{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4XYXTRIQTDYZVEKPXMLIBPVVQA","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":"0a8cb8dafa05ba5c023391f56930b9e956719ce9a4d556e84dbc3152fb8baaf5","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T13:02:12Z","title_canon_sha256":"fe8b48fba75fef2c4b0c53d1c861ca508e5fcaa41f1edb762d92be462529fcd0"},"schema_version":"1.0","source":{"id":"2406.12572","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12572","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12572v3","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12572","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"pith_short_12","alias_value":"4XYXTRIQTDYZ","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"pith_short_16","alias_value":"4XYXTRIQTDYZVEKP","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"pith_short_8","alias_value":"4XYXTRIQ","created_at":"2026-07-05T09:20:44Z"}],"graph_snapshots":[{"event_id":"sha256:71e6011d1b474fd2ad3bf5a0a39f0039d42a1569689e0d47683d92cd5698bd65","target":"graph","created_at":"2026-07-05T09:20:44Z","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/2406.12572/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce Mathador-LM, a new benchmark for evaluating the mathematical reasoning on large language models (LLMs), combining ruleset interpretation, planning, and problem-solving. This benchmark is inspired by the Mathador game, where the objective is to reach a target number using basic arithmetic operations on a given set of base numbers, following a simple set of rules. We show that, across leading LLMs, we obtain stable average performance while generating benchmark instances \\emph{dynamically}, following a target difficulty level. Thus, our benchmark alleviates concerns about test-set l","authors_text":"Amir Moeini, Dan Alistarh, Eldar Kurtic","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T13:02:12Z","title":"Mathador-LM: A Dynamic Benchmark for Mathematical Reasoning on Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12572","kind":"arxiv","version":3},"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:b536f85cd4594f43bb7057017ee419c96639b4015c87e17518eb4b997babd4db","target":"record","created_at":"2026-07-05T09:20:44Z","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":"0a8cb8dafa05ba5c023391f56930b9e956719ce9a4d556e84dbc3152fb8baaf5","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T13:02:12Z","title_canon_sha256":"fe8b48fba75fef2c4b0c53d1c861ca508e5fcaa41f1edb762d92be462529fcd0"},"schema_version":"1.0","source":{"id":"2406.12572","kind":"arxiv","version":3}},"canonical_sha256":"e5f179c51098f19a914fbb1680beb5803bbc1e7a393e382c693fad98e80b8e81","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e5f179c51098f19a914fbb1680beb5803bbc1e7a393e382c693fad98e80b8e81","first_computed_at":"2026-07-05T09:20:44.486688Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:20:44.486688Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jdQMOWlH48f3uAGD0II44QhcS6WgsO1/2VQdRxGS7RjzUi6ydWrkvJvaKFdvoJ2RJWxM1wJjvw4sw3wDZLuMAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:20:44.487138Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.12572","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b536f85cd4594f43bb7057017ee419c96639b4015c87e17518eb4b997babd4db","sha256:71e6011d1b474fd2ad3bf5a0a39f0039d42a1569689e0d47683d92cd5698bd65"],"state_sha256":"8e6bd4c58891244881275e4aaadc7505440ea80279879f248aa5778e6211ff86"}