Trending Research
Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving
In autonomous driving, predicting future events in advance and evaluating the foreseeable risks empowers autonomous vehicles to better plan their actions, enhancing safety and efficiency on the road.
Adversarial Diffusion Distillation
stability-ai/generative-models
•
•
We introduce Adversarial Diffusion Distillation (ADD), a novel training approach that efficiently samples large-scale foundational image diffusion models in just 1-4 steps while maintaining high image quality.
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
Based on this observation, we further sparsify delta parameters of multiple SFT homologous models with DARE and subsequently merge them into a single model by parameter averaging.
LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching
envision-research/luciddreamer
•
•
The recent advancements in text-to-3D generation mark a significant milestone in generative models, unlocking new possibilities for creating imaginative 3D assets across various real-world scenarios.
Pair then Relation: Pair-Net for Panoptic Scene Graph Generation
king159/pair-net
•
•
Panoptic Scene Graph (PSG) is a challenging task in Scene Graph Generation (SGG) that aims to create a more comprehensive scene graph representation using panoptic segmentation instead of boxes.
SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis
A lifelike talking head requires synchronized coordination of subject identity, lip movements, facial expressions, and head poses.
Sketch Video Synthesis
Understanding semantic intricacies and high-level concepts is essential in image sketch generation, and this challenge becomes even more formidable when applied to the domain of videos.
White-Box Transformers via Sparse Rate Reduction
ma-lab-berkeley/crate
•
•
NeurIPS 2023
Particularly, we show that the standard transformer block can be derived from alternating optimization on complementary parts of this objective: the multi-head self-attention operator can be viewed as a gradient descent step to compress the token sets by minimizing their lossy coding rate, and the subsequent multi-layer perceptron can be viewed as attempting to sparsify the representation of the tokens.
LoRA: Low-Rank Adaptation of Large Language Models
longyuewangdcu/chinese-llama-2
•
•
ICLR 2022
We propose Low-Rank Adaptation, or LoRA, which freezes the pre-trained model weights and injects trainable rank decomposition matrices into each layer of the Transformer architecture, greatly reducing the number of trainable parameters for downstream tasks.
Video-Bench: A Comprehensive Benchmark and Toolkit for Evaluating Video-based Large Language Models
Video-based large language models (Video-LLMs) have been recently introduced, targeting both fundamental improvements in perception and comprehension, and a diverse range of user inquiries.
