本期的 15 篇论文如下:
[00:25] 🎨 DDT: Decoupled Diffusion Transformer(解耦扩散Transformer)
[01:05] 🎬 GenDoP: Auto-regressive Camera Trajectory Generation as a Director of Photography(GenDoP:基于自回归的相机轨迹生成,如同电影摄影师一般)
[01:49] 🔍 OLMoTrace: Tracing Language Model Outputs Back to Trillions of Training Tokens(OLMoTrace:将语言模型的输出追溯到数万亿的训练文本)
[02:28] 🖼 A Unified Agentic Framework for Evaluating Conditional Image Generation(用于评估条件图像生成的统一代理框架)
[03:11] 🤔 Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?(缺失前提加剧过度思考:推理模型是否正在丧失批判性思维能力?)
[03:57] 🗣 FantasyTalking: Realistic Talking Portrait Generation via Coherent Motion Synthesis(FantasyTalking:通过连贯运动合成生成逼真会说话的人像)
[04:34] 🧐 A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility(冷静看待语言模型推理的进展:陷阱与可复现性之路)
[05:15] 🖼 OmniCaptioner: One Captioner to Rule Them All(万能字幕器:一统天下的字幕生成器)
[05:57] 🧩 Are We Done with Object-Centric Learning?(以对象为中心的学习是否已经结束?)
[06:35] 🤖 Self-Steering Language Models(自导向语言模型)
[07:09] 🇷 RuOpinionNE-2024: Extraction of Opinion Tuples from Russian News Texts(RuOpinionNE-2024:从俄语新闻文本中提取观点元组)
[07:51] 🤖 Masked Scene Modeling: Narrowing the Gap Between Supervised and Self-Supervised Learning in 3D Scene Understanding(掩码场景建模:缩小3D场景理解中监督学习和自监督学习之间的差距)
[08:30] 👂 DiTaiListener: Controllable High Fidelity Listener Video Generation with Diffusion(DiTaiListener:基于扩散模型的可控高保真听者视频生成)
[09:05] 🤖 VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning(VideoChat-R1:通过强化微调增强时空感知能力)
[09:47] 🤖 WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments(WildGS-SLAM:动态环境下的单目高斯溅射SLAM)

【关注我们】
您还可以在以下平台找到我们,获得播客内容以外更多信息
小红书: AI速递

空空如也
暂无小宇宙热门评论