论文分享 | 个性化相关研究进展

我们从2024-11-11到2024-12-10的29篇文章中精选出5篇优秀的工作分享给读者。

1. Robot Behavior Personalization from Sparse User Feedback
2. A Stack-Propagation Framework for Low-Resource Personalized Dialogue Generation
3. DreamBlend: Advancing Personalized Fine-tuning of Text-to-Image Diffusion Models
4. Personality-Guided Code Generation Using Large Language Models
5. ReMix: Training Generalized Person Re-identification on a Mixture of Data

1.Robot Behavior Personalization from Sparse User Feedback

Authors: Maithili Patel, Sonia Chernova

https://arxiv.org/abs/2410.19219

论文摘要

As service robots become more general-purpose, they will need to adapt to their users’ preferences over a large set of all possible tasks that they can perform. This includes preferences regarding which actions the users prefer to delegate to robots as opposed to doing themselves. Existing personalization approaches require task-specific data for each user. To handle diversity across all household tasks and users, and nuances in user preferences across tasks, we propose to learn a task adaptation function independently, which can be used in tandem with any universal robot policy to customize robot behavior. We create Task Adaptation using Abstract Concepts (TAACo) framework. TAACo can learn to predict the user’s preferred manner of assistance with any given task, by mediating reasoning through a representation composed of abstract concepts built based on user feedback. TAACo can generalize to an open set of household tasks from small amount of user feedback and explain its inferences through intuitive concepts. We evaluate our model on a dataset we collected of 5 people’s preferences, and show that TAACo outperforms GPT-4 by 16% and a rule-based system by 54%, on prediction accuracy, with 40 samples of user feedback.

论文简评

这篇论文提出了一种名为TAACo的方法，旨在通过利用抽象概念来个性化机器人行为，并解决家庭任务中适应不同用户偏好的挑战。该方法的目标是将用户的偏好抽象化，并提供对决策的直观解释。实验结果表明，TAACo的表现优于现有的GPT-4等模型，在预测准确性方面表现出色。

总体而言，这篇论文的主要贡献在于提出了一个创新的机制，用于基于抽象概念进行推理和解释。它有效地解决了机器人行为个性化的问题，并在实际应用中取得了显著成果。因此，TAACo是一个值得研究和推广的技术成果。

2.A Stack-Propagation Framework for Low-Resource Personalized Dialogue Generation

Authors: Haoyu Song, Wei-Nan Zhang, Kaiyan Zhang, Ting Liu

https://arxiv.org/abs/2410.20174

论文摘要

With the resurgent interest in building open-domain dialogue systems, the dialogue generation task has attracted increasing attention over the past few years. This task is usually formulated as a conditional generation problem, which aims to generate a natural and meaningful response given dialogue contexts and specific constraints, such as persona. Maintaining a consistent persona is essential for dialogue systems to gain trust from users. Although tremendous advancements have been made, traditional persona-based dialogue models are typically trained by leveraging a large number of persona-dense dialogue examples. Yet, such persona-dense training data are expensive to obtain, leading to a limited scale. This work presents a novel approach to learning from limited training examples by regarding consistency understanding as a regularization of response generation. To this end, we propose a novel stack-propagation framework for learning a generation and understanding pipeline. Specifically, the framework stacks a Transformer encoder and two Transformer decoders, where the first decoder models response generation and the second serves as a regularizer that jointly models response generation and consistency understanding. The proposed framework can benefit from the stacked encoder and decoders to learn from much smaller personalized dialogue data while maintaining competitive performance. Under different low-resource settings, subjective and objective evaluations prove that the stack-propagation framework outperforms strong baselines in response quality and persona consistency and largely overcomes the shortcomings of traditional models that heavily rely on persona-dense dialogue data.

论文简评

这篇论文提出的堆叠传播框架对于低资源个性化对话生成具有重要意义。该方法通过堆叠架构中的Transformer模型集成响应生成与一致性理解，有效地解决了个性化对话系统中维持一致性的挑战。作者声称，在有限训练数据条件下，他们的方法比现有模型表现出更好的性能，能够有效生成符合角色一致性的响应。此外，实验结果表明，该方法在各种低资源设置下的有效性得到了证明。综上所述，本文不仅提出了一个有效的解决方案，还展示了其在不同条件下的优越性，为未来的研究提供了宝贵的经验和启示。

3.DreamBlend: Advancing Personalized Fine-tuning of Text-to-Image Diffusion Models

Authors: Shwetha Ram, Tal Neiman, Qianli Feng, Andrew Stuart, Son Tran, Trishul Chilimbi

https://arxiv.org/abs/2411.19390

论文摘要

Given a small number of images of a subject, personalized image generation techniques can fine-tune large pre-trained text-to-image diffusion models to generate images of the subject in novel contexts, conditioned on text prompts. In doing so, a trade-off is made between prompt fidelity, subject fidelity, and diversity. As the pre-trained model is fine-tuned, earlier checkpoints synthesize images with low subject fidelity but high prompt fidelity and diversity. In contrast, later checkpoints generate images with low prompt fidelity and diversity but high subject fidelity. This inherent trade-off limits the prompt fidelity, subject fidelity, and diversity of generated images. In this work, we propose \emph{DreamBlend} to combine the prompt fidelity from earlier checkpoints and the subject fidelity from later checkpoints during inference. We perform a cross-attention guided image synthesis from a later checkpoint, guided by an image generated by an earlier checkpoint for the same prompt. This enables the generation of images with better subject fidelity, prompt fidelity, and diversity on challenging prompts, outperforming state-of-the-art fine-tuning methods.

