Towards Large Model Feature Coding

• URL: http://arxiv.org/abs/2605.24025v1
• Date: Wed, 20 May 2026 13:39:48 GMT
• Title: Towards Large Model Feature Coding
• Abstract summary: We present a benchmark and evaluation framework for large model feature coding (LaMoFC)<n>We first build the feature dataset LaMoFCBench, covering diverse task requirements across 4 categories and 16 scenarios.<n>We then specify representative split points according to practical application scenarios to extract intermediate features.
• Score: 90.89390892560094
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large models have delivered remarkable performance across a wide range of perception and generation tasks, yet practical deployment is increasingly constrained by computational and memory budgets, as well as privacy requirements. Split execution alleviates these constraints by partitioning computation across devices, but it inevitably introduces intensive transmission and storage of intermediate features. Unlike conventional feature coding for CNNs that typically targets homogeneous spatial activation maps, modern large models generate heterogeneous features with varying statistical distributions and compression tolerances, e.g., multi-level/multi-modal representations and autoregressive context caches. These characteristics necessitate treating large model feature coding (LaMoFC) as a fundamental system component and call for a systematic evaluation framework. In this paper, we present a comprehensive benchmark and evaluation framework for LaMoFC. We first build the feature dataset LaMoFCBench, covering diverse task requirements across 4 categories and 16 scenarios while integrating widelyadopted architectures and various split-computing settings. We then specify representative split points according to practical application scenarios to extract intermediate features, establishing a unified pipeline for fair and reproducible comparisons. Finally, we benchmark mainstream universal feature codecs, exposing the profound misalignment between existing coding paradigms and the heterogeneous nature of large model features. These findings reveal that LaMoFC demands a fundamental departure from existing paradigms, and LaMoFCBench provides the shared empirical foundation to drive this transition. The data and code will be available at https://github.com/lartpang/LaMoFCBench.

Related papers

• MSCM-net: A hyperspectral image classiffcation method based on multi-scale convolution and Mamba [10.40281964904095]
  We propose an innovative hyperspectral image classification model, MSCM-net.<n>It consists of a model architecture combining multi-scale CNN and Mamba.<n>It can achieve advanced classification performance while reducing computational complexity.
  arXiv  Detail & Related papers  (2026-07-30T14:27:59Z)
• Disentangle-then-Align: Non-Iterative Hybrid Multimodal Image Registration via Cross-Scale Feature Disentanglement [19.39282311652453]
  Multimodal image registration is a fundamental task and a prerequisite for downstream cross-modal analysis.<n>We formulate hybrid multimodal registration as jointly learning a stable shared feature space and a unified hybrid transformation.<n>A shared backbone with Modality-Specific Batch Normalization (MSBN) extracts multi-scale features, while a Cross-scale Disentanglement and Adaptive Projection (CDAP) module suppresses modality-private cues and projects shared features into a stable subspace for matching.
  arXiv  Detail & Related papers  (2026-03-20T04:03:38Z)
• NExT-OMNI: Towards Any-to-Any Omnimodal Foundation Models with Discrete Flow Matching [64.10695425442164]
  We introduce NExT-OMNI, an open-source omnimodal foundation model that achieves unified modeling through discrete flow paradigms.<n>Trained on large-scale interleaved text, image, video, and audio data, NExT-OMNI delivers competitive performance on multimodal generation and understanding benchmarks.<n>To advance further research, we release training details, data protocols, and open-source both the code and model checkpoints.
  arXiv  Detail & Related papers  (2025-10-15T16:25:18Z)
• An Efficient and Mixed Heterogeneous Model for Image Restoration [71.85124734060665]
  Current mainstream approaches are based on three architectural paradigms: CNNs, Transformers, and Mambas.<n>We propose RestorMixer, an efficient and general-purpose IR model based on mixed-architecture fusion.
  arXiv  Detail & Related papers  (2025-04-15T08:19:12Z)
• ContextFormer: Redefining Efficiency in Semantic Segmentation [48.81126061219231]
  Convolutional methods, although capturing local dependencies well, struggle with long-range relationships.<n>Vision Transformers (ViTs) excel in global context capture but are hindered by high computational demands.<n>We propose ContextFormer, a hybrid framework leveraging the strengths of CNNs and ViTs in the bottleneck to balance efficiency, accuracy, and robustness for real-time semantic segmentation.
  arXiv  Detail & Related papers  (2025-01-31T16:11:04Z)
• FlexiCrackNet: A Flexible Pipeline for Enhanced Crack Segmentation with General Features Transfered from SAM [24.99233476254989]
  FlexiCrackNet is a novel pipeline that seamlessly integrates traditional deep learning paradigms with the strengths of large-scale pre-trained models.<n>Experiments show that FlexiCrackNet outperforms state-of-the-art methods, excels in zero-shot generalization, computational efficiency, and segmentation robustness.<n>These advancements underscore the potential of FlexiCrackNet for real-world applications in automated crack detection and comprehensive structural health monitoring systems.
  arXiv  Detail & Related papers  (2025-01-31T02:37:09Z)
• A Refreshed Similarity-based Upsampler for Direct High-Ratio Feature Upsampling [54.05517338122698]
  A popular similarity-based feature upsampling pipeline has been proposed, which utilizes a high-resolution feature as guidance.<n>We propose an explicitly controllable query-key feature alignment from both semantic-aware and detail-aware perspectives.<n>We develop a fine-grained neighbor selection strategy on HR features, which is simple yet effective for alleviating mosaic artifacts.
  arXiv  Detail & Related papers  (2024-07-02T14:12:21Z)
• Transformer Architecture for NetsDB [0.0]
  We create an end-to-end implementation of a transformer for deep learning model serving in NetsDB. We load out weights from our model for distributed processing, deployment, and efficient inferencing.
  arXiv  Detail & Related papers  (2024-05-08T04:38:36Z)
• Slimmable Domain Adaptation [112.19652651687402]
  We introduce a simple framework, Slimmable Domain Adaptation, to improve cross-domain generalization with a weight-sharing model bank. Our framework surpasses other competing approaches by a very large margin on multiple benchmarks.
  arXiv  Detail & Related papers  (2022-06-14T06:28:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.