K-LITE: Learning Transferable Visual Models with External Knowledge
Comments NeurIPS 2022 camera ready
作者
Trevor Darrell
Computer Vision
Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectly
Comments ECCV 2022. Code and models are available here: https://github.com/facebookresearch/reliable_vqa
Real-World Robot Learning with Masked Visual Pre-training
Comments CoRL 2022; Project page: https://tetexiao.com/projects/real-mvp
Studying Bias in GANs through the Lens of Race
Comments ECCV 2022. Project Page: https://neerja.me/bias-gans/
Exploring Simple and Transferable Recognition-Aware Image Processing
Comments IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
Prior Knowledge-Guided Attention in Self-Supervised Vision Transformers
Visual Prompting via Image Inpainting
Comments Project page: https://yossigandelsman.github.io/visual_prompt
TL;DW? Summarizing Instructional Videos with Task Relevance & Cross-Modal Saliency
Comments Accepted to ECCV 2022. Website: https://medhini.github.io/ivsum/
Towards Learning to Play Piano with Dexterous Hands and Touch
Disentangled Action Recognition with Knowledge Bases
Comments NAACL 2022
Structured Video Tokens @ Ego4D PNR Temporal Localization Challenge 2022
Comments Ego4D CVPR22 Object State Localization challenge. arXiv admin note: substantial text overlap with arXiv:2206.06346
Visual Attention Emerges from Recurrent Sparse Reconstruction
Comments Released code: https://github.com/bfshi/VARS
Object-Region Video Transformers
Comments CVPR 2022
Voxel-informed Language Grounding
Comments ACL 2022
ReCLIP: A Strong Zero-Shot Baseline for Referring Expression Comprehension
Comments ACL 2022
Twitter-COMMs: Detecting Climate, COVID, and Military Multimodal Misinformation
Comments 11 pages, 6 figures
Anytime Dense Prediction with Confidence Adaptivity
Comments Published in ICLR 2022
Contrastive Test-Time Adaptation
Comments CVPR 2022 camera-ready version
Learning to Listen: Modeling Non-Deterministic Dyadic Facial Motion
Learning to Detect Every Thing in an Open World
Comments Project page is available at https://ksaito-ut.github.io/openworld_ldet/
Explaining Reinforcement Learning Policies through Counterfactual Trajectories
Comments Accepted at ICML HILL 2021 Workshop
Masked Visual Pre-training for Motor Control
Comments Code and videos at: https://tetexiao.com/projects/mvp
A ConvNet for the 2020s
Comments CVPR 2022; Code: https://github.com/facebookresearch/ConvNeXt
On Guiding Visual Attention with Language Specification
Comments 14 pages, 9 figures
Un-Mix: Rethinking Image Mixtures for Unsupervised Visual Representation Learning
Comments AAAI 2022 camera ready version with Appendix (add a formula example with InfoNCE). Code is available at: https://github.com/szq0214/Un-Mix
Region-level Active Detector Learning
CLIP-It! Language-Guided Video Summarization
Comments Neurips 2021. Website at https://medhini.github.io/clip_it/
Journal ref Thirty-Fifth Conference on Neural Information Processing Systems. 2021
Regularization Matters in Policy Optimization
Comments Published at ICLR 2021; please cite this paper's ICLR 2021 version at https://github.com/xuanlinli17/iclr2021_rlreg#citation or the arXiv version from "Export Bibtex Citation" on the right, instead of the "2019 OpenReview" version in Google Scholar. Thanks!
Discovering Non-monotonic Autoregressive Orderings with Variational Inference
Comments updated from ICLR 2021, first two authors contributed equally
Early Convolutions Help Transformers See Better
Comments NeurIPS 2021
