AI Paper Atlas

Research map / Benchmarks / RefCOCO testB

RefCOCO testB leaderboard

75 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on RefCOCO testB. Best accuracy: 92 (Florence-2-L, CVPR 2024).

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.

RefCOCO testB · accuracy

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

19 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

RefCOCO testB · mIoU

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

#mIoU ↑MethodPaperTopic
180.7GenMask
WAN-2.1 DiT 1.3B + Qwen2.5-VL-7B
GenMask: Adapting DiT for Segmentation via Direct Mask Generation
CVPR 2026 · Table 1
Autoregressive and latent generation
276.6UniRES
CLIP
Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation
CVPR 2024 · Table 4
Open-set and few-shot detection
376.3Ours-LWIMFRIS: WIndow Mamba Fusion and Parameter Efficient Tuning for Referring Image Segmentation
ICLR 2026 · Table 2
CNN features
475.67TCRT
Swin-B
Task-aware Cross-modal Feature Refinement Transformer with Large Language Models for Visual Grounding
CVPR 2025 · Table 2
Human and embodied modeling
572.3RefFormerReferencing Where to Focus: Improving Visual Grounding with Referential Query
NeurIPS 2024 · Table 2
Human and embodied modeling
670.94LAVT
Swin-B + BERT
Language-Aware Vision Transformer for Referring Segmentation
TPAMI 2024 · Table 7
Cross-modal learning
757.25WeakMCN†WeakMCN: Multi-task Collaborative Network for Weakly Supervised Referring Expression Comprehension and Segmentation
CVPR 2025 · Table 1
Weakly supervised detection and segmentation
856.97DViN
YOLOv5-Seg
DViN: Dynamic Visual Routing Network for Weakly Supervised Referring Expression Comprehension
CVPR 2025 · Table 6
Human and embodied modeling
946.28PPT
CLIP ViT-B/16 + SAM ViT-B
Curriculum Point Prompting for Weakly-Supervised Referring Image Segmentation
CVPR 2024 · Table 1
Weakly supervised detection and segmentation
1045.19Ours
SAM, CLIP ViT-B/16
Hybrid Global-Local Representation with Augmented Spatial Guidance for Zero-Shot Referring Image Segmentation
CVPR 2025 · Table 1
CLIP and zero-shot
4 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

RefCOCO testB · cIoU

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

#cIoU ↑MethodPaperTopic
184.7ALToLLM-8B (AL)
InternVL-2.5-8B
ALTo: Adaptive-Length Tokenizer for Autoregressive Mask Generation
NeurIPS 2025 · Table: Performance comparison
Multimodal LLMs
284InstructSeg
Mipha-3B
InstructSeg: Unifying Instructed Visual Segmentation with Multi-modal Large Language Models
ICCV 2025 · Table 2
Video-language understanding
379.5CoPRS-7B
Qwen2.5-VL-7B
CoPRS: Learning Positional Prior from Chain-of-Thought for Reasoning Segmentation
ICLR 2026 · Table 1
Multimodal LLMs
478.4DenseMLLM
DenseMLLM-4B
DenseMLLM: Standard Multimodal LLMs for Dense Prediction
ICML 2026 · Table 1
Multimodal LLMs
577.7UGround-7BUGround: Towards Unified Visual Grounding with Unrolled Transformers
ICML 2026 · Table 3
Multimodal LLMs
676.57SegAgent-Qwen+SimpleClick
Qwen-VL-7B
SegAgent: Exploring Pixel Understanding Capabilities in MLLMs by Imitating Human Annotator Trajectories
CVPR 2025 · RES comparison table
Multimodal LLMs
773.2READ
LLaVA1.5-7B
Reasoning to Attend: Try to Understand How <SEG> Token Works
CVPR 2025 · RefCOCO(+/g) results table
Multimodal LLMs
872.2Vitron
Vicuna 7B v1.5
Vitron: A Unified Pixel-level Vision LLM for Understanding, Generating, Segmenting, Editing
NeurIPS 2024 · Table: Results of referring im
Multimodal LLMs
970.3RESAnything
Pixtral 12B
RESAnything: Attribute Prompting for Arbitrary Referring Segmentation
NeurIPS 2025 · Table: Quantitative results on
Open-set and few-shot detection
1069.5NExT-Chat
Vicuna-1.5
NExT-Chat: An LMM for Chat, Detection and Segmentation
ICML 2024 · Table 2
Open-set and few-shot detection
2 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

RefCOCO testB · Acc@0.5

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

2 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

RefCOCO testB · oIoU

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

#oIoU ↑MethodPaperTopic
171.35SaFiRe
Swin-B + BERT
SaFiRe: Saccade-Fixation Reiteration with Mamba for Referring Image Segmentation
NeurIPS 2025 · Table: Comparison of RIS Metho
Open-set and few-shot detection
271.05MagNet
Swin-B
Mask Grounding for Referring Image Segmentation
CVPR 2024 · Table 1
Human and embodied modeling
371.04LQMFormer
Swin-B
LQMFormer: Language-aware Query Mask Transformer for Referring Image Segmentation
CVPR 2024 · Table 1
Human and embodied modeling
470.89ReMamber
Mamba-B
ReMamber: Referring Image Segmentation with Mamba Twister
ECCV 2024 · Table 2
Cross-modal learning
570.58CoupAlign
Swin-B
CoupAlign: Coupling Word-Pixel with Sentence-Mask Alignments for Referring Image Segmentation
NeurIPS 2022 · Table 1
Open-set and few-shot detection
670.16DMMI
Swin-B
Beyond One-to-One: Rethinking the Referring Image Segmentation
ICCV 2023 · Table 1
Open-set and few-shot detection
770.06LAVT + SADLR
LAVT
Semantics-Aware Dynamic Localization and Refinement for Referring Image Segmentation
AAAI 2023 · Table 1
Open-set and few-shot detection
868.79LAVT
Swin-B + BERT
Language-Aware Vision Transformer for Referring Segmentation
TPAMI 2024 · Table 1
Cross-modal learning
968.79LAVT
Swin-B + BERT
LAVT: Language-Aware Vision Transformer for Referring Image Segmentation
CVPR 2022 · Table 1
Human and embodied modeling
1067.46InstructPix2Pix-SD2.0 + ADDP
Stable Diffusion 2.0
Aligning Generative Denoising with Discriminative Objectives Unleashes Diffusion for Visual Perception
ICLR 2025 · Table 2
Fast diffusion sampling

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