English
Related papers

Related papers: All is Not Lost: LLM Recovery without Checkpoints

200 papers

Best-of-N decoding methods instruct large language models (LLMs) to generate multiple solutions, score each using a scoring function, and select the highest scored as the final answer to mathematical reasoning problems. However, this…

Computation and Language · Computer Science 2024-10-18 Zhenyu Wu , Qingkai Zeng , Zhihan Zhang , Zhaoxuan Tan , Chao Shen , Meng Jiang

Using tools by Large Language Models (LLMs) is a promising avenue to extend their reach beyond language or conversational settings. The number of tools can scale to thousands as they enable accessing sensory information, fetching updated…

Information Retrieval · Computer Science 2024-12-06 Mohammad Kachuee , Sarthak Ahuja , Vaibhav Kumar , Puyang Xu , Xiaohu Liu

Phase retrieval is a nonlinear inverse problem that arises in a wide range of imaging modalities, from electron microscopy to Fourier ptychography. In particular, the reconstruction is facilitated when the sensing matrix is i.i.d. random,…

Low rank model arises from a wide range of applications, including machine learning, signal processing, computer algebra, computer vision, and imaging science. Low rank matrix recovery is about reconstructing a low rank matrix from…

Numerical Analysis · Mathematics 2018-09-12 Jian-Feng Cai , Ke Wei

Large language model (LLM) agents on multi-step tasks suffer reasoning degradation, looping, drift, stuck states, at rates up to 30% on hard tasks. Current solutions include hard step limits (abrupt) or LLM-as-judge monitoring (10-15%…

Artificial Intelligence · Computer Science 2026-04-16 Rafflesia Khan , Nafiul Islam Khan

Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of tasks, but their deployment is often constrained by substantial memory footprints and computational costs. While prior work has achieved…

Machine Learning · Computer Science 2026-02-04 Jiangyong Yu , Xiaomeng Han , Xing Hu , Chen Xu , Zhe Jiang , Dawei Yang

Self-powered intermittent systems typically adopt runtime checkpointing as a means to accumulate computation progress across power cycles and recover system status from power failures. However, existing approaches based on the checkpointing…

Operating Systems · Computer Science 2019-10-14 Wei-Ming Chen , Tei-Wei-Kuo , Pi-Cheng Hsiu

This study presents a meshless-based local reanalysis (MLR) method. The purpose of this study is to extend reanalysis methods to the Kriging interpolation meshless method due to its high efficiency. In this study, two reanalysis methods:…

Computational Engineering, Finance, and Science · Computer Science 2017-11-15 Zhenxing Cheng , Hu Wang

Verification-guided self-improvement has recently emerged as a promising approach to improving the accuracy of large language model (LLM) outputs. However, existing approaches face a trade-off between inference efficiency and accuracy:…

Computation and Language · Computer Science 2026-03-24 Yuran Li , Di Wu , Benoit Boulet

Foundation models (FMs) such as CLIP have demonstrated impressive zero-shot performance across various tasks by leveraging large-scale, unsupervised pre-training. However, they often inherit harmful or unwanted knowledge from noisy…

Computer Vision and Pattern Recognition · Computer Science 2025-06-04 Zeliang Zhang , Gaowen Liu , Charles Fleming , Ramana Rao Kompella , Chenliang Xu

As computers reach exascale and beyond, the incidence of faults will increase. Solutions to this problem are an active research topic. We focus on strategies to make the preconditioned conjugate gradient (PCG) solver resilient against node…

Distributed, Parallel, and Cluster Computing · Computer Science 2020-07-09 Carlos Pachajoa , Christina Pacher , Markus Levonyak , Wilfried N. Gansterer

Most Continual Learning (CL) methods maintain performance on earlier tasks by storing exemplars in a replay buffer, introducing memory overhead that scales with the number of tasks and raising privacy concerns in regulated domains. We…

Machine Learning · Computer Science 2026-05-11 Mohammad Ali Vahedifar , Qi Zhang

Large language models (LLMs) have demonstrated remarkable success across various application domains, but their enormous sizes and computational demands pose significant challenges for deployment on resource-constrained edge devices. To…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-03-20 Kai Zhang , Hengtao He , Shenghui Song , Jun Zhang , Khaled B. Letaief

LLMs have achieved remarkable breakthroughs in reasoning, insights, and tool use, but chaining these abilities into extended processes at the scale of those routinely executed by humans, organizations, and societies has remained out of…

Fault localization, the process of identifying the software components responsible for failures, is essential but often time-consuming. Recent advances in Large Language Models (LLMs) have enabled fault localization without extensive defect…

Software Engineering · Computer Science 2025-06-05 Inseok Yeo , Duksan Ryu , Jongmoon Baik

Models of sensory processing and learning in the cortex need to efficiently assign credit to synapses in all areas. In deep learning, a known solution is error backpropagation, which however requires biologically implausible weight…

Neurons and Cognition · Quantitative Biology 2024-02-05 Kevin Max , Laura Kriener , Garibaldi Pineda García , Thomas Nowotny , Ismael Jaras , Walter Senn , Mihai A. Petrovici

Optimizing LLM-based workflows is typically formulated as a global search, where candidate workflows are evaluated based on a scalar metric. This paradigm, however, suffers from a critical flaw: information collapse. By reducing rich,…

Artificial Intelligence · Computer Science 2025-10-14 Jusheng Zhang , Kaitong Cai , Qinglin Zeng , Ningyuan Liu , Stephen Fan , Ziliang Chen , Keze Wang

Checkpoint merging is a technique for combining multiple model snapshots into a single superior model, potentially reducing training time for large language models. This paper explores checkpoint merging in the context of…

Machine Learning · Computer Science 2025-04-29 Shi Jie Yu , Sehyun Choi

Unlike traditional central training, federated learning (FL) improves the performance of the global model by sharing and aggregating local models rather than local data to protect the users' privacy. Although this training approach appears…

Machine Learning · Computer Science 2022-01-27 Jiahui Geng , Yongli Mou , Feifei Li , Qing Li , Oya Beyan , Stefan Decker , Chunming Rong

Schedule-Free Learning has shown promise as a practical anytime training method for machine learning, showing success across dozens of standard benchmark problems. However, strong performance for LLM training has only been demonstrated at…

Machine Learning · Computer Science 2026-05-20 Aaron Defazio