Modern large language model (LLM) applications exhibit diverse service-level objectives (SLOs), from low-latency requirements in interactive coding assistants to more relaxed constraints in data wrangling tasks. Existing LLM serving systems, which rely on uniform batching and scheduling strategies, often fail to meet these heterogeneous SLOs concurrently. We present AdaServe, the first LLM serving system designed to support efficient multi-SLO serving through SLO-customized speculative decoding. AdaServe formulates multi-SLO serving as a constrained optimization problem and introduces a hardware-aware algorithm that constructs a speculation tree tailored to each request's latency target. It features a speculate-select-verify pipeline that enables fine-grained control over decoding speed while maximizing system throughput. AdaServe further adapts to workload variation by dynamically adjusting speculation parameters. Evaluations across diverse workloads show that AdaServe reduces SLO violations by up to 4.3X and improves goodput by up to 1.9X compared to the best-performing baselines, highlighting its effectiveness in multi-SLO serving.

Large language models (LLMs) have driven significant advancements across diverse NLP tasks, with long-context models gaining prominence for handling extended inputs. However, the expanding key-value (KV) cache size required by Transformer architectures intensifies the memory constraints, particularly during the decoding phase, creating a significant bottleneck. Existing sparse attention mechanisms designed to address this bottleneck have two limitations: (1) they often fail to reliably identify the most relevant tokens for attention, and (2) they overlook the spatial coherence of token selection across consecutive Transformer layers, which can lead to performance degradation and substantial overhead in token selection. This paper introduces TidalDecode, a simple yet effective algorithm and system for fast and accurate LLM decoding through position persistent sparse attention. TidalDecode leverages the spatial coherence of tokens selected by existing sparse attention methods and introduces a few token selection layers that perform full attention to identify the tokens with the highest attention scores, while all other layers perform sparse attention with the pre-selected tokens. This design enables TidalDecode to substantially reduce the overhead of token selection for sparse attention without sacrificing the quality of the generated results. Evaluation on a diverse set of LLMs and tasks shows that TidalDecode closely matches the generative performance of full attention methods while reducing the LLM decoding latency by up to $2.1\times$.

Recent advances in inference-time compute have significantly improved performance on complex tasks by generating long chains of thought (CoTs) using Large Reasoning Models (LRMs). However, this improved accuracy comes at the cost of high inference latency due to the length of generated reasoning sequences and the autoregressive nature of decoding. Our key insight in tackling these overheads is that LRM inference, and the reasoning that it embeds, is highly tolerant of approximations: complex tasks are typically broken down into simpler steps, each of which brings utility based on the semantic insight it provides for downstream steps rather than the exact tokens it generates. Accordingly, we introduce SpecReason, a system that automatically accelerates LRM inference by using a lightweight model to (speculatively) carry out simpler intermediate reasoning steps and reserving the costly base model only to efficiently assess (and potentially correct) the speculated outputs. Importantly, SpecReason's focus on exploiting the semantic flexibility of thinking tokens in preserving final-answer accuracy is complementary to prior speculation techniques, most notably speculative decoding, which demands token-level equivalence at each step. Across a variety of cross-domain reasoning benchmarks, SpecReason achieves 1.4-3.0$\times$ speedup over vanilla LRM inference while improving accuracy by 0.4-9.0%. Compared to speculative decoding without SpecReason, their combination yields an additional 8.8-58.0% latency reduction. We open-source SpecReason at \url{https://anonymous.4open.science/r/specreason/}.

Zhengxin Zhang, Dan Zhao, Xupeng Miao, Gabriele Oliaro, Zhihao Zhang, Qing Li, Yong Jiang, Zhihao Jia. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

This paper introduces SpecInfer, a system that accelerates generative large language model (LLM) serving with tree-based speculative inference and verification. The key idea behind SpecInfer is leveraging small speculative models to predict the LLM's outputs; the predictions are organized as a token tree, whose nodes each represent a candidate token sequence. The correctness of all candidate token sequences represented by a token tree is verified against the LLM in parallel using a novel tree-based parallel decoding mechanism. SpecInfer uses an LLM as a token tree verifier instead of an incremental decoder, which significantly reduces the end-to-end latency and computational requirement for serving generative LLMs while provably preserving model quality. Our evaluation shows that SpecInfer outperforms existing LLM serving systems by 1.5-2.8× for distributed LLM inference and by 2.6-3.5× for offloading-based LLM inference, while preserving the same generative performance. SpecInfer is publicly available at https://github.com/flexflow/FlexFlow/

This paper introduces RaLMSpec, a framework that accelerates iterative retrieval-augmented language model (RaLM) with *speculative retrieval* and *batched verification*. RaLMSpec further introduces several important systems optimizations, including prefetching, optimal speculation stride scheduler, and asynchronous verification. The combination of these techniques allows RaLMSPec to significantly outperform existing systems. For document-level iterative RaLM serving, evaluation over three LLMs on four QA datasets shows that RaLMSpec improves over existing approaches by $1.75$-$2.39\times$, $1.04$-$1.39\times$, and $1.31$-$1.77\times$ when the retriever is an exact dense retriever, approximate dense retriever, and sparse retriever respectively. For token-level iterative RaLM (KNN-LM) serving, RaLMSpec is up to $7.59\times$ and $2.45\times$ faster than existing methods for exact dense and approximate dense retrievers, respectively.

In this work, we study the experts problem in the distributed setting where an expert's cost needs to be aggregated across multiple servers. Our study considers various communication models such as the message-passing model and the broadcast model, along with multiple aggregation functions, such as summing and taking the $\ell_p$ norm of an expert's cost across servers. We propose the first communication-efficient protocols that achieve near-optimal regret in these settings, even against a strong adversary who can choose the inputs adaptively. Additionally, we give a conditional lower bound showing that the communication of our protocols is nearly optimal. Finally, we implement our protocols and demonstrate empirical savings on the HPO-B benchmarks.

An effective understanding of the environment and accurate trajectory prediction of surrounding dynamic obstacles are indispensable for intelligent mobile systems (e.g., autonomous vehicles, social robots) to achieve safe and high-quality planning when they navigate in highly interactive and crowded scenarios. Due to the existence of frequent interactions and uncertainty in the scene evolution, it is desired for the prediction system to enable relational reasoning on different entities and provide a distribution of future trajectories for each agent. In this paper, we propose a generic generative neural system (called STG-DAT) for multi-agent trajectory prediction involving heterogeneous agents. The system takes a step forward to explicit interaction modeling by incorporating relational inductive biases with a dynamic graph representation and leverages both trajectory and scene context information. We also employ an efficient kinematic constraint layer applied to vehicle trajectory prediction. The constraint not only ensures physical feasibility but also enhances model performance. Moreover, the proposed prediction model can be easily adopted by multi-target tracking frameworks. The tracking accuracy proves to be improved by empirical results. The proposed system is evaluated on three public benchmark datasets for trajectory prediction, where the agents cover pedestrians, cyclists and on-road vehicles. The experimental results demonstrate that our model achieves better performance than various baseline approaches in terms of prediction and tracking accuracy.