AI Paper Atlas

Research map / Benchmarks / Charades-STA

Charades-STA leaderboard

103 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on Charades-STA. Best mIoU: 67.7 (VideoChat-M1 (37B), CVPR 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.

Charades-STA · mIoU

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

#mIoU ↑MethodPaperTopic
167.7VideoChat-M1 (37B)
4-agent group
VideoChat-M1: Collaborative Policy Planning for Video Understanding via Multi-Agent Reinforcement Learning
CVPR 2026 · Table 1
Cooperative multi-agent RL
262.16EMB (ELA, multi-candidate)
VSLNet
Video Activity Localisation with Uncertainties in Temporal Boundary
ECCV 2022 · Table 2
Video-language understanding
361.2SDST (Ours)
InternVideo2
Sparse-Dense Side-Tuner for efficient Video Temporal Grounding
ICCV 2025 · Table 2
Video-language understanding
460.7Video-o3
Qwen2.5-VL-7B
Video-o3: Native Interleaved Clue Seeking for Long Video Multi-Hop Reasoning
ICML 2026 · Table 8
Video-language understanding
560.6VideoChat-R1.5-7B
Qwen2.5-VL-7B
VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception
NeurIPS 2025 · fine-tuning temporal grounding
Multimodal LLMs
660VideoAuto-R1
Qwen2.5-VL-7B
VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice
CVPR 2026 · Table 4
Video-language understanding
759.9VITAL-7B
Qwen2.5-VL-7B
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning
CVPR 2026 · Table 4
Video-language understanding
859.1+ RL (ours)
Qwen2-VL-7B-Base
Incentivizing Versatile Video Reasoning in MLLMs via Data-Efficient Reinforcement Learning
CVPR 2026 · Table 3
Multimodal LLMs
957.8VideoTemp-o3-7B-RL
Qwen2.5-VL-7B
VideoTemp-o3: Harmonizing Temporal Grounding and Video Understanding in Agentic Thinking-with-Videos
ICML 2026 · Table 2
Video-language understanding
1056.2Ours (w/ frame-level)
SlowFast+CLIP-B
Object-Centric Framework for Video Moment Retrieval
AAAI 2026 · Table 1
Video-language understanding
39 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

Charades-STA · R@1, IoU=0.5

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

#R@1, IoU=0.5 ↑MethodPaperTopic
169.93MKTNMulti-Pair Temporal Sentence Grounding via Multi-Thread Knowledge Transfer Network
AAAI 2025 · Table 1
Video-language understanding
268.82SpotVMRFewer Steps, Better Performance: Efficient Cross-Modal Clip Trimming for Video Moment Retrieval Using Language
AAAI 2024 · Table 1
Video-language understanding
364.06SMINStructured Multi-Level Interaction Network for Video Moment Localization via Language Query
CVPR 2021 · Table 1
Video-language understanding
463.98MGSL-NetMemory-Guided Semantic Learning Network for Temporal Sentence Grounding
AAAI 2022 · Table 1
Video-language understanding
559.7DeNet
I3D
Embracing Uncertainty: Decoupling and De-Bias for Robust Temporal Grounding
CVPR 2021 · Table 1
Video-language understanding
658.4TimeChat-VT
TimeChat
On the Consistency of Video Large Language Models in Temporal Comprehension
CVPR 2025 · Table 3
Video-language understanding
754.44Ours-SCDMSemantic Conditioned Dynamic Modulation for Temporal Sentence Grounding in Videos
NeurIPS 2019 · Table 1
Video-language understanding
854.31Ours(Shapley)Rethinking Weakly-supervised Video Temporal Grounding From a Game Perspective
ECCV 2024 · Table 1
Video-language understanding
953.85TCSFYou Can Ground Earlier Than See: An Effective and Efficient Pipeline for Temporal Sentence Grounding in Compressed Videos
CVPR 2023 · Table 2
Video segmentation and tracking
1053.09DRN
I3D
Dense Regression Network for Video Grounding
CVPR 2020 · Table: Comparisons with state-
Video-language understanding
23 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

