用大模型驱动自动进化，为触觉感知设计更优神经网络。
TacEvo: Self-Evolving Architecture Discovery for Robotic Tactile Perception via LLM-Driven Quality-Diversity Search
- 大模型生成代码变异和交叉，结合多样性搜索优化网络结构。
- 20代内性能提升56.1%（力估计）和96.1%（纹理分类），94.5%可训练率。
- 适合机器人触觉、自适应架构设计等研究者使用。
基于视觉的触觉传感将接触引起的表面形变转化为图像，使机器人能够推断接触力与精细表面纹理，这是传统视觉无法获取的信息。然而，触觉图像具有传感器和物理特性依赖性，有效网络结构通常需要专家直觉和大量人工迭代。现有神经架构搜索（NAS）方法虽能减轻负担，但计算开销大且受限于手工设计的搜索空间，限制了架构新颖性和多样性。我们提出TacEvo，一种自进化架构发现框架，通过下游反馈持续改进网络设计。TacEvo利用大语言模型（LLM）生成代码级变异和交叉，结合MAP-Elites质量-多样性循环，保留多样化的优秀架构，并优先复用持续产生改进的提示。探索由两个行为描述符引导：架构多样性与效率比，鼓励覆盖结构变化与计算量权衡。在ViTacTip力回归与条纹分类任务中，TacEvo实现96.0%/94.5%的自主生成可训练率，20代内验证适应度提升56.1%/96.1%。20次种子的高保真评估显示，其在力预测上达到专家基线水平，在细粒度条纹分类上表现更优。结果表明，大模型驱动的自进化搜索是人工智能辅助特定机器人感知科学发现的可行范式。
原文摘要 · Abstract (English)
Vision-based tactile sensing converts contact-induced surface deformation into images, enabling robots to infer contact forces and fine surface textures that are not accessible through conventional vision alone. However, tactile images are sensor- and physics-specific, so effective architectures often require expert intuition and extensive manual iteration. Existing neural architecture search (NAS) pipelines can reduce this burden, but they are often computationally expensive and restricted to hand-designed search spaces, which limits architectural novelty and diversity. We introduce TacEvo, a self-evolving architecture discovery framework that improves network designs from downstream feedback. TacEvo uses an LLM to generate code-level mutations and crossovers, and a MAP-Elites quality-diversity loop that preserves diverse elite architectures while preferentially reusing prompts that consistently yield improvements. Exploration is guided by two behavioural descriptors, Architectural Diversity and Efficiency Ratio, which encourage coverage across structural variations and compute-size trade-offs. On ViTacTip force regression and grating classification, TacEvo achieves high autonomous generation reliability (96.0%/94.5% trainable) and improves best validation fitness over 20 generations by 56.1%/96.1%. In a 20-seed post-search high-fidelity evaluation, TacEvo matches the expert baseline on force prediction and outperforms it on fine-grained grating classification. These results suggest that LLM-driven self-evolving search constitutes a practical paradigm for AI-assisted scientific discovery in specialised robotic sensing.
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。