Order Matters in the Presence of Dataset Imbalance for Multilingual Learning
Paper
•
2312.06134
•
Published
•
2
LLM
-
-
Efficient Monotonic Multihead Attention
Paper • 2312.04515 • Published • 6 -
Contrastive Decoding Improves Reasoning in Large Language Models
Paper • 2309.09117 • Published • 37 -
Exploring Format Consistency for Instruction Tuning
Paper • 2307.15504 • Published • 7 -
Learning Universal Predictors
Paper • 2401.14953 • Published • 19 -
EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty
Paper • 2401.15077 • Published • 19 -
SliceGPT: Compress Large Language Models by Deleting Rows and Columns
Paper • 2401.15024 • Published • 69 -
Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities
Paper • 2401.14405 • Published • 12 -
Deconstructing Denoising Diffusion Models for Self-Supervised Learning
Paper • 2401.14404 • Published • 17 -
Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data
Paper • 2401.10891 • Published • 60 -
Time is Encoded in the Weights of Finetuned Language Models
Paper • 2312.13401 • Published • 20 -
Unsupervised Universal Image Segmentation
Paper • 2312.17243 • Published • 19 -
Reasons to Reject? Aligning Language Models with Judgments
Paper • 2312.14591 • Published • 17 -
Unlocking Pre-trained Image Backbones for Semantic Image Synthesis
Paper • 2312.13314 • Published • 7 -
Cached Transformers: Improving Transformers with Differentiable Memory Cache
Paper • 2312.12742 • Published • 12 -
In-Context Learning Creates Task Vectors
Paper • 2310.15916 • Published • 42 -
Controlled Decoding from Language Models
Paper • 2310.17022 • Published • 14 -
CapsFusion: Rethinking Image-Text Data at Scale
Paper • 2310.20550 • Published • 25 -
Tell Your Model Where to Attend: Post-hoc Attention Steering for LLMs
Paper • 2311.02262 • Published • 10 -
Memory Augmented Language Models through Mixture of Word Experts
Paper • 2311.10768 • Published • 16 -
SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding
Paper • 2310.15308 • Published • 22 -
An Image is Worth Multiple Words: Learning Object Level Concepts using Multi-Concept Prompt Learning
Paper • 2310.12274 • Published • 12 -
Language Modeling Is Compression
Paper • 2309.10668 • Published • 83 -
Finite Scalar Quantization: VQ-VAE Made Simple
Paper • 2309.15505 • Published • 21 -
Vision Transformers Need Registers
Paper • 2309.16588 • Published • 78 -
SLiMe: Segment Like Me
Paper • 2309.03179 • Published • 29 -
Gated recurrent neural networks discover attention
Paper • 2309.01775 • Published • 7 -
One Wide Feedforward is All You Need
Paper • 2309.01826 • Published • 31 -
Semantic-SAM: Segment and Recognize Anything at Any Granularity
Paper • 2307.04767 • Published • 21 -
Scaling MLPs: A Tale of Inductive Bias
Paper • 2306.13575 • Published • 14 -
MSViT: Dynamic Mixed-Scale Tokenization for Vision Transformers
Paper • 2307.02321 • Published • 7 -
CRAG -- Comprehensive RAG Benchmark
Paper • 2406.04744 • Published • 44