AI Paper Atlas

Research map / Benchmarks / WISE

WISE leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on WISE. Best overall: 0.76 (Ours (BAGEL-7B + Process-Driven), ECCV 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.

WISE · overall

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

#overall ↑MethodPaperTopic
10.76Ours (BAGEL-7B + Process-Driven)
BAGEL-7B
Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning
ECCV 2026 · Table 2
Text-to-image diffusion
20.71MindOmni*
Qwen2.5-VL + OmniGen
MindOmni: Unleashing Reasoning Generation in Vision Language Models with RGPO
NeurIPS 2025 · WISE table
Multimodal LLMs
30.67NextStep-1 (Self-CoT)NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale
ICLR 2026 · Table 4
Text-to-image diffusion
40.64SPAR-3BSPAR: Semantic-Pixel Self-Alignment and Adaptive Routing for Unified Multimodal Models
ECCV 2026 · Table 3
Multimodal LLMs
50.63UniLIP-3BUniLiP: Adapting CLIP for Unified Multimodal Understanding, Generation and Editing
ICLR 2026 · Table 4
Multimodal LLMs
60.56MANGO-4B
Qwen2-1.5B
MANGO: Unleashing Image Generation Capability of Unified Multimodal Models
ECCV 2026 · Table 3
Multimodal LLMs
70.54MoS-L (5B)Mixture of States: Routing Token-Level Dynamics for Multimodal Generation
CVPR 2026 · Table 1
Text-to-image diffusion
80.54T2I-R1
Janus-Pro-7B
T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT
NeurIPS 2025 · WISE Result table
Reasoning and chain of thought
90.41Harmon-1.5B
Qwen2.5-1.5B-Instruct + MAR-H
Harmonizing Visual Representations for Unified Multimodal Understanding and Generation
ICCV 2025 · Table 5
Human and embodied modeling
100.33ULM-R1
Janus-Pro-1B
Co-Reinforcement Learning for Unified Multimodal Understanding and Generation
NeurIPS 2025 · Table 1
RL from feedback

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