Trending Research
Data Engineering for Scaling Language Models to 128K Context
franxyao/long-context-data-engineering
•
•
We demonstrate that continual pretraining of the full model on 1B-5B tokens of such data is an effective and affordable strategy for scaling the context length of language models to 128K.
Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model
hustvl/vim
•
•
The results demonstrate that Vim is capable of overcoming the computation & memory constraints on performing Transformer-style understanding for high-resolution images and it has great potential to be the next-generation backbone for vision foundation models.
EasySpider: A No-Code Visual System for Crawling the Web
As such, web-crawling is an essential tool for both computational and non-computational scientists to conduct research.
An End-to-End Structure with Novel Position Mechanism and Improved EMD for Stock Forecasting
As a branch of time series forecasting, stock movement forecasting is one of the challenging problems for investors and researchers.
A Survey on Vision Mamba: Models, Applications and Challenges
ruixxxx/awesome-vision-mamba-models
•
•
To help keep pace with the rapid advancements in computer vision, this paper aims to provide a comprehensive review of visual Mamba approaches.
PLAID SHIRTTT for Large-Scale Streaming Dense Retrieval
hltcoe/colbert-x
•
•
PLAID, an efficient implementation of the ColBERT late interaction bi-encoder using pretrained language models for ranking, consistently achieves state-of-the-art performance in monolingual, cross-language, and multilingual retrieval.
GPU-based Private Information Retrieval for On-Device Machine Learning Inference
Together, for various on-device ML applications such as recommendation and language modeling, our system on a single V100 GPU can serve up to $100, 000$ queries per second -- a $>100 \times$ throughput improvement over a CPU-based baseline -- while maintaining model accuracy.
ESRL: Efficient Sampling-based Reinforcement Learning for Sequence Generation
hiyouga/llama-efficient-tuning
•
•
Applying Reinforcement Learning (RL) to sequence generation models enables the direct optimization of long-term rewards (\textit{e. g.,} BLEU and human feedback), but typically requires large-scale sampling over a space of action sequences.
MagicPose: Realistic Human Poses and Facial Expressions Retargeting with Identity-aware Diffusion
In this work, we propose MagicPose, a diffusion-based model for 2D human pose and facial expression retargeting.
FREB-TQA: A Fine-Grained Robustness Evaluation Benchmark for Table Question Answering
To investigate these aspects, we create and publish a novel TQA evaluation benchmark in English.
