Trending Research
PrefixQuant: Static Quantization Beats Dynamic through Prefixed Outliers in LLMs
Specifically, PrefixQuant identifies high-frequency outlier tokens and prefixes them in the KV cache, preventing the generation of outlier tokens during inference and simplifying quantization.
PuLID: Pure and Lightning ID Customization via Contrastive Alignment
We propose Pure and Lightning ID customization (PuLID), a novel tuning-free ID customization method for text-to-image generation.
ControlAR: Controllable Image Generation with Autoregressive Models
Firstly, we explore control encoding for AR models and propose a lightweight control encoder to transform spatial inputs (e. g., canny edges or depth maps) into control tokens.
MOSEL: 950,000 Hours of Speech Data for Open-Source Speech Foundation Model Training on EU Languages
hlt-mt/mosel
•
•
The rise of foundation models (FMs), coupled with regulatory efforts addressing their risks and impacts, has sparked significant interest in open-source models.
Grounded-VideoLLM: Sharpening Fine-grained Temporal Grounding in Video Large Language Models
whb139426/grounded-video-llm
•
•
Video Large Language Models (Video-LLMs) have demonstrated remarkable capabilities in coarse-grained video understanding, however, they struggle with fine-grained temporal grounding.
Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
qwenlm/qwen2-vl
•
•
We present the Qwen2-VL Series, an advanced upgrade of the previous Qwen-VL models that redefines the conventional predetermined-resolution approach in visual processing.
Ranked #3 on
Temporal Relation Extraction
on Vinoground
EFM3D: A Benchmark for Measuring Progress Towards 3D Egocentric Foundation Models
The advent of wearable computers enables a new source of context for AI that is embedded in egocentric sensor data.
Ranked #1 on
3D Reconstruction
on Aria Synthetic Environments
AWT: Transferring Vision-Language Models via Augmentation, Weighting, and Transportation
MCG-NJU/AWT
•
•
Pre-trained vision-language models (VLMs) have shown impressive results in various visual classification tasks.
A Spark of Vision-Language Intelligence: 2-Dimensional Autoregressive Transformer for Efficient Finegrained Image Generation
chenllliang/dnd-transformer
•
•
This work tackles the information loss bottleneck of vector-quantization (VQ) autoregressive image generation by introducing a novel model architecture called the 2-Dimensional Autoregression (DnD) Transformer.
Fast Inference from Transformers via Speculative Decoding
Inference from large autoregressive models like Transformers is slow - decoding K tokens takes K serial runs of the model.
