MIRA-Math: A Benchmark for Minimal Information Requesting and Mathematical Reasoning

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Charbel Al Bateh, Samer Saab
• Venue: arXiv: cs.AI
• Published: 2026-07-08
• Collected: 2026-07-09
• URL: http://arxiv.org/abs/2607.07391v1
• PDF: https://arxiv.org/pdf/2607.07391v1
• DOI: UNKNOWN

Plain-Language Brief

The researchers introduce MIRA-Math, a benchmark for testing AI models' ability to ask for and integrate missing facts to solve mathematical problems. The benchmark contains 2310 instances from 22 mathematical families.

What the Document Says

• MIRA-Math is a benchmark for minimal information requesting in mathematical reasoning.
• It tests AI models' ability to ask for missing facts and integrate them into solutions.
• The benchmark contains 2310 instances from 22 mathematical families.
• These families include algebra, probability, linear systems, discrete structures, signal processing, Markov chains, circuits, interpolation, and numerical boundary-value problems.

Key Concepts Mentioned

• MIRA-Math
• mathematical reasoning benchmark
• minimal information requesting
• AI models
• mathematical families

Claims Made by the Document

• MIRA-Math tests AI models' ability to ask for and integrate missing facts.
• The benchmark contains instances from various mathematical fields.

Evidence Used

• The authors describe MIRA-Math as a benchmark for minimal information requesting in mathematical reasoning.
• They mention that the benchmark contains 2310 instances from 22 mathematical families.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Mathematical reasoning benchmarks typically provide all facts needed to solve each problem, while interactive benchmarks often mix reasoning with tools, retrieval, and long-horizon dialogue. We introduce MIRA-Math, a benchmark for a narrower diagnostic capability: solving mathematical problems whose full latent state has a unique answer, but whose solver-facing view is missing exactly one necessary atomic fact. The solver must request the missing information in natural language under a strict budget and then integrate the returned fact into an exact final answer. A fixed constrained LLM responder sees only the dataset-provided atomic fact and must either offer the quoted fact when the request matches it, or decline otherwise. Thus, instance generation, typed hint specifications, validation, and final-answer verification are deterministic, while request metrics are measured under a fixed LLM-mediated responder channel. MIRA-Math contains 2{,}310 generated instances from 22 typed mathematical families spanning algebra, probability, linear systems, discrete structures, signal processing, Markov chains, circuits, interpolation, and numerical boundary-value problems. Experiments across frontier and small models show that request success and final-answer accuracy are separable: models may ask for the right fact yet fail the downstream computation, or fail before obtaining the canonical hint. We release generators, verifiers, prompts, run metadata, and dataset documentation to support reproducible evaluation of minimal information requesting in mathematical reasoning.