{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VGEWUJPVJWZJJ4OPREGUAI7BB4","short_pith_number":"pith:VGEWUJPV","schema_version":"1.0","canonical_sha256":"a9896a25f54db294f1cf890d4023e10f00fefd6fce3f1ab2f3e62ea9395a7da7","source":{"kind":"arxiv","id":"2403.04706","version":1},"attestation_state":"computed","paper":{"title":"Common 7B Language Models Already Possess Strong Math Capabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chen Li, Han Hu, Houwen Peng, Jingcheng Hu, Nanning Zheng, Weiqi Wang, Yixuan Wei, Zheng Zhang","submitted_at":"2024-03-07T18:00:40Z","abstract_excerpt":"Mathematical capabilities were previously believed to emerge in common language models only at a very large scale or require extensive math-related pre-training. This paper shows that the LLaMA-2 7B model with common pre-training already exhibits strong mathematical abilities, as evidenced by its impressive accuracy of 97.7% and 72.0% on the GSM8K and MATH benchmarks, respectively, when selecting the best response from 256 random generations. The primary issue with the current base model is the difficulty in consistently eliciting its inherent mathematical capabilities. Notably, the accuracy f"},"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":"2403.04706","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-07T18:00:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"dcb906ac9389ffa17a31373b5a1a2e88e61f7810f60e48df640a359569c135c4","abstract_canon_sha256":"f09df90a0d453bf6f1edc68cb926a85c48b5a7c66dbd1912de54479ac09b2381"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:53:26.938774Z","signature_b64":"CPKOQjyrS8r/y8rrfuaetCVj/WJzouQavHVHf2oKkGeG8/zmR3pWgk5690Em9WntBtNbSJ7OQcVaxn21NylVCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9896a25f54db294f1cf890d4023e10f00fefd6fce3f1ab2f3e62ea9395a7da7","last_reissued_at":"2026-07-05T07:53:26.938363Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:53:26.938363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Common 7B Language Models Already Possess Strong Math Capabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chen Li, Han Hu, Houwen Peng, Jingcheng Hu, Nanning Zheng, Weiqi Wang, Yixuan Wei, Zheng Zhang","submitted_at":"2024-03-07T18:00:40Z","abstract_excerpt":"Mathematical capabilities were previously believed to emerge in common language models only at a very large scale or require extensive math-related pre-training. This paper shows that the LLaMA-2 7B model with common pre-training already exhibits strong mathematical abilities, as evidenced by its impressive accuracy of 97.7% and 72.0% on the GSM8K and MATH benchmarks, respectively, when selecting the best response from 256 random generations. The primary issue with the current base model is the difficulty in consistently eliciting its inherent mathematical capabilities. Notably, the accuracy f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.04706","kind":"arxiv","version":1},"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/2403.04706/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":"2403.04706","created_at":"2026-07-05T07:53:26.938419+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.04706v1","created_at":"2026-07-05T07:53:26.938419+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.04706","created_at":"2026-07-05T07:53:26.938419+00:00"},{"alias_kind":"pith_short_12","alias_value":"VGEWUJPVJWZJ","created_at":"2026-07-05T07:53:26.938419+00:00"},{"alias_kind":"pith_short_16","alias_value":"VGEWUJPVJWZJJ4OP","created_at":"2026-07-05T07:53:26.938419+00:00"},{"alias_kind":"pith_short_8","alias_value":"VGEWUJPV","created_at":"2026-07-05T07:53:26.938419+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12578","citing_title":"MARD: Mirror-Augmented Reasoning Distillation for Mechanism-Level Drug-Drug Interaction Prediction","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":132,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20369","citing_title":"DEL: Digit Entropy Loss for Numerical Learning of Large Language Models","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2406.18629","citing_title":"Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2406.20094","citing_title":"Scaling Synthetic Data Creation with 1,000,000,000 Personas","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19015","citing_title":"FedProxy: Federated Fine-Tuning of LLMs via Proxy SLMs and Heterogeneity-Aware Fusion","ref_index":132,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4","json":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4.json","graph_json":"https://pith.science/api/pith-number/VGEWUJPVJWZJJ4OPREGUAI7BB4/graph.json","events_json":"https://pith.science/api/pith-number/VGEWUJPVJWZJJ4OPREGUAI7BB4/events.json","paper":"https://pith.science/paper/VGEWUJPV"},"agent_actions":{"view_html":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4","download_json":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4.json","view_paper":"https://pith.science/paper/VGEWUJPV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.04706&json=true","fetch_graph":"https://pith.science/api/pith-number/VGEWUJPVJWZJJ4OPREGUAI7BB4/graph.json","fetch_events":"https://pith.science/api/pith-number/VGEWUJPVJWZJJ4OPREGUAI7BB4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4/action/storage_attestation","attest_author":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4/action/author_attestation","sign_citation":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4/action/citation_signature","submit_replication":"https://pith.science/pith/VGEWUJPVJWZJJ4OPREGUAI7BB4/action/replication_record"}},"created_at":"2026-07-05T07:53:26.938419+00:00","updated_at":"2026-07-05T07:53:26.938419+00:00"}