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
Yung-Sung Chuang, Yujia Xie, Hongyin Luo +3 authors
A decoding strategy that contrasts logits from different transformer layers reduces hallucinations in large language models without external knowledge or fine-tuning.
Hanoona Rasheed, Muhammad Maaz, Sahal Shaji +7 authors
GLaMM is a multimodal model that generates visually grounded language responses with object segmentation masks from both text and optional visual prompts.
Quan Sun, Yufeng Cui, Xiaosong Zhang +8 authors
A large-scale generative multimodal model with 37 billion parameters demonstrates strong few-shot in-context learning and achieves state-of-the-art performance on multimodal tasks through scaling-up and instruction tuning.
Jessy Lin, Yuqing Du, Olivia Watkins +4 authors
Dynalang, a multimodal agent, learns to predict future text and image representations using language hints to improve task performance and enrich its understanding.
Aohan Zeng, Mingdao Liu, Rui Lu +4 authors
AgentTuning enhances the agent capabilities of open LLMs through instruction-tuning without compromising their general abilities, demonstrating performance comparable to commercial models on unseen tasks.
Jie Huang, Xinyun Chen, Swaroop Mishra +4 authors
This paper evaluates the intrinsic self-correction capabilities of Large Language Models and identifies limitations in their ability to improve responses without external feedback.
Meredith Ringel Morris, Jascha Sohl-dickstein, Noah Fiedel +5 authors
The framework proposes a classification system for AGI capabilities and behavior using levels of performance and generality, with principles to assess risks and measure progress.
Xu Zhao, Wenchao Ding, Yongqi An +5 authors
A speed-up method using a CNN detector with an instance segmentation branch achieves comparable performance to SAM with 50 times faster runtime by training on a small subset of SAM's dataset.
Zhiliang Peng, Wenhui Wang, Li Dong +4 authors
Kosmos-2, a Multimodal Large Language Model, combines text-to-visual grounding with existing MLLM capabilities, enabling tasks like referring expression comprehension and phrase grounding.
Chong Mou, Xintao Wang, Jiechong Song +2 authors
DragonDiffusion, a novel image editing method, enables precise manipulation of generated or real images using classifier guidance and multi-scale attention mechanisms without requiring fine-tuning.
Lianghui Zhu, Xinggang Wang, Xinlong Wang
Large Language Models fine-tuned as scalable judges (JudgeLM) achieve state-of-the-art performance in evaluating open-ended benchmarks through a comprehensive dataset and benchmark, enhancing judgment efficiency and accuracy.
Yew Ken Chia, Guizhen Chen, Luu Anh Tuan +2 authors
Contrastive chain of thought, utilizing both valid and invalid reasoning examples, improves language model reasoning and generalization compared to conventional methods.
Yilin Zhao, Xinbin Yuan, Shanghua Gao +4 authors
A framework for generating anthropomorphized personas with diverse voices and appearances from text descriptions using LLMs and generative models, with improved face landmark detection for automatic animation.
Pei Zhou, Aman Madaan, Srividya Pranavi Potharaju +9 authors
LLMs perform well in inferring beliefs but struggle translating these into actions; Foresee and Reflect (FaR) improves their action guidance in social scenarios.
Yangyi Huang, Hongwei Yi, Yuliang Xiu +4 authors
TeCH reconstructs high-fidelity 3D human models from single images using descriptive text prompts, a personalized Text-to-Image diffusion model, and a hybrid DMTet representation, outperforming existing methods.
Han Lin, Abhay Zala, Jaemin Cho +1 authors
VideoDirectorGPT combines LLMs for video content planning with a layout-controlled video generator, enabling temporally consistent multi-scene video generation from text prompts.
Peng Wang, Yichun Shi
ImageDream is a multi-view diffusion model for 3D object generation that uses image prompts to produce high-quality 3D models with improved visual geometry accuracy.
Raymond Li, Loubna Ben Allal, Yangtian Zi +64 authors
StarCoder, a 15.5B parameter LLM trained on 1 trillion tokens, outperforms other open Code LLMs across multiple languages and fine-tuned Python, with safety enhancements and publicly available under the Open Responsible AI Model license.
Chengshu Li, Jacky Liang, Andy Zeng +7 authors
Chain of Code (CoC) enhances language models' reasoning capabilities by using pseudocode and selective code emulation, enabling them to tackle complex semantic and logical tasks more effectively.
Griffin Adams, Alexander Fabbri, Faisal Ladhak +2 authors
CoD prompt in GPT-4 improves summary density, informativeness, and human preference over vanilla prompts, with tradeoffs between readability and detail.
Bo Li, Peiyuan Zhang, Jingkang Yang +3 authors
OtterHD-8B, an advanced multimodal model from Fuyu-8B, excels in processing high-resolution inputs and discerning detailed spatial relationships through MagnifierBench, an evaluation framework highlighting the importance of vision encoder flexibility.
Telmo Pessoa Pires, António V. Lopes, Yannick Assogba +1 authors
The study shows that the Feed Forward Network in Transformers is highly redundant and its reduction or sharing within the model can lead to improved accuracy and latency without significant loss of performance.
Anas Awadalla, Irena Gao, Josh Gardner +13 authors
OpenFlamingo, an open-source family of autoregressive vision-language models, replicates DeepMind's Flamingo performance across multiple datasets.
Tear Gosling, Alpin Dale, Yinhe Zheng
PIPPA, a community-driven partially-synthetic dataset, provides over a million utterances across numerous conversation sessions to enhance role-play and conversational AI research and development.
Xingchao Liu, Xiwen Zhang, Jianzhu Ma +2 authors
Rectified Flow is used to develop an ultra-fast one-step text-to-image generator named InstaFlow, achieving high image quality with significantly reduced inference time compared to existing methods.
Yu Shu, Siwei Dong, Guangyao Chen +5 authors
LLaSM, an end-to-end trained large multi-modal speech-language model, enhances human interaction with AI by following speech-and-language instructions using cross-modal conversational abilities.
Qihang Fan, Huaibo Huang, Mingrui Chen +2 authors
The proposed RMT model, combining RetNet and Transformer architectures, introduces explicit spatial distance priors and coordinate-wise decomposition to achieve exceptional performance in computer vision tasks.
Shiwei Zhang, Jiayu Wang, Yingya Zhang +6 authors
The cascaded I2VGen-XL approach uses hierarchical encoders and text-image pairs to improve the semantic accuracy, continuity, and clarity of generated videos.
Ziyang Luo, Can Xu, Pu Zhao +7 authors
WizardCoder, a Code LLM fine-tuned with complex instructions using Evol-Instruct, outperforms other open-source and closed LLMs on several code generation benchmarks.
Hadi Alzayer, Kevin Zhang, Brandon Feng +2 authors
A method for reconstructing 3D scenes beyond a camera's line of sight using eye reflections is proposed, refining cornea poses, radiance fields, and iris textures.
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