AI Paper Atlas

Research map / Benchmarks / TabMWP

TabMWP leaderboard

19 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on TabMWP. Best accuracy: 99.2 (TableMix (ours), 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.

TabMWP · accuracy

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

#accuracy ↑MethodPaperTopic
199.2TableMix (ours)
Qwen2.5-VL-7B
TableMix: Enhancing Multimodal Table Reasoning in MLLMs from a Data-Centric Perspective
CVPR 2026 · Table 1
Multimodal LLMs
297.1MM-Table-R1-3B
Qwen2.5-VL-3B
Multimodal Table Understanding with Difficulty-aware Reinforcement Learning
AAAI 2026 · Table 1
Multimodal LLMs
396.75Turbo
Ovis2-8B
Multimodal Tabular Reasoning with Privileged Structured Information
NeurIPS 2025 · Main results table
Multimodal LLMs
496.52VISTA-R1
InternVL3-8B
Scaling Agentic Reinforcement Learning for Tool-Integrated Reasoning in VLMs
CVPR 2026 · Table 1
Multimodal LLMs
596.21Llama 3 trained with illustrative solutions
Llama 3 8B
Template-Driven LLM-Paraphrased Framework for Tabular Math Word Problem Generation
AAAI 2025 · Table 3
Agents, code and math
693.7TabFlash
InternVL-2.5 3B
TabFlash: Efficient Table Understanding with Progressive Question Conditioning and Token Focusing
AAAI 2026 · Table 1
Multimodal LLMs
790KTCE
GPT-3.5-Turbo
Automated Creation of Reusable and Diverse Toolsets for Enhancing LLM Reasoning
AAAI 2025 · Table 2
Agents, code and math
889.2GPT-3.5-turbo (ReGAL)
GPT-3.5-turbo
ReGAL: Refactoring Programs to Discover Generalizable Abstractions
ICML 2024 · Table 1
Agents, code and math
988.4CRAFT
GPT-3.5-Turbo
CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets
ICLR 2024 · Experimental Results table
Agents, code and math
1088.3SynTab-LLaVA
Vicuna-1.5 7B
SynTab-LLaVA: Enhancing Multimodal Table Understanding with Decoupled Synthesis
CVPR 2025 · Table 1
Multimodal LLMs
9 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.