Related papers: Inference Scaling Reshapes AI Governance
Today, AI is being increasingly used to help human experts make decisions in high-stakes scenarios. In these scenarios, full automation is often undesirable, not only due to the significance of the outcome, but also because human experts…
Skill routing is an important component in large-scale conversational systems. In contrast to traditional rule-based skill routing, state-of-the-art systems use a model-based approach to enable natural conversations. To provide supervision…
Scaling up language models has been empirically shown to improve performance on a wide range of downstream tasks. However, if we were to observe worse performance as a function of scale ("inverse scaling") on certain tasks, this would…
A recent trend in deep learning algorithms has been towards training large scale models, having high parameter count and trained on big dataset. However, robustness of such large scale models towards real-world settings is still a…
This research addresses the critical lack of comprehensive studies on feature scaling by systematically evaluating 12 scaling techniques - including several less common transformations - across 14 different Machine Learning algorithms and…
Barroso's seminal contributions in energy-proportional warehouse-scale computing launched an era where modern datacenters have become more energy efficient and cost effective than ever before. At the same time, modern AI applications have…
Compute governance proposals often rely on the assumption that frontier AI training requires large, detectable computing clusters. However, recent advances in distributed training algorithms could allow developers to conduct frontier-scale…
Sampling-based search, a simple paradigm for utilizing test-time compute, involves generating multiple candidate responses and selecting the best one -- typically by having models self-verify each response for correctness. In this paper, we…
Rehearsal is one of the key techniques for mitigating catastrophic forgetting and has been widely adopted in continual learning algorithms due to its simplicity and practicality. However, the theoretical understanding of how rehearsal scale…
Benchmark datasets are crucial for evaluating approaches to scheduling or dispatching in the semiconductor industry during the development and deployment phases. However, commonly used benchmark datasets like the Minifab or SMT2020 lack the…
Inference-time scaling has emerged as a powerful way to improve large language model (LLM) performance by generating multiple candidate responses and selecting among them. However, existing work on dynamic allocation for test-time compute…
Modern applications increasingly rely on inference serving systems to provide low-latency insights with a diverse set of machine learning models. Existing systems often utilize resource elasticity to scale with demand. However, many…
Index structures are important for efficient data access, which have been widely used to improve the performance in many in-memory systems. Due to high in-memory overheads, traditional index structures become difficult to process the…
While artificial intelligence has the potential to process vast amounts of data, generate new insights, and unlock greater productivity, its widespread adoption may entail unforeseen consequences. We identify conditions under which AI, by…
Artificial intelligence is reshaping the organization and practice of research in ways that extend far beyond gains in productivity. AI systems now accelerate discovery, reorganize scholarly labour, and mediate access to expanding…
Deep learning recommendation models have grown to the terabyte scale. Traditional serving schemes--that load entire models to a single server--are unable to support this scale. One approach to support this scale is with distributed serving,…
Modeling user preferences has been mainly addressed by looking at users' interaction history with the different elements available in the system. Tailoring content to individual preferences based on historical data is the main goal of…
One of the most striking findings in modern research on large language models (LLMs) is that scaling up compute during training leads to better results. However, less attention has been given to the benefits of scaling compute during…
It is commonly believed that scaling language models should commit a significant space or time cost, by increasing the parameters (parameter scaling) or output tokens (inference-time scaling). We introduce the third and more…
The possibility of a rapid, "software-only" intelligence explosion brought on by AI's recursive self-improvement (RSI) is a subject of intense debate within the AI community. This paper presents an economic model and an empirical estimation…