Charades-STA · R@1, IoU=0.7

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

#R@1, IoU=0.7 ↑MethodPaperTopic
147.39SpotVMRFewer Steps, Better Performance: Efficient Cross-Modal Clip Trimming for Video Moment Retrieval Using Language
AAAI 2024 · Table 1
Video-language understanding
243.09MARN
C3D
Exploring Motion and Appearance Information for Temporal Sentence Grounding
AAAI 2022 · Table 3
Video segmentation and tracking
341.03MGSL-NetMemory-Guided Semantic Learning Network for Temporal Sentence Grounding
AAAI 2022 · Table 1
Video-language understanding
440.75SMINStructured Multi-Level Interaction Network for Video Moment Localization via Language Query
CVPR 2021 · Table 1
Video-language understanding
538.52DeNet
I3D
Embracing Uncertainty: Decoupling and De-Bias for Robust Temporal Grounding
CVPR 2021 · Table 1
Video-language understanding
637.2TCSFYou Can Ground Earlier Than See: An Effective and Efficient Pipeline for Temporal Sentence Grounding in Compressed Videos
CVPR 2023 · Table 2
Video segmentation and tracking
734.48MHSTHypotheses Tree Building for One-Shot Temporal Sentence Localization
AAAI 2023 · Table 2
Video-language understanding
833.43Ours-SCDMSemantic Conditioned Dynamic Modulation for Temporal Sentence Grounding in Videos
NeurIPS 2019 · Table 1
Video-language understanding
931.75DRN
I3D
Dense Regression Network for Video Grounding
CVPR 2020 · Table: Comparisons with state-
Video-language understanding
1029.25UMT video+audioUMT: Unified Multi-Modal Transformers for Joint Video Moment Retrieval and Highlight Detection
CVPR 2022 · Table: Comparison on Charades-
Video-language understanding
18 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

Charades-STA · R1@0.5

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

#R1@0.5 ↑MethodPaperTopic
172.5Invert4TVG
Qwen2.5-VL-7B
Invert4TVG: A Temporal Video Grounding Framework with Inversion Tasks Preserving Action Understanding Ability
ICLR 2026 · Table 1
Action recognition
272.1Non-linear reward shaping
Qwen2.5-VL-7B-Instruct
TempSamp-R1: Effective Temporal Sampling with Reinforcement Fine-Tuning for Video LLMs
NeurIPS 2025 · Table 4
Reasoning and chain of thought
372SDST (Ours)
InternVideo2
Sparse-Dense Side-Tuner for efficient Video Temporal Grounding
ICCV 2025 · Table 2
Video-language understanding
470.14SnAG w/ TD-DETRThe Devil is in the Spurious Correlations: Boosting Moment Retrieval with Dynamic Learning
ICCV 2025 · Table 2
Video-language understanding
570VideoLoom-8B
InternVL3-8B
VideoLoom: A Video Large Language Model for Joint Spatial-Temporal Understanding
ICML 2026 · Table 1
Video-language understanding
665.1Ours (w/ frame-level)
SlowFast+CLIP-B
Object-Centric Framework for Video Moment Retrieval
AAAI 2026 · Table 1
Video-language understanding
764.5SpaceVLLM-7BSpaceVLLM: Endowing Multimodal Large Language Model with Spatio-Temporal Video Grounding Capability
AAAI 2026 · Table 6
Video-language understanding
862.61CVACVA: Context-aware Video-text Alignment for Video Temporal Grounding
CVPR 2026 · Table 3
Video-language understanding
961.72QD-DETR+Ours (SHINE)
QD-DETR
SHINE: Saliency-aware HIerarchical NEgative Ranking for Compositional Temporal Grounding
ECCV 2024 · Appendix Table (Charades-STA)
Video-language understanding
1061.34Sim-DETRSim-DETR: Unlock DETR for Temporal Sentence Grounding
ICCV 2025 · Table 2
Video-language understanding
12 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

