{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7IKDFPQIEKALA74YZ4BWOQDWIB","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":"713271090429f7c7fe31f21e78811b016fcd52de31e1813a2381bd55465a84d8","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-11T07:29:56Z","title_canon_sha256":"d784e60246792da41abb727dd47a340520da84124b2f5111e5baf6e7a9a59cd3"},"schema_version":"1.0","source":{"id":"2504.08312","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.08312","created_at":"2026-07-05T10:47:46Z"},{"alias_kind":"arxiv_version","alias_value":"2504.08312v1","created_at":"2026-07-05T10:47:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.08312","created_at":"2026-07-05T10:47:46Z"},{"alias_kind":"pith_short_12","alias_value":"7IKDFPQIEKAL","created_at":"2026-07-05T10:47:46Z"},{"alias_kind":"pith_short_16","alias_value":"7IKDFPQIEKALA74Y","created_at":"2026-07-05T10:47:46Z"},{"alias_kind":"pith_short_8","alias_value":"7IKDFPQI","created_at":"2026-07-05T10:47:46Z"}],"graph_snapshots":[{"event_id":"sha256:14556db4c26668ae8c64d753f9756618bdce80b87e40e1457597873cad3dcf9e","target":"graph","created_at":"2026-07-05T10:47:46Z","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/2504.08312/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Sorting is a tedious but simple task for human intelligence and can be solved fairly easily algorithmically. However, for Large Language Models (LLMs) this task is surprisingly hard, as some properties of sorting are among known weaknesses of LLMs: being faithful to the input data, logical comparisons between values, and strictly differentiating between syntax (used for sorting) and semantics (typically learned by embeddings). Within this paper, we describe the new SortBench benchmark for LLMs that comes with different difficulties and that can be easily scaled in terms of difficulty. We apply","authors_text":"Steffen Herbold","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-11T07:29:56Z","title":"SortBench: Benchmarking LLMs based on their ability to sort lists"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.08312","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":"sha256:47458502c26c4dafd28dabb0f74d06dc503929fe9ebfc342c053494324b22dfa","target":"record","created_at":"2026-07-05T10:47:46Z","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":"713271090429f7c7fe31f21e78811b016fcd52de31e1813a2381bd55465a84d8","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-11T07:29:56Z","title_canon_sha256":"d784e60246792da41abb727dd47a340520da84124b2f5111e5baf6e7a9a59cd3"},"schema_version":"1.0","source":{"id":"2504.08312","kind":"arxiv","version":1}},"canonical_sha256":"fa1432be082280b07f98cf036740764043c7ae0eb1c003003eddc6196e7bb417","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fa1432be082280b07f98cf036740764043c7ae0eb1c003003eddc6196e7bb417","first_computed_at":"2026-07-05T10:47:46.914575Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:47:46.914575Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rn2hm/CqZVpNXD0M3jWyrpJ8WfU679ksR1Hk1v87qbAo2XJU8lD5jp4yk6CnoT/BER1TrflvnpIXu0yrMsqiDg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:47:46.915056Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.08312","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:47458502c26c4dafd28dabb0f74d06dc503929fe9ebfc342c053494324b22dfa","sha256:14556db4c26668ae8c64d753f9756618bdce80b87e40e1457597873cad3dcf9e"],"state_sha256":"cf3038e7b4b5f1ce51d01ef96a0ecf45266a71ee3e563459765f2909827a6090"}