AI Paper Atlas

Research map / Benchmarks / LVIS minival zero-shot

LVIS minival zero-shot leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on LVIS minival zero-shot. Best AP: 62.5 (T-Rex2 (Visual-I box prompt), TPAMI 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.

LVIS minival zero-shot · AP

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

#AP ↑MethodPaperTopic
162.5T-Rex2 (Visual-I box prompt)
Swin-L
T-Rex2++: Toward Generic Object Perception via Text-Visual Prompt Synergy
TPAMI 2026 · Table 2
Open-set and few-shot detection
255WeDetect-LargeWeDetect: Fast Open-Vocabulary Object Detection as Retrieval
CVPR 2026 · Table 1
Open-set and few-shot detection
354.9T-Rex2 (Text)
Swin-L
T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy
ECCV 2024 · Table 1
Open-set and few-shot detection
447.6CP-DETR-T
Swin-T
CP-DETR: Concept Prompt Guide DETR Toward Stronger Universal Object Detection
AAAI 2025 · Table 1
Open-set and few-shot detection
540.4SRA-Det
Swin-T
SRA-Det: Learning Omni-Grained Open-Vocabulary Detection Beyond Category Names
CVPR 2026 · Table 1
Open-set and few-shot detection
639.5DesCo-FIBER
Swin-B
DesCo: Learning Object Recognition with Rich Language Descriptions
NeurIPS 2023 · Table 2
Open-set and few-shot detection
738.6DetCLIP-L
Swin-L
DetCLIP: Dictionary-Enriched Visual-Concept Paralleled Pre-training for Open-world Detection
NeurIPS 2022 · Table 1
Open-set and few-shot detection
837.2ExpAlign (OG)ExpAlign: Expectation-Guided Vision–Language Alignment for Open-Vocabulary Grounding
ICML 2026 · Table 1
CLIP and zero-shot
935.4YOLO-World-L + CC3M-Lite
YOLOv8-L
YOLO-World: Real-Time Open-Vocabulary Object Detection
CVPR 2024 · Table 2
Open-set and few-shot detection
1033.9Grounding DINO L
Swin-L
Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
ECCV 2024 · Table 3
Open-set and few-shot detection

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