{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7V7IEDWWTVQGKAYHOJVLIVTHWT","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":"5b7d247802fcad1aef3bf4b19a6064888bb6e2681031da07d51d0bd4f705c2a8","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-03T19:18:06Z","title_canon_sha256":"1257fa62537be955d96612c748e48fd3331eb0101805b3da6b3be74d9dcbdc03"},"schema_version":"1.0","source":{"id":"2505.01903","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01903","created_at":"2026-07-05T11:17:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01903v2","created_at":"2026-07-05T11:17:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01903","created_at":"2026-07-05T11:17:45Z"},{"alias_kind":"pith_short_12","alias_value":"7V7IEDWWTVQG","created_at":"2026-07-05T11:17:45Z"},{"alias_kind":"pith_short_16","alias_value":"7V7IEDWWTVQGKAYH","created_at":"2026-07-05T11:17:45Z"},{"alias_kind":"pith_short_8","alias_value":"7V7IEDWW","created_at":"2026-07-05T11:17:45Z"}],"graph_snapshots":[{"event_id":"sha256:938451e4ddca88a3d9e5eec1735d5722d01b2d309efa2d1885d54641e61aa2f7","target":"graph","created_at":"2026-07-05T11:17:45Z","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/2505.01903/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are increasingly used to generate distractors for multiple-choice questions (MCQs), especially in domains like math education. However, existing approaches are limited in ensuring that the generated distractors are consistent with common student errors. We propose LookAlike, a method that improves error-distractor consistency via preference optimization. Our two main innovations are: (a) mining synthetic preference pairs from model inconsistencies, and (b) alternating supervised fine-tuning (SFT) with Direct Preference Optimization (DPO) to stabilize training. Unli","authors_text":"Alexander Scarlatos, Andrew Lan, Nigel Fernandez, Nisarg Parikh, Simon Woodhead","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-03T19:18:06Z","title":"LookAlike: Consistent Distractor Generation in Math MCQs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01903","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":"sha256:04c346239c79ddbdc38d3bcf8a0e9b640c50a7c0c12fcc92a0ef8eb199111b20","target":"record","created_at":"2026-07-05T11:17:45Z","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":"5b7d247802fcad1aef3bf4b19a6064888bb6e2681031da07d51d0bd4f705c2a8","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-03T19:18:06Z","title_canon_sha256":"1257fa62537be955d96612c748e48fd3331eb0101805b3da6b3be74d9dcbdc03"},"schema_version":"1.0","source":{"id":"2505.01903","kind":"arxiv","version":2}},"canonical_sha256":"fd7e820ed69d60650307726ab45667b4efbbcac9863d25552b5192aa6dba3b2c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fd7e820ed69d60650307726ab45667b4efbbcac9863d25552b5192aa6dba3b2c","first_computed_at":"2026-07-05T11:17:45.127866Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:17:45.127866Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YGVUA4Id5CbdTPnNViFKOFZm4Qf8/AJ7E79auf3ZUS8N14DZwxBaxcHxC8koyVUhUlQ7jUSb2xy9Wk8lWO3yCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:17:45.128371Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.01903","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:04c346239c79ddbdc38d3bcf8a0e9b640c50a7c0c12fcc92a0ef8eb199111b20","sha256:938451e4ddca88a3d9e5eec1735d5722d01b2d309efa2d1885d54641e61aa2f7"],"state_sha256":"b37738531059a39236556f035c18734267f134047398448e883d99436a14621d"}