Tootfinder

MetaVLA: Unified Meta Co-training For Efficient Embodied Adaption
Chen Li, Zhantao Yang, Han Zhang, Fangyi Chen, Chenchen Zhu, Anudeepsekhar Bolimera, Marios Savvides
https://arxiv.org/abs/2510.05580

MetaVLA: Unified Meta Co-training For Efficient Embodied Adaption
Vision-Language-Action (VLA) models show promise in embodied reasoning, yet remain far from true generalists-they often require task-specific fine-tuning, and generalize poorly to unseen tasks. We propose MetaVLA, a unified, backbone-agnostic post-training framework for efficient and scalable alignment. MetaVLA introduces Context-Aware Meta Co-Training, which consolidates diverse target tasks into a single fine-tuning stage while leveraging structurally diverse auxiliary tasks to improve in-dom…

Crosslisted article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[2/4]:
- LightCache: Memory-Efficient, Training-Free Acceleration for Video Generation
Yang Xiao, Gen Li, Kaiyuan Deng, Yushu Wu, Zheng Zhan, Yanzhi Wang, Xiaolong Ma, Bo Hui

VLMQ: Efficient Post-Training Quantization for Large Vision-Language Models via Hessian Augmentation
Yufei Xue, Yushi Huang, Jiawei Shao, Jun Zhang
https://arxiv.org/abs/2508.03351

VLMQ: Efficient Post-Training Quantization for Large Vision-Language Models via Hessian Augmentation
Post-training quantization (PTQ) has emerged as an effective approach for compressing large models and accelerating their inference without retraining. While PTQ has been extensively studied in the context of large language models (LLMs), its applicability to vision-language models (VLMs) remains underexplored. In this paper, we identify a modality discrepancy (\emph{i.e.}, limited text tokens \emph{vs.} excessive and redundant vision tokens) of VLMs. However, existing Hessian-based LLM PTQ met…

Geometry of Distance Protection
Josh A. Taylor, Alejandro D. Dom\'inguez-Garc\'ia
https://arxiv.org/abs/2510.04379 https://arxiv.org/pdf/2510.04379…

Geometry of Distance Protection
Distance relays detect faults on transmission lines. They face uncertainty from the fault's location and resistance, as well as the current from the line's remote terminal. In this paper, we aggregate this uncertainty with the Minkowski sum. This allows us to explicitly model the power grid surrounding the relay's line, and in turn accommodate any mix of synchronous machines and inverter-based resources. To make the relay's task easier, inverters can inject perturbations, or auxiliary signals, …

FlashRecovery: Fast and Low-Cost Recovery from Failures for Large-Scale Training of LLMs
Haijun Zhang, Jinxiang Wang, Zhenhua Yu, Yanyong Zhang, Xuejie Ji, Kaining Mao, Jun Zhang, Yaqing Zhang, Ting Wu, Fei Jie, Xiemin Huang, Zhifang Cai, Junhua Cheng, Shuwei Wang, Wei Li, Xiaoming Bao, Hua Xu, Shixiong Zhao, Jun Li, Hongwei Sun, Ziyang Zhang, Yi Xiong, Chunsheng Li

FlashRecovery: Fast and Low-Cost Recovery from Failures for Large-Scale Training of LLMs
Large language models (LLMs) have made a profound impact across various fields due to their advanced capabilities. However, training these models at unprecedented scales requires extensive AI accelerator clusters and sophisticated parallelism strategies, which pose significant challenges in maintaining system reliability over prolonged training periods. A major concern is the substantial loss of training time caused by inevitable hardware and software failures. To address these challenges, we p…

A general framework for knowledge integration in machine learning for electromagnetic scattering using quasinormal modes
Viktor A. Lilja, Albin J. Sv\"ardsby, Timo Gahlmann, Philippe Tassin
https://arxiv.org/abs/2509.06130

A general framework for knowledge integration in machine learning for electromagnetic scattering using quasinormal modes
Neural networks have been demonstrated to be able to provide fast surrogate models that can accelerate the inverse design of optical and electromagnetic devices. Nevertheless, such neural networks can be unreliable and normally require extreme amounts of data for training. Here we show that these limitations can be alleviated by constraining neural-network models using prior knowledge about the governing physics. We propose a universal physics-informed neural network for electromagnetic scatter…

C2AL: Cohort-Contrastive Auxiliary Learning for Large-scale Recommendation Systems
Mertcan Cokbas, Ziteng Liu, Zeyi Tao, Chengkai Zhang, Elder Veliz, Qin Huang, Ellie Wen, Huayu Li, Qiang Jin, Murat Duman, Benjamin Au, Guy Lebanon, Sagar Chordia
https://arxiv.org/abs/2510.02215

C2AL: Cohort-Contrastive Auxiliary Learning for Large-scale Recommendation Systems
Training large-scale recommendation models under a single global objective implicitly assumes homogeneity across user populations. However, real-world data are composites of heterogeneous cohorts with distinct conditional distributions. As models increase in scale and complexity and as more data is used for training, they become dominated by central distribution patterns, neglecting head and tail regions. This imbalance limits the model's learning ability and can result in inactive attention we…

Fulcrum: Optimizing Concurrent DNN Training and Inferencing on Edge Accelerators
Prashanthi S. K., Saisamarth Taluri, Pranav Gupta, Amartya Ranjan Saikia, Kunal Kumar Sahoo, Atharva Vinay Joshi, Lakshya Karwa, Kedar Dhule, Yogesh Simmhan
https://arxiv.org/abs/2509.20205

Fulcrum: Optimizing Concurrent DNN Training and Inferencing on Edge Accelerators
The proliferation of GPU accelerated edge devices like Nvidia Jetsons and the rise in privacy concerns are placing an emphasis on concurrent DNN training and inferencing on edge devices. Inference and training have different computing and QoS goals. But edge accelerators like Jetson do not support native GPU sharing and expose 1000s of power modes. This requires careful time-sharing of concurrent workloads to meet power--performance goals, while limiting costly profiling. In this paper, we desi…

Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
Prashanthi S. K., Sai Anuroop Kesanapalli, Yogesh Simmhan
https://arxiv.org/abs/2509.20160

Characterizing the Performance of Accelerated Jetson Edge Devices for Training Deep Learning Models
Deep Neural Networks (DNNs) have had a significant impact on domains like autonomous vehicles and smart cities through low-latency inferencing on edge computing devices close to the data source. However, DNN training on the edge is poorly explored. Techniques like federated learning and the growing capacity of GPU-accelerated edge devices like NVIDIA Jetson motivate the need for a holistic characterization of DNN training on the edge. Training DNNs is resource-intensive and can stress an edge's…

HAPT: Heterogeneity-Aware Automated Parallel Training on Heterogeneous Clusters
Antian Liang, Zhigang Zhao, Kai Zhang, Xuri Shi, Chuantao Li, Chunxiao Wang, Zhenying He, Yinan Jing, X. Sean Wang
https://arxiv.org/abs/2509.24859

HAPT: Heterogeneity-Aware Automated Parallel Training on Heterogeneous Clusters
With the rapid evolution of GPU architectures, the heterogeneity of model training infrastructures is steadily increasing. In such environments, effectively utilizing all available heterogeneous accelerators becomes critical for distributed model training. However, existing frameworks, which are primarily designed for homogeneous clusters, often exhibit significant resource underutilization when deployed on heterogeneous accelerators and networks. In this paper, we present Hapt, an automated pa…