Scaling up language models to billions of parameters has opened up possibilities for in-
context learning, allowing instruction tuning and few-shot learning on tasks that the model …

Large language models (LLMs) are restricted to reason in the" language space", where they
typically express the reasoning process with a chain-of-thought (CoT) to solve a complex …

In human cognition theory, human thinking is governed by two systems: the fast and intuitive
System 1 and the slower but more deliberative System 2. Recent studies have shown that …

Current long context large language models (LLMs) can process inputs up to 100,000
tokens, yet struggle to generate outputs exceeding even a modest length of 2,000 words …

LLMs are typically trained to answer user questions or follow instructions similarly to how
human experts respond. However, in the standard alignment framework they lack the basic …

Numina is an open AI4Maths initiative dedicated to advancing both artificial and human
intelligence in the field of mathematics. In this paper, we present the NuminaMath dataset, a …

Large Language Models (LLMs) have emerged as transformative tools in artificial
intelligence (AI), exhibiting remarkable capabilities across diverse tasks such as text …

Large Language Models (LLMs) have demonstrated strong performance in handling
complex tasks requiring both extensive knowledge and reasoning abilities. However, the …

While vision-language models (VLMs) have demonstrated remarkable performance across
various tasks combining textual and visual information, they continue to struggle with fine …

Large language models (LLMs) demonstrate exceptional capabilities, yet still face the
hallucination issue. Typical text generation approaches adopt an auto-regressive generation …