Developing Adaptive Context Compression Techniques for Large Language Models (LLMs) in Long-Running Interactions
This addresses computational efficiency and memory issues for users of LLMs in long interactions, though it is incremental as it builds on existing compression methods.
The paper tackled performance degradation in large language models during long-running interactions by developing an adaptive context compression framework, achieving consistent improvements in conversational stability and retrieval performance while reducing token usage and inference latency.
Large Language Models (LLMs) often experience performance degradation during long-running interactions due to increasing context length, memory saturation, and computational overhead. This paper presents an adaptive context compression framework that integrates importance-aware memory selection, coherence-sensitive filtering, and dynamic budget allocation to retain essential conversational information while controlling context growth. The approach is evaluated on LOCOMO, LOCCO, and LongBench benchmarks to assess answer quality, retrieval accuracy, coherence preservation, and efficiency. Experimental results demonstrate that the proposed method achieves consistent improvements in conversational stability and retrieval performance while reducing token usage and inference latency compared with existing memory and compression-based approaches. These findings indicate that adaptive context compression provides an effective balance between long-term memory preservation and computational efficiency in persistent LLM interactions