{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PHCMAHGK4PPTIWORWOWSILMP3N","short_pith_number":"pith:PHCMAHGK","schema_version":"1.0","canonical_sha256":"79c4c01ccae3df3459d1b3ad242d8fdb7daca57d4fb0d79c824ac4bb22a1f878","source":{"kind":"arxiv","id":"2306.13651","version":2},"attestation_state":"computed","paper":{"title":"Bring Your Own Data! Self-Supervised Evaluation for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aniruddha Saha, John Kirchenbauer, Jonas Geiping, Khalid Saifullah, Manli Shu, Micah Goldblum, Neel Jain, Tom Goldstein, Yuxin Wen","submitted_at":"2023-06-23T17:59:09Z","abstract_excerpt":"With the rise of Large Language Models (LLMs) and their ubiquitous deployment in diverse domains, measuring language model behavior on realistic data is imperative. For example, a company deploying a client-facing chatbot must ensure that the model will not respond to client requests with profanity. Current evaluations approach this problem using small, domain-specific datasets with human-curated labels. These evaluation sets are often sampled from a narrow and simplified distribution, and data sources can unknowingly be leaked into the training set which can lead to misleading evaluations. To"},"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":"2306.13651","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-23T17:59:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0671cf794722ddd2af295d4faa1e5cadac2a683b95319706c97aef72a404e3d5","abstract_canon_sha256":"950978af0406cdb502c12a2f669896cebc28e34db8fd210a7c8d81994f6f2469"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:26:03.381813Z","signature_b64":"/OG17hZR5smzlbdWZdg+XqmBcTka6PrCAmysmog9Dkr/3baRXXyuWcjNpNwQ4zb3i/dWlpmvZYvkTlC6Mc1yBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"79c4c01ccae3df3459d1b3ad242d8fdb7daca57d4fb0d79c824ac4bb22a1f878","last_reissued_at":"2026-07-05T06:26:03.381419Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:26:03.381419Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bring Your Own Data! Self-Supervised Evaluation for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aniruddha Saha, John Kirchenbauer, Jonas Geiping, Khalid Saifullah, Manli Shu, Micah Goldblum, Neel Jain, Tom Goldstein, Yuxin Wen","submitted_at":"2023-06-23T17:59:09Z","abstract_excerpt":"With the rise of Large Language Models (LLMs) and their ubiquitous deployment in diverse domains, measuring language model behavior on realistic data is imperative. For example, a company deploying a client-facing chatbot must ensure that the model will not respond to client requests with profanity. Current evaluations approach this problem using small, domain-specific datasets with human-curated labels. These evaluation sets are often sampled from a narrow and simplified distribution, and data sources can unknowingly be leaked into the training set which can lead to misleading evaluations. To"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.13651","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/2306.13651/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":"2306.13651","created_at":"2026-07-05T06:26:03.381474+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.13651v2","created_at":"2026-07-05T06:26:03.381474+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.13651","created_at":"2026-07-05T06:26:03.381474+00:00"},{"alias_kind":"pith_short_12","alias_value":"PHCMAHGK4PPT","created_at":"2026-07-05T06:26:03.381474+00:00"},{"alias_kind":"pith_short_16","alias_value":"PHCMAHGK4PPTIWOR","created_at":"2026-07-05T06:26:03.381474+00:00"},{"alias_kind":"pith_short_8","alias_value":"PHCMAHGK","created_at":"2026-07-05T06:26:03.381474+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2401.02458","citing_title":"Data-Centric Foundation Models in Computational Healthcare: A Survey","ref_index":126,"is_internal_anchor":false},{"citing_arxiv_id":"2406.04244","citing_title":"Benchmark Data Contamination of Large Language Models: A Survey","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2309.00614","citing_title":"Baseline Defenses for Adversarial Attacks Against Aligned Language Models","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N","json":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N.json","graph_json":"https://pith.science/api/pith-number/PHCMAHGK4PPTIWORWOWSILMP3N/graph.json","events_json":"https://pith.science/api/pith-number/PHCMAHGK4PPTIWORWOWSILMP3N/events.json","paper":"https://pith.science/paper/PHCMAHGK"},"agent_actions":{"view_html":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N","download_json":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N.json","view_paper":"https://pith.science/paper/PHCMAHGK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.13651&json=true","fetch_graph":"https://pith.science/api/pith-number/PHCMAHGK4PPTIWORWOWSILMP3N/graph.json","fetch_events":"https://pith.science/api/pith-number/PHCMAHGK4PPTIWORWOWSILMP3N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N/action/storage_attestation","attest_author":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N/action/author_attestation","sign_citation":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N/action/citation_signature","submit_replication":"https://pith.science/pith/PHCMAHGK4PPTIWORWOWSILMP3N/action/replication_record"}},"created_at":"2026-07-05T06:26:03.381474+00:00","updated_at":"2026-07-05T06:26:03.381474+00:00"}