相关论文: On the Origin of Algorithmic Progress in AI

The accelerating development and deployment of AI technologies depend on the continued ability to scale their infrastructure. This has implied increasing amounts of monetary investment and natural resources. Frontier AI applications have…

We investigate large language model performance across five orders of magnitude of compute scaling in eleven recent model architectures. We show that average benchmark performance, aggregating over many individual tasks and evaluations as…

As AI model size grows, neural scaling laws have become a crucial tool to predict the improvements of large models when increasing capacity and the size of original (human or natural) training data. Yet, the widespread use of popular models…

Understanding how humans evaluate robot behavior during human-robot interactions is crucial for developing socially aware robots that behave according to human expectations. While the traditional approach to capturing these evaluations is…

Driven by recent advances in artificial intelligence (AI), a growing literature has demonstrated the potential for using large language models (LLMs) as scalable surrogates to generate human-like responses in many business applications. Two…

We study the effects of approximate inference on the performance of Thompson sampling in the $k$-armed bandit problems. Thompson sampling is a successful algorithm for online decision-making but requires posterior inference, which often…

Rapid advancements in large language models (LLMs) have the potential to assist in scientific progress. A critical capability toward this endeavor is the ability to reproduce existing work. To evaluate the ability of AI agents to reproduce…

The substantial increase in AI model training has considerable environmental implications, mandating more energy-efficient and sustainable AI practices. On the one hand, data-centric approaches show great potential towards training…

After a proof of concept using Dropbox(tm), a free storage and synchronization service, showed that an evolutionary algorithm using several dissimilar computers connected via WiFi or Ethernet had a good scaling behavior in terms of…

The fields of machine learning and mathematical optimization increasingly intertwined. The special topic on supervised learning and convex optimization examines this interplay. The training part of most supervised learning algorithms can…

How software developers interact with Artificial Intelligence (AI)-powered tools, including Large Language Models (LLMs), plays a vital role in how these AI-powered tools impact them. While overreliance on AI may lead to long-term negative…

Power-law scaling indicates that large-scale training with uniform sampling is prohibitively slow. Active learning methods aim to increase data efficiency by prioritizing learning on the most relevant examples. Despite their appeal, these…

Test-time scaling (TTS) -- the dynamic allocation of compute during inference -- is a promising direction for improving reasoning in large language models (LLMs). However, a systematic comparison of well-known TTS strategies under identical…

Recent developments in sequential experimental design look to construct a policy that can efficiently navigate the design space, in a way that maximises the expected information gain. Whilst there is work on achieving tractable policies for…

Research on scaling large language models (LLMs) has primarily focused on model parameters and training data size, overlooking the role of vocabulary size. We investigate how vocabulary size impacts LLM scaling laws by training models…

Augmented Large Language Models (LLMs) enhance the capabilities of standalone LLMs by integrating external data sources through API calls. In interactive LLM applications, efficient scheduling is crucial for maintaining low request…

Post-training alignment has produced dozens of competing algorithms -- DPO, SimPO, KTO, GRPO, and others -- yet practitioners lack controlled comparisons to guide algorithm selection. We present OXRL, a unified framework implementing 51…

While scaling laws for Large Language Models (LLMs) traditionally focus on proxy metrics like pretraining loss, predicting downstream task performance has been considered unreliable. This paper challenges that view by proposing a direct…

Large language models (LLMs) excel at diverse tasks, but their deployment on resource-constrained devices remains challenging. Existing methods like quantization, pruning, and distillation can reduce memory footprint but often demand…

As Large Language Models (LLMs) move from curated training sets into open-ended real-world environments, a fundamental limitation emerges: static training cannot keep pace with continual deployment environment change. Scaling training-time…