Related papers: StepCache: Step-Level Reuse with Lightweight Verif…

The rise of million-token, agent-based applications has placed unprecedented demands on large language model (LLM) inference services. The long-running nature of these tasks increases their susceptibility to hardware and software faults,…

We introduce SwiftCache, a "fresh" learning-based caching framework designed for content distribution networks (CDNs) featuring distributed front-end local caches and a dynamic back-end database. Users prefer the most recent version of the…

We propose cache steering, a lightweight method for implicit steering of language models via a one-shot intervention applied directly to the key-value cache. To validate its effectiveness, we apply cache steering to induce chain-of-thought…

Long-context decoding in LLMs is IO-bound: each token re-reads an ever-growing KV cache. Prior accelerations cut bytes via compression, which lowers fidelity, or selection/eviction, which restricts what remains accessible, and both can…

Handling faults is a growing concern in HPC. In future exascale systems, it is projected that silent undetected errors will occur several times a day, increasing the occurrence of corrupted results. In this article, we propose SEDAR, which…

In RL post-training of LLM agents, calls to external tools take several seconds or even minutes, leaving allocated GPUs idle and inflating post-training time and cost. While many tool invocations repeat across parallel rollouts and could in…

The Key-Value (KV) cache is integral to efficient autoregressive inference in large language models (LLMs), yet its unbounded growth in stateful multi-turn scenarios presents major challenges. This paper examines the interplay between KV…

This paper presents an efficient speech enhancement (SE) approach that reuses a processing block repeatedly instead of conventional stacking. Rather than increasing the number of blocks for learning deep latent representations, repeating a…

Large Language Models have achieved strong performance on reasoning tasks with objective answers by generating step-by-step solutions, but diagnosing where a multi-step reasoning trace might fail remains difficult. Confidence estimation…

Large reasoning models (LRMs) that generate long chains of thought now perform well on multi-step math, science, and coding tasks. However, their behavior is still unstable and hard to interpret, and existing analysis tools struggle with…

Many computational factors limit broader deployment of large language models. In this paper, we focus on a memory bottleneck imposed by the key-value (KV) cache, a computational shortcut that requires storing previous KV pairs during…

Caching has the potential to be of significant benefit for accessing large language models (LLMs) due to their high latencies which typically range from a small number of seconds to well over a minute. Furthermore, many LLMs charge money…

Large Language Models (LLMs) rely heavily on Key-Value (KV) caching to minimize inference latency. However, standard KV caches are context-dependent: reusing a cached document in a new context requires recomputing KV states to account for…

Chain-of-Thought reasoning significantly improves the performance of large language models on complex tasks, but incurs high inference latency due to long generation traces. Step-level speculative reasoning aims to mitigate this cost, yet…

Randomized, skewed caches (RSCs) such as CEASER-S have recently received much attention to defend against contention-based cache side channels. By randomizing and regularly changing the mapping(s) of addresses to cache sets, these…

Recent progress in large language models (LLMs) has advanced automatic code generation, yet most approaches rely on direct, single-step translation from problem descriptions to code, disregarding structured software engineering practices.…

Large Language Models (LLMs) excel across a variety of language tasks yet are constrained by limited input lengths and high computational costs. Existing approaches\textemdash such as relative positional encodings (e.g., RoPE, ALiBi) and…

Spin-Transfer Torque RAM (STTRAM) is promising for cache applications. However, it brings new data security issues that were absent in volatile memory counterparts such as Static RAM (SRAM) and embedded Dynamic RAM (eDRAM). This is…

Latency Based Tiling provides a systems based approach to deriving approximate tiling solution that maximizes locality while maintaining a fast compile time. The method uses triangular loops to characterize miss ratio scaling of a machine…

Training large-scale image recognition models is computationally expensive. This raises the question of whether there might be simple ways to improve the test performance of an already trained model without having to re-train or fine-tune…