Towards Being Parameter-Efficient: A Stratified Sparsely Activated Transformer with Dynamic Capacity
π₯ Github: https://github.com/fe1ixxu/stratified_mixture_of_experts
β© Paper: https://arxiv.org/pdf/2305.02176v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/flores-200
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π3π₯2β€1
Hybrid and Collaborative Passage Reranking
π₯ Github: https://github.com/zmzhang2000/hybrank
β© Paper: https://arxiv.org/pdf/2305.09313v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/natural-questions
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
β€6π₯2π1
Dynamic Sparse Training with Structured Sparsity
π₯ Github: https://github.com/calgaryml/condensed-sparsity
β© Paper: https://arxiv.org/pdf/2305.02299v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/cifar-10
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π5β€4π1
Concurrent Misclassification and Out-of-Distribution Detection for Semantic Segmentation via Energy-Based Normalizing Flow
π₯ Github: https://github.com/gudovskiy/flowenedet
β© Paper: https://arxiv.org/pdf/2305.09610v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/coco
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯6π2β€1
Online Continual Learning Without the Storage Constraint
π₯ Github: https://github.com/drimpossible/acm
β© Paper: https://arxiv.org/pdf/2305.09253v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/yfcc100m
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π7π₯2β€1
Pre-Training to Learn in Context
π₯ Github: https://github.com/thu-coai/picl
β© Paper: https://arxiv.org/pdf/2305.09137v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/sst
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π6β€2π₯1
This media is not supported in your browser
VIEW IN TELEGRAM
DB-Football
π₯ Github: https://github.com/shanghai-digital-brain-laboratory/db-football
β© Paper: https://arxiv.org/pdf/2305.09458v1.pdf
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π7β€1π₯°1
LGTM
π₯ Github: https://github.com/twinkle0331/lgtm
β© Paper: https://arxiv.org/pdf/2305.09651v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/sst
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π6π₯2
SSSegmenation
π₯ Github: https://github.com/segmentationblwx/sssegmentation
β© Paper: https://arxiv.org/pdf/2305.17091v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/cityscapes
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π8β€1π₯1
Meta-prediction Model for Distillation-Aware NAS on Unseen Datasets
π₯ Github: https://github.com/cownowan/dass
β© Paper: https://arxiv.org/pdf/2305.16948v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/cub-200-2011
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π9β€1π₯1
Occ-BEV: Multi-Camera Unified Pre-training via 3D Scene Reconstruction
π₯ Github: https://github.com/chaytonmin/occ-bev
β© Paper: https://arxiv.org/pdf/2305.18829v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/imagenet
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π6π₯1π₯°1
Jax OffBench
π₯ Github: https://github.com/sail-sg/offbench
β© Paper: https://arxiv.org/pdf/2306.00972v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/d4rl
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π5β€1π₯1
TabEAE
π₯ Github: https://github.com/stardust-hyx/tabeae
β© Paper: https://arxiv.org/pdf/2306.00502v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/wikievents
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π₯7π2β€1
(Almost) Provable Error Bounds Under Distribution Shift via Disagreement Discrepancy
π₯ Github: https://github.com/erosenfeld/disagree_discrep
β© Paper: https://arxiv.org/pdf/2306.00312v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/cifar-10
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
β€8π2π₯1
BUOL: A Bottom-Up Framework with Occupancy-aware Lifting for Panoptic 3D Scene Reconstruction From A Single Image
π₯ Github: https://github.com/chtsy/buol
β© Paper: https://arxiv.org/pdf/2306.00965v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/matterport3d
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π6π₯°2π₯1
Asymmetric VQGAN
π₯ Github: https://github.com/buxiangzhiren/asymmetric_vqgan
β© Paper: https://arxiv.org/pdf/2306.04632v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/imagenet
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π9β€1π₯1
GP-UNIT - Official PyTorch Implementation
π₯ Github: https://github.com/williamyang1991/gp-unit
β© Paper: https://arxiv.org/pdf/2306.04636v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/celeba-hq
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π8π₯1π₯°1
Galactic: Scaling End-to-End Reinforcement Learning for Rearrangement
at 100k Steps-Per-Second
π₯ Github: https://github.com/facebookresearch/galactic
β© Paper: https://arxiv.org/pdf/2306.07552v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/vizdoom
@ArtificialIntelligencedl
at 100k Steps-Per-Second
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π8π₯1π₯°1
Semi-supervised learning made simple with self-supervised clustering [CVPR 2023]
π₯ Github: https://github.com/pietroastolfi/suave-daino
β© Paper: https://arxiv.org/pdf/2306.07483v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/imagenet
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π5β€4π₯1
Semi-supervised learning made simple with self-supervised clustering [CVPR 2023]
π₯ Github: https://github.com/Ruixinhua/ExplainableNRS
β© Paper: https://arxiv.org/pdf/2306.07506v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/mind
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π11β€2π₯1
LabelBench: A Comprehensive Framework for Benchmarking Label-Efficient Learning
π₯ Github: https://github.com/efficienttraining/labelbench
β© Paper: https://arxiv.org/pdf/2306.09910v1.pdf
π¨ Dataset: https://paperswithcode.com/dataset/cifar-10
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
π6β€2π₯°1