Related papers: TIDE: Temporal Incremental Draft Engine for Self-I…

Modern web applications--from real-time content recommendation and dynamic pricing to CDN optimization--increasingly rely on time-series forecasting to deliver personalized experiences to billions of users. Large-scale Transformer-based…

Recent work has improved language models (LMs) remarkably by equipping them with a non-parametric memory component. However, most existing approaches only introduce mem-ories at testing time or represent them using a separately trained…

This paper presents "Predictive Pipelined Decoding (PPD)," an approach that speeds up greedy decoding in Large Language Models (LLMs) while maintaining the exact same output as the original decoding. Unlike conventional strategies, PPD…

Diffusion models have emerged as the mainstream approach for visual generation. However, these models typically suffer from sample inefficiency and high training costs. Consequently, methods for efficient finetuning, inference and…

Rich textual and topological information of textual graphs need to be modeled in real-world applications such as webpages, e-commerce, and academic articles. Practitioners have been long following the path of adopting a shallow text encoder…

Subject-driven image generation (SDIG) aims to manipulate specific subjects within images while adhering to textual instructions, a task crucial for advancing text-to-image diffusion models. SDIG requires reconciling the tension between…

Tree-based speculative decoding accelerates autoregressive generation by verifying multiple draft candidates in parallel, but this advantage weakens for sparse Mixture-of-Experts (MoE) models. As the draft tree grows, different branches…

Although Large Language Models (LLMs) have made significant progress in code generation, they still struggle with code generation tasks in specific scenarios. These scenarios usually necessitate the adaptation of LLMs to fulfill specific…

The enhancement of Visual Language Models (VLMs) has traditionally relied on knowledge distillation from larger, more capable models. This dependence creates a fundamental bottleneck for improving state-of-the-art systems, particularly when…

Inference efficiency in Large Language Models (LLMs) is fundamentally limited by their serial, autoregressive generation, especially as reasoning becomes a key capability and response sequences grow longer. Speculative decoding (SD) offers…

Speculative decoding (SD) accelerates LLM inference by verifying draft tokens in parallel. However, this method presents a critical trade-off: it improves throughput in low-load, memory-bound systems but degrades performance in high-load,…

Diffusion LLMs (dLLMs) have recently emerged as a powerful alternative to autoregressive LLMs (AR-LLMs) with the potential to operate at significantly higher token generation rates. However, currently available open-source dLLMs often…

Tool-Integrated Reasoning (TIR) enables large language models (LLMs) to improve their internal reasoning ability by integrating external tools. However, models employing TIR often display suboptimal behaviors, such as insufficient or…

Cloud-based Large Language Model (LLM) services often face challenges in achieving low inference latency and meeting Service Level Objectives (SLOs) under dynamic request patterns. Speculative decoding, which exploits lightweight models for…

Diffusion large language models (dLLMs) offer parallel decoding and bidirectional context, but state-of-the-art dLLMs require billions of parameters for competitive performance. While existing distillation methods for dLLMs reduce inference…

With the increasingly giant scales of (causal) large language models (LLMs), the inference efficiency comes as one of the core concerns along the improved performance. In contrast to the memory footprint, the latency bottleneck seems to be…

Text-to-image person re-identification (TIReID) aims to retrieve the target person from an image gallery via a textual description query. Recently, pre-trained vision-language models like CLIP have attracted significant attention and have…

Speculative decoding is a powerful technique for reducing the latency of Large Language Models (LLMs), offering a fault-tolerant framework that enables the use of highly compressed draft models. In this work, we introduce Self-Distilled…

LLM-based equation discovery offers a promising route to recovering symbolic laws from data, but many systems still rely on generation-centered loops that propose candidates, fit parameters, score results, and reuse selected examples. Such…

Speculative decoding accelerates large language model inference by proposing tokens with a lightweight draft model and selectively accepting them using a target model. This work introduces DropMatch, a novel approach that matches draft…