AI Paper Atlas

Research map / Benchmarks / MathQA

MathQA leaderboard

17 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MathQA. Best accuracy: 85.4 (ShorterBetter-7B (α=2), NeurIPS 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

MathQA · accuracy

17 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
185.4ShorterBetter-7B (α=2)
DeepSeek-R1-Distill-Qwen-7B
ShorterBetter: Guiding Reasoning Models to Find Optimal Inference Length for Efficient Reasoning
NeurIPS 2025 · 7B results table
Reasoning and chain of thought
280.4Ours
Llama-3.1-8B-Instruct
Translate Policy to Language: Flow Matching Generated Rewards for LLM Explanations
ICLR 2026 · Table 1
Agents, code and math
377.79IGLB
True/False softmax score
Multicalibration for Confidence Scoring in LLMs
ICML 2024 · Table 1
Language, safety and interpretability
476.4MEML-GRPO (Expert0)
Qwen2.5-1.5B-Math
MEML-GRPO: Heterogeneous Multi-Expert Mutual Learning for RLVR Advancement
AAAI 2026 · Table 4
Reasoning and chain of thought
561.34HCInfer
Qwen-30B-A3B (4-bit)
HCInfer: An Efficient Inference System via Heterogeneous Error Compensation for Resource-Constrained Devices
NeurIPS 2026 · Table 1
Quantization
660Complex CoT + Vote Complex
Code-davinci-002
Complexity-Based Prompting for Multi-step Reasoning
ICLR 2023 · Main results table
Reasoning and chain of thought
758.8Qwen3-8B-Base + ED-SFT
Qwen3-8B-Base
Echoes as Anchors: Probabilistic Costs and Attention Refocusing in LLM Reasoning
ICLR 2026 · Table 3
Reasoning and chain of thought
856.62C3oT
LLaMA-2-Chat-13B
C3oT: Generating Shorter Chain-of-Thought Without Compromising Effectiveness
AAAI 2025 · Main Results table
Reasoning and chain of thought
951.42LATE FUSION+MNL
Qwen2.5-0.5B-Instruct; Qwen2.5-1.5B
Multimodal Negative Learning
NeurIPS 2025 · MathQA table
Cross-modal learning
1040.6QJL
Llama-3-8B
QJL: 1-Bit Quantized JL Transform for KV Cache Quantization with Zero Overhead
AAAI 2025 · LM-eval results table
Quantization
7 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.