LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference
Reviewed by Pithpith:DVJ4QDECopen to challenge →
read the original abstract
The inference of transformer-based large language models consists of two sequential stages: 1) a prefilling stage to compute the KV cache of prompts and generate the first token, and 2) a decoding stage to generate subsequent tokens. For long prompts, the KV cache must be computed for all tokens during the prefilling stage, which can significantly increase the time needed to generate the first token. Consequently, the prefilling stage may become a bottleneck in the generation process. An open question remains whether all prompt tokens are essential for generating the first token. To answer this, we introduce a novel method, LazyLLM, that selectively computes the KV for tokens important for the next token prediction in both the prefilling and decoding stages. Contrary to static pruning approaches that prune the prompt at once, LazyLLM allows language models to dynamically select different subsets of tokens from the context in different generation steps, even though they might be pruned in previous steps. Extensive experiments on standard datasets across various tasks demonstrate that LazyLLM is a generic method that can be seamlessly integrated with existing language models to significantly accelerate the generation without fine-tuning. For instance, in the multi-document question-answering task, LazyLLM accelerates the prefilling stage of the LLama 2 7B model by 2.34x while maintaining accuracy.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
Long Context Pre-Training with Lighthouse Attention
Lighthouse Attention enables faster long-context pre-training via gradient-free symmetrical hierarchical compression of QKV while preserving causality, followed by a short full-attention recovery that yields lower los...
-
PrefixWall: Mitigating Prefix Caching Side Channels in Shared LLM Systems
PrefixWall mitigates APC side channels in multi-tenant LLM systems via selective prefix isolation, delivering up to 70% higher cache reuse and 30% lower latency than full-isolation baselines.
-
ProactiveLLM: Learning Active Interaction for Streaming Large Language Models
ProactiveLLM enables active interaction in streaming LLMs by learning semantic sufficiency cues from partial inputs through mask-based modeling and synchronized privileged self-distillation without external supervision.
-
UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification
UniPrefill accelerates LLM prefill via block-wise dynamic sparsification, achieving up to 2.1x TTFT speedup while supporting hybrid architectures and native vLLM continuous batching.
-
HISA: Efficient Hierarchical Indexing for Fine-Grained Sparse Attention
HISA speeds up fine-grained sparse attention indexers via block-then-token hierarchy, delivering substantial speedups at 64K context with no training and quality matching the original DSA on long-context benchmarks.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.