Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks150 from public data
- Earthformer: Exploring Space-Time Transformers for Earth System Forecasting416
The idea is to decompose the data into cuboids and apply cuboid-level self-attention in parallel, which achieves state-of-the-art performance on real-world benchmarks about precipitation nowcasting and El Nino/Southern Oscillation forecasting.
- 365
- AIM: Adapting Image Models for Efficient Video Action Recognition273
This work proposes a novel method to Adapt pre-trained Image Models (AIM) for efficient video understanding, and shows that the proposed AIM can achieve competitive or even better performance than prior arts with substantially fewer tunable parameters on four video action recognition benchmarks.
- Towards Geospatial Foundation Models via Continual Pretraining169
This work investigates the potential of continual pretraining from large-scale ImageNet-22k models and proposes a multi-objective continual pretraining paradigm, which leverages the strong representations of ImageNet while simultaneously providing the freedom to learn valuable in-domain features.
- Visual Prompt Tuning for Test-time Domain Adaptation124
This work tackles the practical yet challenging test-time adaptation (TTA) problem, where a model adapts to the target domain without accessing the source data, with a simple recipe called Data-efficient Prompt Tuning (DePT).
- Can We Use Arbitrary Objects to Attack LiDAR Perception in Autonomous Driving?102
This paper investigates an easier way to perform effective adversarial attacks with high flexibility and good stealthiness against LiDAR perception in autonomous driving and proposes a novel attack framework based on which the attacker can identify a few adversarial locations in the physical space.
- 84
- Parsing Tweets into Universal Dependencies77
It is shown that it is challenging to deliver consistent annotation due to ambiguity in understanding and explaining tweets and proposed a new method to distill an ensemble of 20 transition-based parsers into a single one that achieves an improvement of 2.2 in LAS over the un-ensembled baseline and outperforms parsers that are state-of-the-art on other treebanks in both accuracy and speed.
- 75
- Adversarial Attacks against LiDAR Semantic Segmentation in Autonomous Driving73
This paper proposes a novel adversarial attack framework based on which the attacker can easily fool LiDAR semantic segmentation by placing some simple objects at some locations in the physical space, and conducts extensive real-world experiments to evaluate the performance of the proposed attack framework.
- Partial and Asymmetric Contrastive Learning for Out-of-Distribution Detection in Long-Tailed Recognition70
This work demonstrates that existing OOD detection methods commonly suffer from significant performance degradation when the training set is long-tail distributed, and proposes Partial and Asymmetric Supervised Contrastive Learning (PASCL), which explicitly encourages the model to distinguish between tail-class in-distribution samples and OOD samples.
- A Closer Look at Few-Shot Crosslingual Transfer: The Choice of Shots Matters70
This work conducts a large-scale experimental study on 40 sets of sampled few shots for six diverse NLP tasks across up to 40 languages, and shows that a straightforward full model finetuning approach is quite effective for few-shot transfer, outperforming several state-of-the-art few- shot approaches.
- Unsupervised Semantic Segmentation with Self-supervised Object-centric Representations63
A methodology based on unsupervised saliency masks and self-supervised feature clustering to kickstart object discovery followed by training a semantic segmentation network on pseudo-labels to bootstrap the system on images with multiple objects is proposed.
- ViewCo: Discovering Text-Supervised Segmentation Masks via Multi-View Semantic Consistency62
The text-to-views consistency modeling benefits the dense assignment of the visual features by encouraging different crops to align with the same text, while the cross-view segmentation consistency modeling provides additional self-supervision, overcoming the limitation of ambiguous text supervision for segmentation masks.
- Land use classification using convolutional neural networks applied to ground-level images53
This work presents a general framework that uses ground-level images from Flickr for land use mapping, and utilizes high-level semantic image features extracted using deep learning, specifically convolutional neural networks, which allow it to achieve upwards of 76% accuracy on a challenging eight class land use maps problem.
- 50
- 48
- CoupAlign: Coupling Word-Pixel with Sentence-Mask Alignments for Referring Image Segmentation48
CoupAlign is proposed, a simple yet effective multi-level visual-semantic alignment method to couple sentence-mask alignment with word-pixel alignment to enforce object mask constraint for achieving more accurate localization and segmentation.
- Towards Backdoor Attacks against LiDAR Object Detection in Autonomous Driving47
A novel backdoor attack strategy is proposed based on which the attacker can achieve the attack goal by poisoning a small number of point cloud samples and it is physically realizable, and it allows the attacker to easily perform the attack using some common objects as the triggers.
- MetaWave: Attacking mmWave Sensing with Meta-material-enhanced Tags39
This paper proposes and design a novel passive mmWave attack with low-cost and easily obtainable meta-material tags for both vanish and ghost attack types, and demonstrates that tags made of ordinal material can be leveraged to precisely tamper the mmWave echo signal and spoof the range, angle, and speed sensing measurements.
- Malicious Attacks against Multi-Sensor Fusion in Autonomous Driving38
The first study on the vulnerability of multi-sensor fusion systems that employ LiDAR, camera, and radar is presented, which proposes a novel attack method that can simultaneously attack all three types of sensing modalities using a single type of adversarial object.
- Fine-Grained Visual–Text Prompt-Driven Self-Training for Open-Vocabulary Object Detection36
This article proposes a simple but effective fine-grained visual-text prompt-driven self-training paradigm for open-vocabulary detection (VTP-OVD) that introduces a fine-grained visual–text prompt adapting stage to enhance the current self-training paradigm with a more powerful fine-grained alignment.
- A Systematic Study of Leveraging Subword Information for Learning Word Representations35
This work proposes a general framework for learning subword-informed word representations that allows for easy experimentation with different segmentation and composition components, also including more advanced techniques based on position embeddings and self-attention.
- Contrastive Instruction-Trajectory Learning for Vision-Language Navigation33
A Contrastive Instruction-Trajectory Learning (CITL) framework that explores invariance across similar data samples and variance across different ones to learn distinctive representations for robust navigation and can be easily integrated with VLN backbones to form a new learning paradigm and achieve better generalizability in unseen environments.
- 29
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.