升级视觉主干网络却无注意力机制，反而让图像描述模型更差。
When Better Eyes Lead to Blindness: A Diagnostic Study of the Information Bottleneck in CNN-LSTM Image Captioning Models
- 用五种模型迭代对比，发现瓶颈在单一向量传递信息
- 最终模型在MS COCO上达到BLEU-4 31.4，超越多个基础基准
- 适合想理解视觉-语言模型设计原理的研究者
图像描述任务融合计算机视觉与自然语言处理，需同时理解视觉场景与语言结构。本文系统性地推进了从基础CNN-LSTM到先进注意力模型的五阶段发展过程。从Genesis开始，逐步构建至Nexus——采用EfficientNetV2B3主干和动态注意力机制的先进模型。实验表明，在经典CNN-LSTM范式中，仅提升视觉主干而未引入注意力机制，反而会因单向量信息瓶颈导致性能下降。这一发现验证了从固定编码转向注意力机制的必要性。模型在MS COCO 2017数据集上训练，最终实现BLEU-4 31.4，超越多个基础基准，证明了迭代设计的有效性。本研究为理解现代视觉-语言任务的核心架构原则提供了清晰可复现的蓝图。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。