AI Paper Atlas

Research map / Benchmarks / MathVista mini

MathVista mini leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MathVista mini. Best accuracy: 82.1 (CARE, CVPR 2026).

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.

MathVista mini · accuracy

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

#accuracy ↑MethodPaperTopic
182.1CARE
Qwen3-VL-8B-Instruct
CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal Reasoning
CVPR 2026 · Table 1
Reasoning and chain of thought
281.7MMFineReason-8B (SFT+RL)
Qwen3-VL-8B-Instruct
MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods
NeurIPS 2026 · Table 3
Multimodal LLMs
381.4Bee-8B-RL
Qwen3-8B
Bee: A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs
ICLR 2026 · Table 2
Multimodal LLMs
478R-4B-RL (A-T)
R-4B
R-4B: Incentivizing General-Purpose Auto-Thinking in MLLMs via Bi-Mode Annealing and Reinforce Learning
CVPR 2026 · Table 1
Reasoning and chain of thought
572.7GThinker-7B
Qwen2.5-VL-7B
GThinker: Towards General Multimodal Reasoning via Cue-Guided Rethinking
CVPR 2026 · Table 2
Multimodal LLMs
670Thyme-VL-7B
Qwen2.5-VL-7B
Thyme: Think Beyond Images
ICLR 2026 · Table 1
Multimodal LLMs
769.78FHI + class-conditioned rescaling (Ours)
Kimi-VL-A3B-Thinking
Reallocating Attention Across Layers to Reduce Multimodal Hallucination
CVPR 2026 · Table 1
Multimodal LLMs
868.4LaCoT-Qwen-7B
Qwen2.5-VL-7B
Latent Chain-of-Thought for Visual Reasoning
NeurIPS 2025 · Table: Test accuracy on visual
Reasoning and chain of thought
967.9Monet-SFT + VLPO + Attn (Ours)
Monet-7B
Leveraging Latent Visual Reasoning in Silence
NeurIPS 2026 · Table 1
Multimodal LLMs
1063.4Circle-RoPE
Qwen2.5-VL-3B
Circle-RoPE: Cone-like Decoupled Rotary Positional Embedding for Vision-Language Models
ICML 2026 · Table (main comparison)
Multimodal LLMs

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