Charades-STA · R@0.5

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

#R@0.5 ↑MethodPaperTopic
163.03Three-stream DRFT
LGI
End-to-end Multi-modal Video Temporal Grounding
NeurIPS 2021 · Table 2
Video-language understanding
260.6+T2SGrid-FT
Qwen2-VL-7B
T2SGrid: Temporal-to-Spatial Gridification for Video Temporal Grounding
CVPR 2026 · Table 1
Video-language understanding
359.46LGI (Ours)Local-Global Video-Text Interactions for Temporal Grounding
CVPR 2020 · Table 1
Video-language understanding
459.1VideoMind
Qwen2-VL-7B
VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning
ICLR 2026 · Table 5
Video-language understanding
558.2MomentSeg-3B
Qwen2.5-VL-3B
MomentSeg: Moment-Centric Sampling for Enhanced Referring Video Object Segmentation
ECCV 2026 · Table 2
Video segmentation and tracking
657.93RGTRDiversifying Query: Region-Guided Transformer for Temporal Sentence Grounding
AAAI 2025 · Table 2
Video-language understanding
754.72MMDist TeacherLocal-Global Multi-Modal Distillation for Weakly-Supervised Temporal Video Grounding
AAAI 2024 · Table 1
Video-language understanding
853.77OmniD-TeacherOmnipotent Distillation with LLMs for Weakly-Supervised Natural Language Video Localization: When Divergence Meets Consistency
AAAI 2024 · Table 3
Language, safety and interpretability
953.26Ours (w/ Swin-B)
Swin-B
ProTeGe: Untrimmed Pretraining for Video Temporal Grounding by Video Temporal Grounding
CVPR 2023 · Table 1a
Video-language understanding
1052.69HUAL (50%, 5)
SeqPAN
Are Binary Annotations Sufficient? Video Moment Retrieval via Hierarchical Uncertainty-Based Active Learning
CVPR 2023 · Table 1
Video-language understanding
8 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

Charades-STA · R@5, IoU=0.5

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

#R@5, IoU=0.5 ↑MethodPaperTopic
193.21MGSL-NetMemory-Guided Semantic Learning Network for Temporal Sentence Grounding
AAAI 2022 · Table 1
Video-language understanding
291.24DeNet
I3D
Embracing Uncertainty: Decoupling and De-Bias for Robust Temporal Grounding
CVPR 2021 · Table 1
Video-language understanding
389.49SMINStructured Multi-Level Interaction Network for Video Moment Localization via Language Query
CVPR 2021 · Table 1
Video-language understanding
489.41UMT video+optical flowUMT: Unified Multi-Modal Transformers for Joint Video Moment Retrieval and Highlight Detection
CVPR 2022 · Table: Comparison on Charades-
Video-language understanding
589.29MHSTHypotheses Tree Building for One-Shot Temporal Sentence Localization
AAAI 2023 · Table 2
Video-language understanding
689.24Ours(Shapley)Rethinking Weakly-supervised Video Temporal Grounding From a Game Perspective
ECCV 2024 · Table 1
Video-language understanding
788.01DRN
C3D
Dense Regression Network for Video Grounding
CVPR 2020 · Table: Comparisons with state-
Video-language understanding
886.32SCANetSCANet: Scene Complexity Aware Network for Weakly-Supervised Video Moment Retrieval
ICCV 2023 · Table 1
Video-language understanding
986.23PPSGaussian Mixture Proposals with Pull-Push Learning Scheme to Capture Diverse Events for Weakly Supervised Temporal Video Grounding
AAAI 2024 · Table 3
Video-language understanding
1086.23MANMAN: Moment Alignment Network for Natural Language Moment Retrieval via Iterative Graph Adjustment
CVPR 2019 · Charades-STA results table
Video-language understanding
6 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

