AI Paper Atlas

Research map / Benchmarks / MMLU

MMLU leaderboard

260 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MMLU. Best accuracy: 95.1 (M2CL, ICLR 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.

MMLU · accuracy

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

204 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

MMLU · score

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

#score ↑MethodPaperTopic
187.1PMA-init, 2e-5 -> 2e-6
Seed-MoE-15B/150B
Model Merging in Pre-training of Large Language Models
NeurIPS 2025 · Table (SFT results)
Language, safety and interpretability
285.3DR-Rubric-8B (BS-3)
Qwen3-8B-SFT
Deep Research as Rubric
NeurIPS 2026 · Table 2
Reasoning and chain of thought
385.13Ultra-253B-Base, aligned
Llama-3.1-405B-Instruct
FFN Fusion: Rethinking Sequential Computation in Large Language Models
NeurIPS 2025 · Table: Ultra-253B-Base perform
Attention and transformers
484.27InfiFPO
Phi-4
InfiFPO: Implicit Model Fusion via Preference Optimization in Large Language Models
NeurIPS 2025 · Table Main Results
Alignment and preferences
579.86MODEL-SAT
Phi-3-Mini 3.8B
Capability Instruction Tuning
AAAI 2025 · Table 2
Instruction and fine-tuning
677.2OVR-7B
Qwen2.5-VL-7B
Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoning
NeurIPS 2025 · Language benchmark table
Multimodal LLMs
775.13SlowFast Sampling
Dream 7B
Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles
ICLR 2026 · Table 1
Language, safety and interpretability
871.2Dream-Instruct + G-Star
Dream-Instruct 7B
Guided Star-Shaped Masked Diffusion
ICML 2026 · Table 1
Fast diffusion sampling
970.08StressPrompt level 6
Phi-3-mini-4k-Instruct
StressPrompt: Does Stress Impact Large Language Models and Human Performance Similarly?
AAAI 2025 · Table 1
AI and society
1069.44Qwen2-7B-Instruct + WSPO
Qwen2-7B-Instruct
Weak-to-Strong Preference Optimization: Stealing Reward from Weak Aligned Model
ICLR 2025 · Reasoning benchmark table
Alignment and preferences
36 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.