{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MH3SDVIEORIIKXJ2VEW6AQL565","short_pith_number":"pith:MH3SDVIE","schema_version":"1.0","canonical_sha256":"61f721d5047450855d3aa92de0417df76ad22664bf64010978eac8fcd4ab6781","source":{"kind":"arxiv","id":"2211.04486","version":1},"attestation_state":"computed","paper":{"title":"Active Example Selection for In-Context Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenhao Tan, Shi Feng, Yiming Zhang","submitted_at":"2022-11-08T19:00:02Z","abstract_excerpt":"With a handful of demonstration examples, large-scale language models show strong capability to perform various tasks by in-context learning from these examples, without any fine-tuning. We demonstrate that in-context learning performance can be highly unstable across samples of examples, indicating the idiosyncrasies of how language models acquire information. We formulate example selection for in-context learning as a sequential decision problem, and propose a reinforcement learning algorithm for identifying generalizable policies to select demonstration examples. For GPT-2, our learned poli"},"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":"2211.04486","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-08T19:00:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f5f931646ea0f74b5b256acdf2e0b6246ec72bf1a27d9c83ee4011711f4d5d38","abstract_canon_sha256":"48760efeb3bb088252b579c51965cd6ee29dc7f7364d0689c6245a4fd55b4374"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:14:38.559056Z","signature_b64":"a55SORGhqFwRZZcVED/aUAfjsM7NZrOcvRD+YeZzgjA7HqnLGPZKfwvVQ/1oDS1XYmo0R6eVy2/uYdAx8fmbDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61f721d5047450855d3aa92de0417df76ad22664bf64010978eac8fcd4ab6781","last_reissued_at":"2026-07-05T05:14:38.558503Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:14:38.558503Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Active Example Selection for In-Context Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chenhao Tan, Shi Feng, Yiming Zhang","submitted_at":"2022-11-08T19:00:02Z","abstract_excerpt":"With a handful of demonstration examples, large-scale language models show strong capability to perform various tasks by in-context learning from these examples, without any fine-tuning. We demonstrate that in-context learning performance can be highly unstable across samples of examples, indicating the idiosyncrasies of how language models acquire information. We formulate example selection for in-context learning as a sequential decision problem, and propose a reinforcement learning algorithm for identifying generalizable policies to select demonstration examples. For GPT-2, our learned poli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.04486","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/2211.04486/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":"2211.04486","created_at":"2026-07-05T05:14:38.558563+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.04486v1","created_at":"2026-07-05T05:14:38.558563+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.04486","created_at":"2026-07-05T05:14:38.558563+00:00"},{"alias_kind":"pith_short_12","alias_value":"MH3SDVIEORII","created_at":"2026-07-05T05:14:38.558563+00:00"},{"alias_kind":"pith_short_16","alias_value":"MH3SDVIEORIIKXJ2","created_at":"2026-07-05T05:14:38.558563+00:00"},{"alias_kind":"pith_short_8","alias_value":"MH3SDVIE","created_at":"2026-07-05T05:14:38.558563+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08068","citing_title":"DICE: Entropy-Regularized Equilibrium Selection for Stable Multi-Agent LLM Coordination","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2510.15079","citing_title":"Assessing Coherency and Consistency of Code Execution Reasoning by Large Language Models","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11699","citing_title":"Legal2LogicICL: Improving Generalization in Transforming Legal Cases to Logical Formulas via Diverse Few-Shot Learning","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565","json":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565.json","graph_json":"https://pith.science/api/pith-number/MH3SDVIEORIIKXJ2VEW6AQL565/graph.json","events_json":"https://pith.science/api/pith-number/MH3SDVIEORIIKXJ2VEW6AQL565/events.json","paper":"https://pith.science/paper/MH3SDVIE"},"agent_actions":{"view_html":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565","download_json":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565.json","view_paper":"https://pith.science/paper/MH3SDVIE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.04486&json=true","fetch_graph":"https://pith.science/api/pith-number/MH3SDVIEORIIKXJ2VEW6AQL565/graph.json","fetch_events":"https://pith.science/api/pith-number/MH3SDVIEORIIKXJ2VEW6AQL565/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565/action/storage_attestation","attest_author":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565/action/author_attestation","sign_citation":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565/action/citation_signature","submit_replication":"https://pith.science/pith/MH3SDVIEORIIKXJ2VEW6AQL565/action/replication_record"}},"created_at":"2026-07-05T05:14:38.558563+00:00","updated_at":"2026-07-05T05:14:38.558563+00:00"}