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。
403 篇论文 · 按点赞排序
Clifton Poth, Hannah Sterz, Indraneil Paul +7 authors
Adapters is an open-source library designed for parameter-efficient and modular transfer learning in large language models, offering a unified interface for various adapter methods.
Xi Chen, Xiao Wang, Lucas Beyer +16 authors
PaLI-3, a smaller vision language model, outperforms larger models on multimodal tasks, particularly localization and text understanding, using SigLIP pretraining and achieves state-of-the-art results in multilingual cross-modal retrieval.
Jianwei Yang, Hao Zhang, Feng Li +3 authors
Set-of-Mark (SoM) enhances visual grounding in large multimodal models using segmented and marked images, demonstrating superior performance on fine-grained vision tasks without full fine-tuning.
Shengnan An, Zexiong Ma, Zeqi Lin +3 authors
LeMa, a learning-from-mistakes approach, enhances LLMs' mathematical reasoning by learning from inaccurate reasoning paths corrected by GPT-4, surpassing SOTA performance on math problems.
Ziyue Jiang, Jinglin Liu, Yi Ren +8 authors
Mega-TTS 2 extends zero-shot text-to-speech with arbitrary-length prompts by incorporating a multi-reference timbre encoder and a prosody language model, enhancing both identity preservation and prosodic control.
Kai Yang, Jian Tao, Jiafei Lyu +6 authors
D3PO method directly fine-tunes diffusion models using human feedback without requiring a separate reward model, reducing computational cost and improving image quality.
Bin Lin, Bin Zhu, Yang Ye +3 authors
Video-LLaVA is a unified large vision-language model that enhances performance across various image and video benchmarks by integrating visual representations into the language feature space.
Haoning Wu, Zicheng Zhang, Erli Zhang +11 authors
The Q-Instruct dataset, enhanced with diverse human feedback, improves low-level visual perception and understanding in multi-modality foundation models.
Hanjia Lyu, Song Jiang, Hanqing Zeng +2 authors
Combining diverse prompts and input augmentation techniques improves personalized content recommendation performance using large language models.
Dave Van Veen, Cara Van Uden, Louis Blankemeier +12 authors
Domain-adapted large language models outperform human experts in various clinical text summarization tasks, improving completeness, correctness, and aligning with physician preferences.
Haohe Liu, Ke Chen, Qiao Tian +2 authors
A diffusion-based generative model, AudioSR, achieves robust audio super-resolution across various audio types and bandwidths, enhancing generation quality for different audio models.
Junha Hyung, Jaeyo Shin, Jaegul Choo
MagiCapture personalizes text-to-image models by integrating subject and style concepts using limited references and an Attention Refocusing loss, enhancing realism and quality in portrait image generation.
Shangchen Zhou, Chongyi Li, Kelvin C. K. Chan +1 authors
ProPainter is a framework that improves video inpainting by combining dual-domain propagation and a mask-guided sparse video Transformer, outperforming existing methods in PSNR while maintaining efficiency.
Josef Dai, Xuehai Pan, Ruiyang Sun +5 authors
Safe RLHF, a novel algorithm for human value alignment in LLMs, improves performance and safety by decoupling human preferences and using a Lagrangian method to balance reward and cost constraints.
Lionel Wong, Gabriel Grand, Alexander K. Lew +4 authors
A framework combining large language models and probabilistic programs translates language into a symbolic language of thought for rational, commonsense reasoning across multiple cognitive domains.
Biao Jiang, Xin Chen, Wen Liu +3 authors
MotionGPT, a unified motion-language model using discrete vector quantization, achieves top performance in various motion-related tasks by treating motion as a language similar to text.
Jianqiang Ren, Chao He, Lin Liu +10 authors
The Make-A-Character (Mach) framework uses large language and vision models to create realistic, 3D avatars from text descriptions efficiently, enabling customization and seamless integration with existing CG pipelines.
Zengzhi Wang, Rui Xia, Pengfei Liu
A new large-scale, high-quality math-centric corpus named MathPile has been introduced to enhance mathematical reasoning abilities of language models.
Weihan Wang, Qingsong Lv, Wenmeng Yu +13 authors
CogVLM, a visual language foundation model, uses a trainable visual expert module to deeply integrate vision and language without compromising NLP performance.
Alexander Borzunov, Max Ryabinin, Artem Chumachenko +5 authors
A decentralized system pools idle compute resources to enable efficient inference and fine-tuning of large language models over a consumer-grade network, showcasing significant performance improvements.
Difei Gao, Lei Ji, Luowei Zhou +4 authors
A multi-modal AI assistant, AssistGPT, with a Plan, Execute, Inspect, and Learn (PEIL) approach integrates LLMs with various tools to handle complex visual-based tasks and achieves state-of-the-art results on benchmarks and beyond.
Chunting Zhou, Pengfei Liu, Puxin Xu +12 authors
A 65B parameter LLaMa language model trained with minimal instructional data matches or outperforms models with extensive human preference modeling in most cases, indicating that pretraining is predominantly responsible for knowledge acquisition.
Animesh Sinha, Bo Sun, Anmol Kalia +14 authors
Style Tailoring is a finetuning method for Latent Diffusion Models that enhances the generation of sticker images, improving visual quality, prompt alignment, and scene diversity.
Elias Frantar, Dan Alistarh
QMoE enables efficient execution of trillion-parameter MoE models on affordable hardware through compression and bespoke GPU decoding.
Fuxiao Liu, Tianrui Guan, Zongxia Li +4 authors
HallusionBench is a benchmark that highlights language hallucination and visual illusion issues in vision-language models (VLMs), showcasing the limitations of current state-of-the-art models like GPT-4V and LLaVA-1.5.
Shuyan Zhou, Frank F. Xu, Hao Zhu +8 authors
WebArena, a realistic and reproducible environment, evaluates the performance of autonomous agents performing complex tasks on websites using benchmark tasks that require human-like reasoning.
Waseem AlShikh, Manhal Daaboul, Kirk Goddard +4 authors
The Instruction Following Score (IFS) is introduced to evaluate and control instruction-following ability in language models, aiding in distinguishing between base and instruct models as well as serving as an early stopping criterion in training.
Miao Hua, Jiawei Liu, Fei Ding +3 authors
DreamTurner uses a novel approach by injecting reference information through subject encoders and self-subject-attention layers to enhance subject-driven image generation, balancing subject learning and model capabilities.
Xiaofei Wang, Manthan Thakker, Zhuo Chen +7 authors
SpeechX integrates neural codec language modeling and multi-task learning to enable versatile speech generation, transformation, and enhancement tasks in both clean and noisy conditions.
En Yu, Liang Zhao, Yana Wei +8 authors
A novel MLLM, Merlin, is introduced which incorporates future modeling through FPT and FIT, enhancing foresight and future reasoning capabilities.
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