When Better Eyes Lead to Blindness: A Diagnostic Study of the Information Bottleneck in CNN-LSTM Image Captioning Models

• URL: http://arxiv.org/abs/2507.18788v2
• Date: Wed, 20 Aug 2025 19:21:58 GMT
• Title: When Better Eyes Lead to Blindness: A Diagnostic Study of the Information Bottleneck in CNN-LSTM Image Captioning Models
• Abstract summary: This paper documents a systematic, iterative development of foundational image captioning models.<n>It progresses from a simple CNN-LSTM encoder-decoder to a competitive attention-based system.<n>The experiments chart the impact of architectural enhancements and demonstrate a key finding within the classic CNN-LSTM paradigm.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Image captioning, situated at the intersection of computer vision and natural language processing, requires a sophisticated understanding of both visual scenes and linguistic structure. While modern approaches are dominated by large-scale Transformer architectures, this paper documents a systematic, iterative development of foundational image captioning models, progressing from a simple CNN-LSTM encoder-decoder to a competitive attention-based system. This paper presents a series of five models, beginning with Genesis and concluding with Nexus, an advanced model featuring an EfficientNetV2B3 backbone and a dynamic attention mechanism. The experiments chart the impact of architectural enhancements and demonstrate a key finding within the classic CNN-LSTM paradigm: merely upgrading the visual backbone without a corresponding attention mechanism can degrade performance, as the single-vector bottleneck cannot transmit the richer visual detail. This insight validates the architectural shift to attention. Trained on the MS COCO 2017 dataset, the final model, Nexus, achieves a BLEU-4 score of 31.4, surpassing several foundational benchmarks and validating the iterative design process. This work provides a clear, replicable blueprint for understanding the core architectural principles that underpin modern vision-language tasks.

