Recent advances in large language models (LLMs) provide robots with contextual reasoning abilities to comprehend human instructions. Yet, current LLM-enabled robots typically depend on cloud-based models or high-performance computing infrastructure, which limit their deployment on robots under unreliable internet environments or with constrained computational resources, such as UAVs and small ground vehicles. Thus, deploying fine-tuned small language models (SLMs) that support onboard deployment offers a promising alternative. This paper introduces Ro-SLM, a framework that enables reliable SLM-driven robot operation by distilling LLMs’ knowledge and reasoning. Ro-SLM starts from dataset synthesis by leveraging LLMs to generate diverse task instructions, produce corresponding ground truth code with minimal human assistance, and augment instructions into real-world application scenarios. Ro-SLM is then fine-tuned with the dataset, in which LLM serves as a reward function to guide the training. Extensive experiments on UAV operation tasks demonstrate that Ro-SLM improves the performance of SLM from being incapable of supporting robotic task planning and code generation to achieving performance that approaches LLM.

State-of-the-art NLP models have demonstrated exceptional performance across various tasks, including sentiment analysis. However, concerns have been raised about their robustness and susceptibility to systematic biases in both training and test data, which may lead to performance challenges when these models encounter out-of-distribution data in real-world applications. Although various data augmentation and adversarial perturbation techniques have shown promise in tackling these issues, prior methods such as word embedding perturbation or synonymous sentence expansion have failed to mitigate the spurious association problem inherent in the original data. Recent counterfactual augmentation methods have attempted to tackle this issue, but they have been limited by rigid rules, resulting in inconsistent context and disrupted semantics. In response to these challenges, we introduce a diffusion-based counterfactual data augmentation (DCA) framework. It utilizes an antonymous paradigm to guide the continuous diffusion model and employs reinforcement learning in combination with contrastive learning to optimize algorithms for generating counterfactual samples with high diversity and quality. Furthermore, we use a dual sentiment classifier to validate the generated antonymous samples and subsequently perform sentiment classification. Our experiments on four benchmark datasets demonstrate that DCA achieves state-of-the-art performance in sentiment classification tasks.

Stance detection aims to identify whether the author of an opinionated text is in favor of, against, or neutral towards a given target. Remarkable success has been achieved when sufficient labeled training data is available. However, it is labor-intensive to annotate sufficient data and train the model for every new target.Therefore, zero-shot stance detection, aiming at identifying stances of unseen targets with seen targets, has gradually attracted attention. Among them, one of the important challenges is to reduce the domain transfer between seen and unseen targets. To tackle this problem, we propose a generative data augmentation approach to generate training samples containing targets and stances for testing data, and map the real samples and generated synthetic samples into the same embedding space with contrastive learning, then perform the final classification based on the augmented data. We evaluate our proposed model on two benchmark datasets. Experimental results show that our approach achieves state-of-the-art performance on most topics in the task of zero-shot stance detection.