{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XMXG7MSMWAFKE6ID4CG6FMO7RL","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":"bea006f78108582c0ecc0df58fe84bbe190b0f3068c27374cfe0019037f4dbd5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T17:09:22Z","title_canon_sha256":"b40f50a44b098c7692e13b50843c3c8af3d5e345e580a6b47afea75deb8d1a07"},"schema_version":"1.0","source":{"id":"2410.08996","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.08996","created_at":"2026-07-05T09:19:20Z"},{"alias_kind":"arxiv_version","alias_value":"2410.08996v1","created_at":"2026-07-05T09:19:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.08996","created_at":"2026-07-05T09:19:20Z"},{"alias_kind":"pith_short_12","alias_value":"XMXG7MSMWAFK","created_at":"2026-07-05T09:19:20Z"},{"alias_kind":"pith_short_16","alias_value":"XMXG7MSMWAFKE6ID","created_at":"2026-07-05T09:19:20Z"},{"alias_kind":"pith_short_8","alias_value":"XMXG7MSM","created_at":"2026-07-05T09:19:20Z"}],"graph_snapshots":[{"event_id":"sha256:1fb6c5ad25f9b4694a19f0b7c521a159992cafec3725c2b45b594fffb86011bb","target":"graph","created_at":"2026-07-05T09:19:20Z","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/2410.08996/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We test whether replacing crowdsource workers with LLMs to write Natural Language Inference (NLI) hypotheses similarly results in annotation artifacts. We recreate a portion of the Stanford NLI corpus using GPT-4, Llama-2 and Mistral 7b, and train hypothesis-only classifiers to determine whether LLM-elicited hypotheses contain annotation artifacts. On our LLM-elicited NLI datasets, BERT-based hypothesis-only classifiers achieve between 86-96% accuracy, indicating these datasets contain hypothesis-only artifacts. We also find frequent \"give-aways\" in LLM-generated hypotheses, e.g. the phrase \"s","authors_text":"Adam Poliak, Grace Proebsting","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T17:09:22Z","title":"Hypothesis-only Biases in Large Language Model-Elicited Natural Language Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.08996","kind":"arxiv","version":1},"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:a0b6acc7e7f1d1f68383ab92447bf5d36174516d101aa3a5749ea55438a8e212","target":"record","created_at":"2026-07-05T09:19:20Z","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":"bea006f78108582c0ecc0df58fe84bbe190b0f3068c27374cfe0019037f4dbd5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-11T17:09:22Z","title_canon_sha256":"b40f50a44b098c7692e13b50843c3c8af3d5e345e580a6b47afea75deb8d1a07"},"schema_version":"1.0","source":{"id":"2410.08996","kind":"arxiv","version":1}},"canonical_sha256":"bb2e6fb24cb00aa27903e08de2b1df8af048e3575a93a4ae4ad84cdd942cfedd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bb2e6fb24cb00aa27903e08de2b1df8af048e3575a93a4ae4ad84cdd942cfedd","first_computed_at":"2026-07-05T09:19:20.452404Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:19:20.452404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FLpnOL1RKVzLeMFAI4OlvmKYcUYS8QnDGLMVvYB6Nag3bXjwZj+PdY/O9pFz31oIutfB518Swghkd15Gj4LxAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:19:20.453184Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.08996","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a0b6acc7e7f1d1f68383ab92447bf5d36174516d101aa3a5749ea55438a8e212","sha256:1fb6c5ad25f9b4694a19f0b7c521a159992cafec3725c2b45b594fffb86011bb"],"state_sha256":"ae8991f2ac1ae4d00ec766714f80917d533bab8e85215d1ac229287095f88f38"}