What research can be pursued with small models trained to complete true programs? Typically, researchers study program synthesis via large language models (LLMs) which introduce issues such as knowing what is in or out of distribution, understanding fine-tuning effects, understanding the effects of tokenization, and higher demand on compute and storage to carry out experiments. We present a system called Cadmus which includes an integer virtual machine (VM), a dataset composed of true programs of diverse tasks, and an autoregressive transformer model that is trained for under \$200 of compute cost. The system can be used to study program completion, out-of-distribution representations, inductive reasoning, and instruction following in a setting where researchers have effective and affordable fine-grained control of the training distribution and the ability to inspect and instrument models. Smaller models working on complex reasoning tasks enable instrumentation and investigations that may be prohibitively expensive on larger models. To demonstrate that these tasks are complex enough to be of interest, we show that these Cadmus models outperform GPT-5 (by achieving 100\% accuracy while GPT-5 has 95\% accuracy) even on a simple task of completing correct, integer arithmetic programs in our domain-specific language (DSL) while providing transparency into the dataset's relationship to the problem. We also show that GPT-5 brings unknown priors into its reasoning process when solving the same tasks, demonstrating a confounding factor that prevents the use of large-scale LLMs for some investigations where the training set relationship to the task needs to be fully understood.
LLaMA-Omni2: LLM-based Real-time Spoken Chatbot with Autoregressive Streaming Speech Synthesis
Key Laboratory of Intelligent Information Processing Institute of Computing Technology, Chinese Academy of Sciences (ICT/CAS)(智能信息处理重点实验室,计算技术研究所,中国科学院);Key Laboratory of AI Safety, Chinese Academy of Sciences(人工智能安全重点实验室,中国科学院);University of Chinese Academy of Sciences, Beijing, China(中国科学院大学,北京,中国)
LLM-guided phase diagram construction through high-throughput experimentation
Center for Basic Research on Materials(材料基础研究中心);National Institute for Materials Science(国家材料科学研究所);Graduate School of Frontier Sciences(前沿科学研究生院);The University of Tokyo(东京大学);RIKEN Center for Advanced Intelligence Project(RIKEN高级智能项目中心);Core Facility Center(核心设施中心);Tohoku University(东北大学);Univ. Grenoble Alpes(格勒诺布尔阿尔卑斯大学);CNRS(法国国家科学研究中心);Grenoble INP(格勒诺布尔INP);SIMaP;Center for Green Research on Energy and Environmental Materials(能源与环境材料绿色研究中心);Nara Institute of Science and Technology(奈良科学技术大学);Research Center for Structural Materials(结构材料研究中心);Graduate School of Medical Life Science(医学生命科学研究生院);Tokyo Institute of Technology(东京技术大学)
Cornell University(康奈尔大学);University of Bern(伯尔尼大学);California Institute of Technology(加州理工学院)
人工智能(AI)实现的自主实验为加速科学发现提供了新的范式。非平衡材料合成是复杂的资源密集型实验的象征,其加速将成为材料发现和开发的分水岭。非平衡合成相图的绘制最近已经通过高通量实验加速,但仍然限制了材料研究,因为参数空间太大而无法穷尽地探索。我们展示了加速合成和探索的亚稳态材料,通过分层自治实验的科学自主推理代理(SARA)。SARA集成了机器人材料合成和表征以及AI方法的层次结构,有效地揭示了加工相图的结构。SARA设计了用于平行材料合成的横向梯度激光尖峰退火(lg-LSA)实验,并采用光学光谱法快速识别相变。多维参数空间的有效探索是通过建立在先进机器学习模型基础上的嵌套主动学习(AL)循环实现的,该模型结合了实验的基础物理学以及端到端的不确定性量化。有了这一点,以及人工智能在多个尺度上的协调,SARA体现了人工智能对复杂科学任务的利用。我们证明了其性能,通过自主映射合成相边界的Bi $_2$ O $_3$系统,导致订单的数量级加速建立的合成相图,其中包括动力学稳定的条件$δ$-Bi $_2$ O $_3$在室温下,一个关键的发展,如固体氧化物燃料电池的电化学技术。
T-CACE: A Time-Conditioned Autoregressive Contrast Enhancement Multi-Task Framework for Contrast-Free Liver MRI Synthesis, Segmentation, and Diagnosis
Department of Computer Science, Toronto Metropolitan University(计算机科学系,多伦多 Metropolitan 大学);School of Biomedical Engineering, Western University(生物医学工程学院,西部大学)