Compressing Transformer inputs into compressd tokens allows running LLMs with improved speed and cost efficiency. Based on the compression method ICAE, we carefully examine the position identifier choices for compressed tokens and also propose a new compression loss. We demonstrate empirically that our proposed methods achieve significantly higher compression ratios (15x compared to 4x for ICAE), while being able to attain comparable reconstruction performance.

Prompt compression is crucial for enhancing inference speed, reducing costs, and improving user experience. However, current methods face challenges such as low compression ratios and potential data leakage during evaluation. To address these issues, we propose 500xCompressor, a method that compresses extensive natural language contexts into a minimum of one single special token. The 500xCompressor introduces approximately 0.3% additional parameters and achieves compression ratios ranging from 6x to 480x. It is designed to compress any text, answer various types of questions, and could be utilized by the original large language model (LLM) without requiring fine-tuning. Initially, 500xCompressor was pretrained on the Arxiv Corpus, followed by fine-tuning on the ArxivQA dataset, and subsequently evaluated on strictly unseen and classical question answering (QA) datasets. The results demonstrate that the LLM retained 62.26-72.89% of its capabilities compared to using non-compressed prompts. This study also shows that not all the compressed tokens are equally utilized and that K V values have significant advantages over embeddings in preserving information at high compression ratios. The highly compressive nature of natural language prompts, even for fine-grained complex information, suggests promising potential for future applications and further research into developing a new LLM language.

We propose the In-context Autoencoder (ICAE), leveraging the power of a large language models (LLM) to compress a long context into short compact memory slots that can be directly conditioned on by the LLM for various purposes. ICAE is first pretrained using both autoencoding and language modeling objectives on massive text data, enabling it to generate memory slots that accurately and comprehensively represent the original context; Then, it is fine-tuned on instruction data for producing desirable responses to various prompts. Experiments demonstrate that our lightweight ICAE, introducing fewer than 1% additional parameters, effectively achieves 4X context compression based on Llama, offering advantages in both improved latency and GPU memory cost during inference, and showing an interesting insight in memorization as well as potential for scalability. These promising results imply a novel perspective on the connection between working memory in cognitive science and representation learning in LLMs, revealing ICAE's significant implications in addressing the long context problem and suggesting further research in LLM context management. Our data, code and model are released at https://github.com/getao/icae.

The goal of intelligent computer-aided engineering (ICAE) is to construct computer programs that capture a significant fraction of an engineer's knowledge. Today, ICAE systems are a goal, not a reality. This article attempts to refine that goal and suggest how to get there. We begin by examining several scenarios of what ICAE systems could be like. Next we describe why ICAE won't evolve directly from current applications of expert system technology to engineering problems.

The goal of intelligent computer-aided engineering (ICAE) is to construct computer programs that capture a significant fraction of an engineer's knowledge. Today, ICAE systems are a goal, not a reality. This article attempts to refine that goal and suggest how to get there. We begin by examining several scenarios of what ICAE systems could be like. Next we describe why ICAE won't evolve directly from current applications of expert system technology to engineering problems. I focus on qualitative physics as a critical area where progress is needed, both in terms of representations and styles of reasoning.