Related papers

• Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification [80.62512020268626]
  UniAR is a unified autoregressive framework where a single visual tokenizer serves as the key bridge between understanding and generation.<n>UniAR adapts a pretrained vision encoder with multi-level feature fusion and a lookup-free bitwise quantization scheme.<n>A diffusion-based visual decoder operates on discrete visual tokens to decode high-fidelity images.
  arXiv  Detail & Related papers  (2026-06-16T17:59:22Z)
• Vision Inference Former: Sustaining Visual Consistency in Multimodal Large Language Models [58.9387276447485]
  Vision Inference Former (VIF) is a lightweight architectural module that establishes a bridge between pure visual representations and the model's output space.<n>We conduct experiments on 14 benchmark tasks covering general reasoning, OCR, table understanding, vision-centric evaluation, and hallucination.
  arXiv  Detail & Related papers  (2026-05-18T10:04:22Z)
• Attention Guided Alignment in Efficient Vision-Language Models [56.20286899428444]
  Large Vision-Language Models (VLMs) rely on effective multimodal alignment between pre-trained vision encoders and Large Language Models (LLMs)<n>This paper presents a comprehensive analysis of attention patterns in efficient VLMs.<n>We introduce Attention-Guided Efficient Vision-Language Models (AGE-VLM), a novel framework that enhances visual grounding through interleaved cross-attention layers.
  arXiv  Detail & Related papers  (2025-11-21T21:36:48Z)
• The Geometry of Cortical Computation: Manifold Disentanglement and Predictive Dynamics in VCNet [0.0]
  This paper introduces Visual Cortex Network (VCNet), a novel neural network architecture.<n> VCNet is framed as a geometric framework that emulates key biological mechanisms.<n>We show that VCNet achieves state-of-the-art accuracy of 92.1% on Spots-10 and 74.4% on the light field dataset.
  arXiv  Detail & Related papers  (2025-08-05T01:52:42Z)
• Foundations and Models in Modern Computer Vision: Key Building Blocks in Landmark Architectures [34.542592986038265]
  This report analyzes the evolution of key design patterns in computer vision by examining six influential papers.<n>We review ResNet, which introduced residual connections to overcome the vanishing gradient problem.<n>We examine the Vision Transformer (ViT), which established a new paradigm by applying the Transformer architecture to sequences of image patches.
  arXiv  Detail & Related papers  (2025-07-31T09:08:11Z)
• Language Embedding Meets Dynamic Graph: A New Exploration for Neural Architecture Representation Learning [38.323486764309]
  We introduce LeDG-Former, an innovative framework that addresses limitations through the synergistic integration of language-based semantic embedding and dynamic graph representation learning.<n>Specifically, we propose a language embedding framework where both neural architectures and hardware platform specifications are projected into a unified semantic space.<n>Our framework achieves superior performance on the cell-structured NAS-Bench-101 and NAS-Bench-201 datasets.
  arXiv  Detail & Related papers  (2025-06-09T13:20:02Z)
• Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Models [18.02840698188587]
  We propose a novel kernel-based method to align CLIP's visual representation with that of DINOv2.<n>Our image-only alignment fine-tuning exhibits significant improvements in zero-shot object recognition, fine-grained spatial reasoning.
  arXiv  Detail & Related papers  (2025-06-03T07:44:43Z)
• Weaving Context Across Images: Improving Vision-Language Models through Focus-Centric Visual Chains [31.828341309787042]
  Vision-language models (VLMs) achieve remarkable success in single-image tasks.<n>Real-world scenarios often involve intricate multi-image inputs, leading to a notable performance decline.<n>We propose Focus-Centric Visual Chain, a novel paradigm that enhances VLMs'perception, comprehension, and reasoning abilities in multi-image scenarios.
  arXiv  Detail & Related papers  (2025-04-28T19:02:18Z)
• Text Speaks Louder than Vision: ASCII Art Reveals Textual Biases in Vision-Language Models [93.46875303598577]
  Vision-language models (VLMs) have advanced rapidly in processing multimodal information, but their ability to reconcile conflicting signals remains underexplored.<n>This work investigates how VLMs process ASCII art, a unique medium where textual elements collectively form visual patterns, potentially creating semantic-visual conflicts.
  arXiv  Detail & Related papers  (2025-04-02T10:47:07Z)
• ELIP: Enhanced Visual-Language Foundation Models for Image Retrieval [83.01358520910533]
  We introduce a new framework that can boost the performance of large-scale pre-trained vision- curation models.<n>The approach, Enhanced Language-Image Pre-training (ELIP), uses the text query, via a simple mapping network, to predict a set of visual prompts.<n>ELIP can easily be applied to the commonly used CLIP, SigLIP and BLIP-2 networks.
  arXiv  Detail & Related papers  (2025-02-21T18:59:57Z)
• DINOv2 Meets Text: A Unified Framework for Image- and Pixel-Level Vision-Language Alignment [20.953645420787527]
  We train a CLIP-like model with only a fraction of the computational cost compared to CLIP.<n>We achieve state-of-the-art results in zero-shot classification and open-vocabulary semantic segmentation.
  arXiv  Detail & Related papers  (2024-12-20T20:46:48Z)
• Can Graph Neural Networks Learn Language with Extremely Weak Text Supervision? [62.12375949429938]
  We propose a multi-modal prompt learning paradigm to adapt pre-trained Graph Neural Networks to downstream tasks and data.<n>Our new paradigm embeds the graphs directly in the same space as the Large Language Models (LLMs) by learning both graph prompts and text prompts simultaneously.<n>We build the first CLIP-style zero-shot classification prototype that can generalize GNNs to unseen classes with extremely weak text supervision.
  arXiv  Detail & Related papers  (2024-12-11T08:03:35Z)
• Emergent Visual-Semantic Hierarchies in Image-Text Representations [13.300199242824934]
  We study the knowledge of existing foundation models, finding that they exhibit emergent understanding of visual-semantic hierarchies. We propose the Radial Embedding (RE) framework for probing and optimizing hierarchical understanding.
  arXiv  Detail & Related papers  (2024-07-11T14:09:42Z)
• Learning Object-Centric Representation via Reverse Hierarchy Guidance [73.05170419085796]
  Object-Centric Learning (OCL) seeks to enable Neural Networks to identify individual objects in visual scenes. RHGNet introduces a top-down pathway that works in different ways in the training and inference processes. Our model achieves SOTA performance on several commonly used datasets.
  arXiv  Detail & Related papers  (2024-05-17T07:48:27Z)
• GiT: Towards Generalist Vision Transformer through Universal Language Interface [94.33443158125186]
  This paper proposes a simple, yet effective framework, called GiT, simultaneously applicable for various vision tasks only with a vanilla ViT. GiT is a multi-task visual model, jointly trained across five representative benchmarks without task-specific fine-tuning.
  arXiv  Detail & Related papers  (2024-03-14T13:47:41Z)
• Hierarchical Graph Pattern Understanding for Zero-Shot VOS [102.21052200245457]
  This paper proposes a new hierarchical graph neural network (GNN) architecture for zero-shot video object segmentation (ZS-VOS) Inspired by the strong ability of GNNs in capturing structural relations, HGPU innovatively leverages motion cues (ie, optical flow) to enhance the high-order representations from the neighbors of target frames.
  arXiv  Detail & Related papers  (2023-12-15T04:13:21Z)
• Divert More Attention to Vision-Language Object Tracking [87.31882921111048]
  We argue that the lack of large-scale vision-language annotated videos and ineffective vision-language interaction learning motivate us to design more effective vision-language representation for tracking. Particularly, in this paper, we first propose a general attribute annotation strategy to decorate videos in six popular tracking benchmarks, which contributes a large-scale vision-language tracking database with more than 23,000 videos. We then introduce a novel framework to improve tracking by learning a unified-adaptive VL representation, where the cores are the proposed asymmetric architecture search and modality mixer (ModaMixer)
  arXiv  Detail & Related papers  (2023-07-19T15:22:06Z)
• Vision Transformers: From Semantic Segmentation to Dense Prediction [139.15562023284187]
  We explore the global context learning potentials of vision transformers (ViTs) for dense visual prediction. Our motivation is that through learning global context at full receptive field layer by layer, ViTs may capture stronger long-range dependency information. We formulate a family of Hierarchical Local-Global (HLG) Transformers, characterized by local attention within windows and global-attention across windows in a pyramidal architecture.
  arXiv  Detail & Related papers  (2022-07-19T15:49:35Z)
• mPLUG: Effective and Efficient Vision-Language Learning by Cross-modal Skip-connections [104.14624185375897]
  mPLUG is a new vision-language foundation model for both cross-modal understanding and generation. It achieves state-of-the-art results on a wide range of vision-language downstream tasks, such as image captioning, image-text retrieval, visual grounding and visual question answering.
  arXiv  Detail & Related papers  (2022-05-24T11:52:06Z)
• The Mind's Eye: Visualizing Class-Agnostic Features of CNNs [92.39082696657874]
  We propose an approach to visually interpret CNN features given a set of images by creating corresponding images that depict the most informative features of a specific layer. Our method uses a dual-objective activation and distance loss, without requiring a generator network nor modifications to the original model.
  arXiv  Detail & Related papers  (2021-01-29T07:46:39Z)
• Embedded Encoder-Decoder in Convolutional Networks Towards Explainable AI [0.0]
  This paper proposes a new explainable convolutional neural network (XCNN) which represents important and driving visual features of stimuli. The experimental results on the CIFAR-10, Tiny ImageNet, and MNIST datasets showed the success of our algorithm (XCNN) to make CNNs explainable.
  arXiv  Detail & Related papers  (2020-06-19T15:49:39Z)

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.