论文简评

DreamBlend是文本到图像扩散模型中的一种方法，旨在通过融合早期和后期检查点提示的忠实度来增强个性化图像生成。该方法利用交叉注意力指导来缓解训练微调过程中通常面临的权衡问题，从而整体改善图像质量，在主题忠实度、提示忠实度以及多样性方面取得更好的效果。

论文的核心是提出了一个创新的方法来处理训练数据中的不均衡性，特别是早期和后期检查点之间的差异。这种融合策略不仅提高了模型在特定领域的表现，而且也促进了模型的泛化能力。实验结果表明，与现有方法相比，DreamBlend在量化评估指标和直观视觉体验上均取得了显著优势。

综上所述，这篇论文提出了一种新颖且有效的方法来解决文本到图像领域中的个性化图像生成问题，并提供了详细的实验验证和分析。通过对不同性能指标的比较，证明了其有效性，为未来的研究提供了有价值的参考。

4.Personality-Guided Code Generation Using Large Language Models

Authors: Yaoqi Guo, Zhenpeng Chen, Jie M. Zhang, Yang Liu, Yun Ma

https://arxiv.org/abs/2411.00006

论文摘要

Code generation, the automatic creation of source code from natural language descriptions, has garnered significant attention due to its potential to streamline software development. Inspired by research linking task-personality alignment with improved development outcomes, we conduct an empirical study on personality-guided code generation using large language models (LLMs). Specifically, we investigate how emulating personality traits appropriate to coding tasks affects LLM performance. We extensively evaluate this approach using seven widely adopted LLMs across four representative datasets. Our results show that personality guidance significantly enhances code generation accuracy, with improved pass rates in 23 out of 28 LLM-dataset combinations. Notably, in 11 cases, the improvement exceeds 5%, and in 5 instances, it surpasses 10%, with the highest gain reaching 12.9%. Additionally, personality guidance can be easily integrated with other prompting strategies to further boost performance.

论文简评

这篇关于如何使用大型语言模型（LLM）来指导代码生成的研究论文，揭示了人格导向的方法可以显著提高代码生成任务的准确率。研究者通过实验比较了七个不同的LLM和四种不同数据集的结果，显示了在编码任务中加入人格指导后代码生成的准确性显著提升。这种创新性方法不仅为软件开发领域提供了新的思路，而且验证了其有效性，表明了这项技术具有实用价值。此外，研究者还采用多样化的LLM和数据集，增强了研究成果的可信度，进一步证明了该方法的有效性。总之，这篇论文展示了利用LLM对代码生成进行个性化指导的可能性，并通过实证研究证实了这种方法的有效性。

5.ReMix: Training Generalized Person Re-identification on a Mixture of Data

Authors: Timur Mamedov, Anton Konushin, Vadim Konushin

https://arxiv.org/abs/2410.21938

论文摘要

Modern person re-identification (Re-ID) methods have a weak generalization ability and experience a major accuracy drop when capturing environment changes. This is because existing multi-camera Re-ID datasets are limited in size and diversity, since such data is difficult to obtain. At the same time, enormous volumes of unlabeled single-camera records are available. Such data can be easily collected, and therefore, it is more diverse. Currently, single-camera data is used only for self-supervised pre-training of Re-ID methods. However, the diversity of single-camera data is suppressed by fine-tuning on limited multi-camera data after pre-training. In this paper, we propose ReMix, a generalized Re-ID method jointly trained on a mixture of limited labeled multi-camera and large unlabeled single-camera data. Effective training of our method is achieved through a novel data sampling strategy and new loss functions that are adapted for joint use with both types of data. Experiments show that ReMix has a high generalization ability and outperforms state-of-the-art methods in generalizable person Re-ID. To the best of our knowledge, this is the first work that explores joint training on a mixture of multi-camera and single-camera data in person Re-ID.

论文简评

总的来说，该篇论文提出了一种名为"ReMix"的新方法，来解决人像识别（Re-ID）中跨环境泛化能力差的问题。论文的主要创新在于提出了联合训练策略，利用有限的标注多摄像头数据和大量未标注的单摄像头数据，以增强模型的泛化能力。此外，作者还对损失函数进行了适应性调整，以更好地处理多样化的数据特性。实验结果表明，相比现有方法，ReMix在不同场景下的表现优于其他竞争对手。综上所述，这篇论文为研究人像识别中的跨环境泛化问题提供了一个新的视角，并且其创新性的解决方案使其具有很高的应用价值。

我们欢迎您在评论区中留下宝贵的建议！包括但不限于：

• 可以提出推文中论文简评的不足！
• 可以分享最近更值得推荐的论文并给出理由！

推荐阅读