Trends · 研究趋势
数据来自 Hugging Face 论文的 AI 提取关键词,按月统计研究方向的增长与热度。
Shelly Bensal, Umar Jamil, Christopher Bryant +5 authors
A method using self-reflection and reinforcement learning improves the performance of large language models, especially with limited feedback, by rewarding self-reflections that lead to better task performance.
Mind Lab, Song Cao, Vic Cao +51 authors
Parameter-efficient fine-tuning can function as a compact substrate for persistent personal models by enabling small trainable adapters to store instance-specific behaviors on top of strong foundation models.
Suriya Gunasekar, Yi Zhang, Jyoti Aneja +16 authors
A new compact Transformer-based large language model for code, phi-1, achieves high accuracy on coding benchmarks despite having fewer parameters than competing models.
Kangsheng Duan, Ziyang Xu, Wenyu Liu +3 authors
A lightweight image inpainting framework achieves high-fidelity results with significantly reduced parameters and inference time through novel local-global interaction blocks and adaptive distillation strategies.
Yicheng Zou, Dongsheng Zhu, Lin Zhu +171 authors
Intern-S1-Pro is a one-trillion-parameter scientific multimodal foundation model that enhances general and scientific capabilities through advanced agent functionalities and specialized task mastery across multiple scientific disciplines.
Zhiyuan Liang, Dongwen Tang, Yuhao Zhou +11 authors
Drag-and-Drop LLMs generate task-specific parameters through prompt-conditioned parameter generation, achieving significant efficiency gains and cross-domain generalization without per-task training.
Sen Xu, Shixi Liu, Wei Wang +6 authors
VibeThinker-3B demonstrates that compact models can achieve state-of-the-art performance on verifiable reasoning tasks through specialized training techniques, challenging conventional scaling assumptions.
Wei Chen, Zhiyuan Li
The Octopus v4 model uses functional tokens to integrate and direct queries to task-specific open-source language models, achieving SOTA performance with models under 10B parameters.
David Bensaïd, Noam Rotstein, Roy Velich +2 authors
SingLoRA, a reformulated low-rank adaptation method, enhances parameter-efficient fine-tuning by learning a single low-rank matrix and its transpose, ensuring stable optimization and reducing parameter count.
Yuchao Gu, Guian Fang, Yuxin Jiang +4 authors
AnyFlow introduces a novel any-step video diffusion distillation framework that improves upon consistency distillation by optimizing full ODE sampling trajectories through flow-map transition learning and backward simulation techniques.
Zhengxuan Wu, Aryaman Arora, Zheng Wang +4 authors
Representation Finetuning (ReFT) methods, exemplified by Low-rank Linear Subspace ReFT (LoReFT), achieve high efficiency and performance by adapting representations in frozen base models, outperforming state-of-the-art Parameter-efficient Fine-tuning (PEFT) methods.
Liliana Hotsko, Yinxi Li, Yuntian Deng +1 authors
Code2LoRA is a hypernetwork framework that generates repository-specific LoRA adapters for code language models, supporting both static and evolving codebases with efficient parameter-efficient fine-tuning.
Eunbi Choi, Kibong Choi, Seokhee Hong +62 authors
K-EXAONE is a multilingual language model with a Mixture-of-Experts architecture that achieves competitive performance on various benchmarks while supporting multiple languages and long-context windows.
Yukang Chen, Shengju Qian, Haotian Tang +4 authors
LongLoRA extends LLM context sizes efficiently using sparse local attention and parameter-efficient fine-tuning, demonstrating strong results with minimal computational cost.
Nikhil Kandpal, Brian Lester, Colin Raffel +24 authors
The Common Pile v0.1 dataset, an 8-terabyte collection of openly licensed text, is used to train competitive 7 billion parameter LLMs.
Sharath Turuvekere Sreenivas, Saurav Muralidharan, Raviraj Joshi +6 authors
Models Llama 3.1 8B and Mistral NeMo 12B are compressed using pruning and distillation, achieving state-of-the-art performance in a reduced parameter setting.
Quy-Anh Dang, Chris Ngo
Reinforcement learning enhances reasoning in small language models with limited resources, demonstrating significant performance improvements in mathematical reasoning tasks.
Ting Jiang, Shaohan Huang, Shengyue Luo +8 authors
MoRA, a high-rank updating method using square matrices, enhances the ability of large language models to learn and memorize new knowledge, especially in memory-intensive tasks, compared to LoRA.
Zihan Wang, Deli Chen, Damai Dai +3 authors
Research on parameter-efficient fine-tuning for Large Language Models with Mixture-of-Experts architecture reveals that task-specific expert tuning can enhance performance and efficiency.
Xudong Lu, Yinghao Chen, Cheng Chen +19 authors
BlueLM-V-3B optimizes the deployment of multimodal large language models on mobile platforms through redesigned dynamic resolution and hardware-aware system optimizations, achieving high performance with a small model size and fast generation speed.
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