{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:UUYTUGXVVNULC4VB55BZADAJCB","short_pith_number":"pith:UUYTUGXV","schema_version":"1.0","canonical_sha256":"a5313a1af5ab68b172a1ef43900c0910602ab79235fb6020345776bdbc88cd00","source":{"kind":"arxiv","id":"2305.09859","version":4},"attestation_state":"computed","paper":{"title":"Smaller Language Models are Better Black-box Machine-Generated Text Detectors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Justus Mattern, Niloofar Mireshghallah, Reza Shokri, Sicun Gao, Taylor Berg-Kirkpatrick","submitted_at":"2023-05-17T00:09:08Z","abstract_excerpt":"With the advent of fluent generative language models that can produce convincing utterances very similar to those written by humans, distinguishing whether a piece of text is machine-generated or human-written becomes more challenging and more important, as such models could be used to spread misinformation, fake news, fake reviews and to mimic certain authors and figures. To this end, there have been a slew of methods proposed to detect machine-generated text. Most of these methods need access to the logits of the target model or need the ability to sample from the target. One such black-box "},"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":"2305.09859","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-17T00:09:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"02077d41a1f03f1c09512c685bc9dcab59002744522b644385d4651b4f945fad","abstract_canon_sha256":"8168a8871a2e590d1a1f020718f829555214e1da4df78158c938fb1e73c3ed35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:46.385455Z","signature_b64":"ppaZpuZZiSpZX22CaSM2/aPdFWG/Jo/KjjkFqH8BERC+dFv7OoacCUGYk20nLzRy5j44k6W4QPgm9FaIHvL3BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5313a1af5ab68b172a1ef43900c0910602ab79235fb6020345776bdbc88cd00","last_reissued_at":"2026-07-05T07:48:46.384965Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:46.384965Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Smaller Language Models are Better Black-box Machine-Generated Text Detectors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Justus Mattern, Niloofar Mireshghallah, Reza Shokri, Sicun Gao, Taylor Berg-Kirkpatrick","submitted_at":"2023-05-17T00:09:08Z","abstract_excerpt":"With the advent of fluent generative language models that can produce convincing utterances very similar to those written by humans, distinguishing whether a piece of text is machine-generated or human-written becomes more challenging and more important, as such models could be used to spread misinformation, fake news, fake reviews and to mimic certain authors and figures. To this end, there have been a slew of methods proposed to detect machine-generated text. Most of these methods need access to the logits of the target model or need the ability to sample from the target. One such black-box "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.09859","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/2305.09859/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":"2305.09859","created_at":"2026-07-05T07:48:46.385021+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.09859v4","created_at":"2026-07-05T07:48:46.385021+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.09859","created_at":"2026-07-05T07:48:46.385021+00:00"},{"alias_kind":"pith_short_12","alias_value":"UUYTUGXVVNUL","created_at":"2026-07-05T07:48:46.385021+00:00"},{"alias_kind":"pith_short_16","alias_value":"UUYTUGXVVNULC4VB","created_at":"2026-07-05T07:48:46.385021+00:00"},{"alias_kind":"pith_short_8","alias_value":"UUYTUGXV","created_at":"2026-07-05T07:48:46.385021+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23190","citing_title":"Hidden Human-Like Nature of Machine-Generated Texts: Theory and Detection Enhancement","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB","json":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB.json","graph_json":"https://pith.science/api/pith-number/UUYTUGXVVNULC4VB55BZADAJCB/graph.json","events_json":"https://pith.science/api/pith-number/UUYTUGXVVNULC4VB55BZADAJCB/events.json","paper":"https://pith.science/paper/UUYTUGXV"},"agent_actions":{"view_html":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB","download_json":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB.json","view_paper":"https://pith.science/paper/UUYTUGXV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.09859&json=true","fetch_graph":"https://pith.science/api/pith-number/UUYTUGXVVNULC4VB55BZADAJCB/graph.json","fetch_events":"https://pith.science/api/pith-number/UUYTUGXVVNULC4VB55BZADAJCB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB/action/storage_attestation","attest_author":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB/action/author_attestation","sign_citation":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB/action/citation_signature","submit_replication":"https://pith.science/pith/UUYTUGXVVNULC4VB55BZADAJCB/action/replication_record"}},"created_at":"2026-07-05T07:48:46.385021+00:00","updated_at":"2026-07-05T07:48:46.385021+00:00"}