Charades-STA · R@0.7

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

#R@0.7 ↑MethodPaperTopic
140.15Three-stream DRFT
LGI
End-to-end Multi-modal Video Temporal Grounding
NeurIPS 2021 · Table 2
Video-language understanding
237.02R2-Tuning
CLIP ViT-B/32
R^2-Tuning: Efficient Image-to-Video Transfer Learning for Video Temporal Grounding
ECCV 2024 · Table 2
Video-language understanding
335.48LGI (Ours)Local-Global Video-Text Interactions for Temporal Grounding
CVPR 2020 · Table 1
Video-language understanding
433VideoTemp-o3-7B-RL
Qwen2.5-VL-7B
VideoTemp-o3: Harmonizing Temporal Grounding and Video Understanding in Agentic Thinking-with-Videos
ICML 2026 · Table 2
Video-language understanding
531.8REVISOR (Ours)
Qwen2.5-VL-7B
REVISOR: Beyond Textual Reflection, Towards Multimodal Introspective Reasoning in Long-Form Video Understanding
CVPR 2026 · Table 2
Video-language understanding
630.38Ours (w/ Swin-B)
Swin-B
ProTeGe: Untrimmed Pretraining for Video Temporal Grounding by Video Temporal Grounding
CVPR 2023 · Table 1a
Video-language understanding
728.9HUAL (50%, 5)
SeqPAN
Are Binary Annotations Sufficient? Video Moment Retrieval via Hierarchical Uncertainty-Based Active Learning
CVPR 2023 · Table 1
Video-language understanding
826MMDist TeacherLocal-Global Multi-Modal Distillation for Weakly-Supervised Temporal Video Grounding
AAAI 2024 · Table 1
Video-language understanding
925.59HiTeA (Qwen2.5-VL)
Qwen2.5-VL-7B
HiTeA: Hierarchical Temporal Alignment for Training-Free Long-Video Temporal Grounding
ICLR 2026 · Table 2
Video-language understanding
1020.94BTDP
EMB
Boundary-Aware Temporal Dynamic Pseudo-Supervision Pairs Generation for Zero-Shot Natural Language Video Localization
AAAI 2025 · Table 1
Video-language understanding
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

Charades-STA · R@5, IoU=0.7

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

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

Charades-STA · R1@0.7

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

3 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

Charades-STA · R@1, IoU=0.3

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

#R@1, IoU=0.3 ↑MethodPaperTopic
171.6MMRGMulti-Modal Relational Graph for Cross-Modal Video Moment Retrieval
CVPR 2021 · Table 1
Cross-modal learning
269.06PPSGaussian Mixture Proposals with Pull-Push Learning Scheme to Capture Diverse Events for Weakly Supervised Temporal Video Grounding
AAAI 2024 · Table 3
Video-language understanding
368.04SCANetSCANet: Scene Complexity Aware Network for Weakly-Supervised Video Moment Retrieval
ICCV 2023 · Table 1
Video-language understanding
466.4CPLWeakly Supervised Temporal Sentence Grounding With Gaussian-Based Contrastive Proposal Learning
CVPR 2022 · Table 1
Video-language understanding
565.92TVP (both prompts)
ResNet-50
Text-Visual Prompting for Efficient 2D Temporal Video Grounding
CVPR 2023 · Table, Charades-STA results
Video-language understanding
658.08PMI-LOC (C+IRV2+A)
C3D+IRV2+audio
Learning Modality Interaction for Temporal Sentence Localization and Event Captioning in Videos
ECCV 2020 · Table 5
Video-language understanding
755.46BPNet
VSLNet backbone
Boundary Proposal Network for Two-stage Natural Language Video Localization
AAAI 2021 · Table: Charades-STA results
Video-language understanding
845.24VLANetVLANet: Video-Language Alignment Network for Weakly-Supervised Video Moment Retrieval
ECCV 2020 · Table 1
Video-language understanding
944.15DSCNetUnsupervised Temporal Video Grounding with Deep Semantic Clustering
AAAI 2022 · Table 1
Video-language understanding
1043.31OursExplore Inter-contrast between Videos via Composition for Weakly Supervised Temporal Sentence Grounding
AAAI 2022 · Table 2
Video-language understanding
1 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.