Dichotomy of Early and Late Phase Implicit Biases Can Provably Induce Grokking
Comments Published as a conference paper at ICLR 2024; 40 pages, 4 figures
期刊&会议
ICLR
International Conference on Learning Representations · 会议 · Machine Learning
Rephrase, Augment, Reason: Visual Grounding of Questions for Vision-Language Models
Comments ICLR 2024 camera-ready (23 pages), Code: https://github.com/archiki/RepARe
TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting
Comments Accepted by ICLR 2024. Camera Ready Version
FasterViT: Fast Vision Transformers with Hierarchical Attention
Comments ICLR'24 Accepted Paper
Low Rank Matrix Completion via Robust Alternating Minimization in Nearly Linear Time
Comments ICLR 2024
Entity-Centric Reinforcement Learning for Object Manipulation from Pixels
Comments ICLR 2024 Spotlight. Videos and code are available on the project website: https://sites.google.com/view/entity-centric-rl
Copilot4D: Learning Unsupervised World Models for Autonomous Driving via Discrete Diffusion
Comments ICLR 2024
Solving Diffusion ODEs with Optimal Boundary Conditions for Better Image Super-Resolution
Comments Accepted by ICLR 2024
Lipsum-FT: Robust Fine-Tuning of Zero-Shot Models Using Random Text Guidance
Comments ICLR 2024
On Difficulties of Attention Factorization through Shared Memory
Comments 2 pages of main content, 8 pages in total, published as a Tiny Paper at ICLR 2024
Observations on Building RAG Systems for Technical Documents
Comments Published as a Tiny Paper at ICLR 2024
HypeBoy: Generative Self-Supervised Representation Learning on Hypergraphs
Comments Published as a conference paper at ICLR 2024
Parameter and Data-Efficient Spectral StyleDCGAN
Comments Notable ICLR Tiny Paper 2024
Thin-Shell Object Manipulations With Differentiable Physics Simulations
Comments ICLR 2024
C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersion
Comments ICLR 2024
An Extensible Framework for Open Heterogeneous Collaborative Perception
Comments Accepted by ICLR 2024. The code and data are open-sourced at https://github.com/yifanlu0227/HEAL
Ito Diffusion Approximation of Universal Ito Chains for Sampling, Optimization and Boosting
Comments Appears in: The Twelfth International Conference on Learning Representations (ICLR 2024). 27 pages, 3 tables. Reference: https://openreview.net/forum?id=fjpfCOV4ru
Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
Comments International Conference on Learning Representations
VDC: Versatile Data Cleanser based on Visual-Linguistic Inconsistency by Multimodal Large Language Models
Comments Accepted to ICLR 2024
AntGPT: Can Large Language Models Help Long-term Action Anticipation from Videos?
Comments ICLR 2024 Camera Ready
G-PECNet: Towards a Generalizable Pedestrian Trajectory Prediction System
Comments Notable ICLR Tiny Paper 2024
LUT-GEMM: Quantized Matrix Multiplication based on LUTs for Efficient Inference in Large-Scale Generative Language Models
Comments ICLR 2024
Embed to Control Partially Observed Systems: Representation Learning with Provable Sample Efficiency
Comments Accepted by ICLR 2022
Negative Label Guided OOD Detection with Pretrained Vision-Language Models
Comments ICLR 2024 Spotlight
Adverb Is the Key: Simple Text Data Augmentation with Adverb Deletion
Comments ICLR 2024 Tiny Papers
Colorful Cutout: Enhancing Image Data Augmentation with Curriculum Learning
Comments ICLR 2024 Tiny Papers
Efficient Modulation for Vision Networks
Comments Accepted by ICLR 2024. Codes are made publically available at https://github.com/ma-xu/EfficientMod
STREAM: Spatio-TempoRal Evaluation and Analysis Metric for Video Generative Models
Comments Our work is accepted to ICLR 2024
Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation
Comments ICLR 2024
SineNet: Learning Temporal Dynamics in Time-Dependent Partial Differential Equations
Comments The Twelfth International Conference on Learning Representations
