH.json","graph_json":"https://pith.science/api/pith-number/B2AVNQO5X6BX2QLWNAE2CY2JQH/graph.json","events_json":"https://pith.science/api/pith-number/B2AVNQO5X6BX2QLWNAE2CY2JQH/events.json","paper":"https://pith.science/paper/B2AVNQO5"},"agent_actions":{"view_html":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH","download_json":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH.json","view_paper":"https://pith.science/paper/B2AVNQO5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13486&json=true","fetch_graph":"https://pith.science/api/pith-number/B2AVNQO5X6BX2QLWNAE2CY2JQH/graph.json","fetch_events":"https://pith.science/api/pith-number/B2AVNQO5X6BX2QLWNAE2CY2JQH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH/action/storage_attestation","attest_author":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH/action/author_attestation","sign_citation":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH/action/citation_signature","submit_replication":"https://pith.science/pith/B2AVNQO5X6BX2QLWNAE2CY2JQH/action/replication_record"}},"created_at":"2026-07-05T07:03:04.361525+00:00","updated_at":"2026-07-05T07:03:04.361525+00:00"}