AI Paper Atlas

Research map / Benchmarks / UCF-101 zero-shot

UCF-101 zero-shot leaderboard

17 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on UCF-101 zero-shot. Best accuracy: 91.5 (IMP-MoE-L, NeurIPS 2023).

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.

UCF-101 zero-shot · accuracy

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

#accuracy ↑MethodPaperTopic
191.5IMP-MoE-LAlternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perception
NeurIPS 2023 · Table 1
Cross-modal learning
288.7MoTE
ViT-L/14
MoTE: Reconciling Generalization with Specialization for Visual-Language to Video Knowledge Transfer
NeurIPS 2024 · Zero-shot video recognition ta
Video-language understanding
383.9ALT-L/14
CLIP ViT-L/14
Align Before Adapt: Leveraging Entity-to-Region Alignments for Generalizable Video Action Recognition
CVPR 2024 · Table 3
Action recognition
481.28GTA-CLIP
CLIP ViT-L/14
Generate, Transduct, Adapt: Iterative Transduction with VLMs
ICCV 2025 · Table 1
CLIP and zero-shot
579.7OST (32 frames)
CLIP ViT-B/16
OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition
CVPR 2024 · Table 1
Action recognition
678.9Storyboard-guided Alignment
CLIP
Storyboard-guided Alignment for Fine-grained Video Action Recognition
NeurIPS 2025 · Zero-shot comparison table
Action recognition
778.7+ MSTA + LCC
ViCLIP
Efficient Transfer Learning for Video-language Foundation Models
CVPR 2025 · Table 4
Video-language understanding
877.6ViCLIP + CONDA
ViCLIP ViT-B/16
Condensed Test-Time Adaptation of VLMs for Action Recognition
CVPR 2026 · Table 1
CLIP and zero-shot
976.8ViFi-CLIP
CLIP ViT-B/16
Fine-Tuned CLIP Models Are Efficient Video Learners
CVPR 2023 · Zero-shot results table
Video-language understanding
1075.1ReGen (BLIP)
BLIP
ReGen: A good Generative Zero-Shot Video Classifier Should be Rewarded
ICCV 2023 · Table 4
Zero-/few-shot and meta-learning
7 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.