Explore · 每周精选
发现最受关注的研究论文,追踪研究趋势,订阅感兴趣的期刊与关键词。
50 篇论文 · 按点赞排序
Chenglei Si, Diyi Yang, Tatsunori Hashimoto
LLM-generated research ideas are perceived as more novel than those from human experts but are deemed slightly less feasible, based on blind evaluations by NLP researchers.
Gongfan Fang, Hongxu Yin, Saurav Muralidharan +5 authors
MaskLLM introduces a learnable pruning method using Gumbel Softmax sampling to achieve N:M sparsity in LLMs, improving inference efficiency and transferability across tasks.
Yi Lu, Jing Nathan Yan, Songlin Yang +6 authors
A controlled evaluation protocol for long-context models reveals that exact fine-tuning methods are effective, while approximate attention methods underperform, and extrapolation remains challenging.
Xingyi Yang, Xinchao Wang
The Kolmogorov-Arnold Transformer replaces MLP layers with Kolmogorov-Arnold Network layers to enhance transformers, overcoming challenges related to inference speed, computation efficiency, and weight initialization through rational basis, group learning, and variance-preserving techniques.
Liqun Yang, Jian Yang, Chaoren Wei +13 authors
A framework uses fine-tuned large language models to guide input mutation in fuzzing, improving the detection of software vulnerabilities.
Di Liu, Meng Chen, Baotong Lu +11 authors
RetrievalAttention is a training-free method that accelerates attention computation in Transformer-based LLMs by using attention-aware vector search to reduce retrieval costs and GPU memory usage.
Zhiyuan Hu, Yuliang Liu, Jinman Zhao +8 authors
LongRecipe extends the context window of large language models efficiently with reduced resources and preserved performance, achieving a 16-fold increase from 8k to 128k context size.
Bin Wang, Chao Xu, Xiaomeng Zhao +15 authors
MinerU is an open-source tool that enhances document content extraction using fine-tuned models and pre/postprocessing rules across diverse document types.
Haoran Que, Feiyu Duan, Liqun He +11 authors
A comprehensive benchmark, HelloBench, and evaluation method, HelloEval, are introduced to assess the long text generation capabilities of Large Language Models, revealing limitations and providing human-aligned evaluations.
Kai Chen, Yunhao Gou, Runhui Huang +27 authors
EMOVA integrates speech, vision, and language capabilities, achieving top performance in benchmarks and enabling emotive spoken dialogue using a disentangled speech tokenizer and style module.
Zayne Sprague, Fangcong Yin, Juan Diego Rodriguez +7 authors
Chain-of-thought prompting enhances large language model performance primarily in math or logic tasks by improving symbolic reasoning, but it is less beneficial in other domains and can be selectively applied to save inference costs.
Mouxiang Chen, Lefei Shen, Zhuo Li +3 authors
The VisionTS model, leveraging a visual masked autoencoder pre-trained on ImageNet, achieves state-of-the-art zero-shot and fine-tuned forecasting performance in time series forecasting without adapting specifically to the time-series domain.
Yunfei Xie, Juncheng Wu, Haoqin Tu +6 authors
OpenAI's o1 demonstrates strong performance in medical tasks but shows weaknesses in reasoning, multilinguality, and evaluation metrics compared to GPT-4.
Dongzhi Jiang, Renrui Zhang, Ziyu Guo +10 authors
MMSearch-Engine enables multimodal search capabilities in Large Multimodal Models (LMMs) using a comprehensive benchmark and demonstrates improved performance over commercial products.
Wenbo Hu, Xiangjun Gao, Xiaoyu Li +5 authors
DepthCrafter generates temporally consistent depth sequences for open-world videos using a video-to-depth model derived from a pre-trained image-to-video diffusion model.
Sihang Li, Jin Huang, Jiaxi Zhuang +7 authors
A hybrid approach combining continual pre-training and supervised fine-tuning creates specialized LLMs for scientific literature, addressing challenges in constructing corpora and generating instructions.
Saba Sturua, Isabelle Mohr, Mohammad Kalim Akram +9 authors
jina-embeddings-v3, a large-scale text embedding model, achieves state-of-the-art performance in multilingual and long-context retrieval tasks using Low-Rank Adaptation and Matryoshka Representation Learning.
Jiongnan Liu, Yutao Zhu, Shuting Wang +6 authors
A novel model constructs user-specific embeddings using a lightweight plug-in module to personalize LLM outputs without fine-tuning, improving performance on various tasks.
Joonghyuk Shin, Daehyeon Choi, Jaesik Park
InstantDrag, an optimization-free pipeline with two networks, performs fast, photo-realistic drag-based image editing without requiring masks or text prompts.
Yejie Wang, Keqing He, Dayuan Fu +11 authors
XCoder, a family of models fine-tuned from LLaMA3, achieves state-of-the-art performance on code instruction tasks using a novel data pruning strategy that addresses data leakage and quality in training datasets.
北京市昌平区探索星信息技术及软件开发工作室
京ICP备2026059466号