A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and Trustworthiness

TL;DR

This survey maps the landscape of Small Language Models (SLMs) in the era of Large Language Models (LLMs), clarifying definitions and proposing a taxonomy of architectures, training methods, and deployment strategies. It details how SLMs can be built from scratch or obtained from LLMs via pruning, distillation, quantization, and low-rank techniques, and it covers enhancement methods such as instruction tuning and data-quality considerations. The paper also inventories a wide range of generic- and domain-specific SLMs, discusses their on-device applications, and explores symbiotic frameworks that pair SLMs with LLMs to improve reliability, efficiency, and privacy. It concludes with trends in trustworthiness, benchmarking, and future directions, emphasizing efficient architectures, domain expansion, and multimodal SLMs. Overall, the work provides a thorough framework for designing, evaluating, and deploying resource-efficient models that complement large-scale systems in privacy-conscious and latency-sensitive contexts.

Abstract

Large language models (LLMs) have demonstrated emergent abilities in text generation, question answering, and reasoning, facilitating various tasks and domains. Despite their proficiency in various tasks, LLMs like PaLM 540B and Llama-3.1 405B face limitations due to large parameter sizes and computational demands, often requiring cloud API use which raises privacy concerns, limits real-time applications on edge devices, and increases fine-tuning costs. Additionally, LLMs often underperform in specialized domains such as healthcare and law due to insufficient domain-specific knowledge, necessitating specialized models. Therefore, Small Language Models (SLMs) are increasingly favored for their low inference latency, cost-effectiveness, efficient development, and easy customization and adaptability. These models are particularly well-suited for resource-limited environments and domain knowledge acquisition, addressing LLMs' challenges and proving ideal for applications that require localized data handling for privacy, minimal inference latency for efficiency, and domain knowledge acquisition through lightweight fine-tuning. The rising demand for SLMs has spurred extensive research and development. However, a comprehensive survey investigating issues related to the definition, acquisition, application, enhancement, and reliability of SLM remains lacking, prompting us to conduct a detailed survey on these topics. The definition of SLMs varies widely, thus to standardize, we propose defining SLMs by their capability to perform specialized tasks and suitability for resource-constrained settings, setting boundaries based on the minimal size for emergent abilities and the maximum size sustainable under resource constraints. For other aspects, we provide a taxonomy of relevant models/methods and develop general frameworks for each category to enhance and utilize SLMs effectively.

Figures (32)

• Figure 1: Overview of Small Language Models.
• Figure 2: Download Statistics Last Month in Huggingface for LLMs with Various Model Sizes, obtained on October 7, 2024.
• Figure 3: A timeline of existing small language models.
• Figure 4: Transformer architecture vaswani2017attention.
• Figure 5: Mamba 1 architecture gu2023mamba.