{"schema":"https://assignee.net/schemas/benchmark-evidence-v1","schema_version":"1.0","contract_version":"benchmark-evidence-v1.0","contract_updated":"2026-06-01","schema_documentation":"https://assignee.net/schemas","changelog_url":"https://assignee.net/changelog","publisher":{"name":"Assignee Research","url":"https://assignee.net"},"html_url":"https://assignee.net/benchmarks/evidence?model=Llama-2&bench=MATH","json_url":"https://assignee.net/benchmarks/evidence.json?model=Llama-2&bench=MATH","model":"Llama-2","benchmark":"MATH","source_count":4,"source_coverage":{"record_count":4,"distinct_source_count":4,"coverage_level":"MODERATE","basis":"distinct public paper URLs or titles in this evidence cluster"},"source_profile":{"source_url_count":4,"missing_source_url_count":0,"domains":["arxiv.org"],"year_min":2023,"year_max":2025,"basis":"public source URLs, source titles, and reported publication years in this evidence cluster"},"reported_range":{"min_score_pct":14.9,"max_score_pct":88.3},"spread_pp":73.4,"severity":"HIGH","entries":[{"model":"Llama-2","benchmark":"MATH","score_pct":88.33,"source_title":"GuiLoMo: Allocating Expert Number and Rank for LoRA-MoE via Bilevel Optimization with GuidedSelection Vectors","source_url":"http://arxiv.org/abs/2506.14646v2","source_domain":"arxiv.org","year":2025},{"model":"Llama-2","benchmark":"MATH","score_pct":28.9,"source_title":"SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models","source_url":"http://arxiv.org/abs/2403.07384v2","source_domain":"arxiv.org","year":2024},{"model":"Llama-2","benchmark":"MATH","score_pct":19.2,"source_title":"Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations","source_url":"http://arxiv.org/abs/2312.08935v3","source_domain":"arxiv.org","year":2023},{"model":"Llama-2","benchmark":"MATH","score_pct":14.9,"source_title":"DONOD: Efficient and Generalizable Instruction Fine-Tuning for LLMs via Model-Intrinsic Dataset Pruning","source_url":"https://arxiv.org/abs/2504.14810","source_domain":"arxiv.org","year":2025}],"interpretation":"This record groups score claims extracted from papers for the same model and benchmark label. A nonzero spread means the public literature reports different values for this cluster.","limitations":["Differences are not automatically errors.","Reported values may differ because of prompts, dataset versions, evaluation protocols, scoring rules, preprocessing, fine-tuning, or reporting conventions.","Source papers remain authoritative for their own claims."]}