{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:Q2NY5NJSA3SGLMR3JMWZTGRFTP","short_pith_number":"pith:Q2NY5NJS","canonical_record":{"source":{"id":"2302.11042","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-21T22:47:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8fe75dd422925f53e21d8ab215242a8ed9fbf8333a3e054beca23221e823ee7c","abstract_canon_sha256":"d224dc1acc895fbf03e97552c0686962139a685608b41e0d93854539cd6af166"},"schema_version":"1.0"},"canonical_sha256":"869b8eb53206e465b23b4b2d999a259bcee30124cab0f180d89db6ad0659eba6","source":{"kind":"arxiv","id":"2302.11042","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.11042","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"arxiv_version","alias_value":"2302.11042v2","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.11042","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"pith_short_12","alias_value":"Q2NY5NJSA3SG","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"pith_short_16","alias_value":"Q2NY5NJSA3SGLMR3","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"pith_short_8","alias_value":"Q2NY5NJS","created_at":"2026-07-05T06:17:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:Q2NY5NJSA3SGLMR3JMWZTGRFTP","target":"record","payload":{"canonical_record":{"source":{"id":"2302.11042","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-21T22:47:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8fe75dd422925f53e21d8ab215242a8ed9fbf8333a3e054beca23221e823ee7c","abstract_canon_sha256":"d224dc1acc895fbf03e97552c0686962139a685608b41e0d93854539cd6af166"},"schema_version":"1.0"},"canonical_sha256":"869b8eb53206e465b23b4b2d999a259bcee30124cab0f180d89db6ad0659eba6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:17:28.089822Z","signature_b64":"7oiBAr5KWdAJFv17btNgERc/BO3jI3NCiLJ56NzCH+n1S/hgcDzgR4kRDiHpx2ndA/J9JA+kMozitm8BAKLzBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"869b8eb53206e465b23b4b2d999a259bcee30124cab0f180d89db6ad0659eba6","last_reissued_at":"2026-07-05T06:17:28.089379Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:17:28.089379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2302.11042","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:17:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z+U8Mh+CID2rd2FFxKLsUiV5Gx82WvXoaEpHRTc7lAxnSQzfvssIeYuyNNJE2M8O7UUu8/6zjc2D3wYu+gSnCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:19:15.702472Z"},"content_sha256":"6c18062e23bd45c8611cbb5fbefcdd5f880ea329b998fb049f583653f26d6f7b","schema_version":"1.0","event_id":"sha256:6c18062e23bd45c8611cbb5fbefcdd5f880ea329b998fb049f583653f26d6f7b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:Q2NY5NJSA3SGLMR3JMWZTGRFTP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"In-context Example Selection with Influences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Eric Wong, Tai Nguyen","submitted_at":"2023-02-21T22:47:45Z","abstract_excerpt":"In-context learning (ICL) is a powerful paradigm emerged from large language models (LLMs). Despite its promises, ICL performance is known to be highly sensitive to input examples. In this work, we use $\\textit{in-context influences}$ to analyze few-shot ICL performance directly from the in-context examples. Our proposed influence-based example selection method can identify both positive and negative examples, outperforming several baselines when evaluated on 9 SuperGLUE tasks. Our analysis uncovers up to a $16.3\\%$ performance gap between using the most negative in-context examples compared t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.11042","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/2302.11042/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:17:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BOODbhQIlQEEa708zKQ0EhFnC1/ezyfQTYqOTGhqqjNgKMJdX2YaGjrH7shf/oQjZcZItHex0LIbFUfP4C1ODQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:19:15.703027Z"},"content_sha256":"da341856b635a89a57fb35659b51732f447742fa9bb9f7bd5d05531e9d1da4d0","schema_version":"1.0","event_id":"sha256:da341856b635a89a57fb35659b51732f447742fa9bb9f7bd5d05531e9d1da4d0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP/bundle.json","state_url":"https://pith.science/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T17:19:15Z","links":{"resolver":"https://pith.science/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP","bundle":"https://pith.science/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP/bundle.json","state":"https://pith.science/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q2NY5NJSA3SGLMR3JMWZTGRFTP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:Q2NY5NJSA3SGLMR3JMWZTGRFTP","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d224dc1acc895fbf03e97552c0686962139a685608b41e0d93854539cd6af166","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-21T22:47:45Z","title_canon_sha256":"8fe75dd422925f53e21d8ab215242a8ed9fbf8333a3e054beca23221e823ee7c"},"schema_version":"1.0","source":{"id":"2302.11042","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.11042","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"arxiv_version","alias_value":"2302.11042v2","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.11042","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"pith_short_12","alias_value":"Q2NY5NJSA3SG","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"pith_short_16","alias_value":"Q2NY5NJSA3SGLMR3","created_at":"2026-07-05T06:17:28Z"},{"alias_kind":"pith_short_8","alias_value":"Q2NY5NJS","created_at":"2026-07-05T06:17:28Z"}],"graph_snapshots":[{"event_id":"sha256:da341856b635a89a57fb35659b51732f447742fa9bb9f7bd5d05531e9d1da4d0","target":"graph","created_at":"2026-07-05T06:17:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2302.11042/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In-context learning (ICL) is a powerful paradigm emerged from large language models (LLMs). Despite its promises, ICL performance is known to be highly sensitive to input examples. In this work, we use $\\textit{in-context influences}$ to analyze few-shot ICL performance directly from the in-context examples. Our proposed influence-based example selection method can identify both positive and negative examples, outperforming several baselines when evaluated on 9 SuperGLUE tasks. Our analysis uncovers up to a $16.3\\%$ performance gap between using the most negative in-context examples compared t","authors_text":"Eric Wong, Tai Nguyen","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-21T22:47:45Z","title":"In-context Example Selection with Influences"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.11042","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6c18062e23bd45c8611cbb5fbefcdd5f880ea329b998fb049f583653f26d6f7b","target":"record","created_at":"2026-07-05T06:17:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d224dc1acc895fbf03e97552c0686962139a685608b41e0d93854539cd6af166","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-21T22:47:45Z","title_canon_sha256":"8fe75dd422925f53e21d8ab215242a8ed9fbf8333a3e054beca23221e823ee7c"},"schema_version":"1.0","source":{"id":"2302.11042","kind":"arxiv","version":2}},"canonical_sha256":"869b8eb53206e465b23b4b2d999a259bcee30124cab0f180d89db6ad0659eba6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"869b8eb53206e465b23b4b2d999a259bcee30124cab0f180d89db6ad0659eba6","first_computed_at":"2026-07-05T06:17:28.089379Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:17:28.089379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7oiBAr5KWdAJFv17btNgERc/BO3jI3NCiLJ56NzCH+n1S/hgcDzgR4kRDiHpx2ndA/J9JA+kMozitm8BAKLzBA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:17:28.089822Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.11042","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6c18062e23bd45c8611cbb5fbefcdd5f880ea329b998fb049f583653f26d6f7b","sha256:da341856b635a89a57fb35659b51732f447742fa9bb9f7bd5d05531e9d1da4d0"],"state_sha256":"be1f65e4aa0665f75aab923cb2289c9f9b65d66f1a015270333f7c2a6fa529c5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mrGOiN40rliN1IS1eNbM9EEwp8tQ+S4UixfPw0FSugW8mIFBq0ThOJJ8J8G19N08gWHIk1zsPhft2Rx55FiJDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T17:19:15.708511Z","bundle_sha256":"f047b59d83de659660751df30a69112a5d3be3415e8ece3d4e353d201223aa12"}}