京东 11.11 红包
Farzi Data: Autoregressive Data Distillation
发布人
打开封面 下载高清视频 观看高清视频 视频下载器
...
Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling
...
Ensemble-Instruct: Generating Instruction-Tuning Data with a Heterogeneous Mixtu
...
Gen4Gen: Generative Data Pipeline for Generative Multi-Concept Composition
...
Ziya2: Data-centric Learning is All LLMs Need
...
Conditional Diffusion Distillation
...
UT5: Pretraining Non autoregressive T5 with unrolled denoising
...
VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Mode
...
GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting
...
FreeControl：实现任意文本到图像扩散模型的无训练空间控制
...
Transformers are Multi-State RNNs
...
基于3D高斯展开的实时少样本视角合成
...
Toward Joint Language Modeling for Speech Units and Text
...
VeRA: Vector-based Random Matrix Adaptation
...
P5: Plug-and-Play Persona Prompting for Personalized Response Selection
...
DeepCache：无需训练加速扩散模型的创新方法
...
Domain Generalization Guided by Gradient Signal to Noise Ratio of Parameters
...
无需相机参数的3D高斯散射：COLMAP-Free 3DGS实现稳健的视角合成与姿态估计
...
DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior
...
单张图像到3D的高效生成：基于分摊生成的3D高斯模型
...
CrossCodeEval: A Diverse and Multilingual Benchmark for Cross-File Code Completi
...
EvoPrompt: AI Prompt Optimizer
...
EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision
...
Improving Summarization with Human Edits
...
Interactive Task Planning with Language Models
...
CLIP作为RNN：无需训练即可分割无数视觉概念
...
TiC-CLIP: Continual Training of CLIP Models
...
稳定分数蒸馏：高质量三维生成新方法
...
大型语言模型在上下文学习中的可靠性提升：结合监督知识的方法
...
Wonder3D: Single Image to 3D using Cross-Domain Diffusion
...
System 2 Attention (is something you might need too)
...
Retrieval meets Long Context Large Language Models
...
ReFT: Reasoning with Reinforced Fine-Tuning
...
综述：大模型的可解释性研究
...
CLEX: Continuous Length Extrapolation for Large Language Models
...
ShortGPT: Layers in Large Language Models are More Redundant Than You Expect
...
Drivable 3D Gaussian Avatars
...
TrustLLM: Trustworthiness in Large Language Models
...
Contrastive Chain-of-Thought Prompting
...
SOTOPIA-$π$: Interactive Learning of Socially Intelligent Language Agents
...
FlowVid：利用不完美光流实现一致性视频到视频合成