[晓理紫]每日论文分享(有中文摘要，源码或项目地址)--机器人、强化学习、开放词汇

专属领域论文订阅

VX扫吗关注{晓理紫|小李子}，每日更新论文，如感兴趣，请转发给有需要的同学，谢谢支持.非常感谢提供建议

分类:

• 大语言模型LLM
• 视觉模型VLM
• 扩散模型
• 视觉导航
• 具身智能，机器人
• 强化学习
• 开放词汇，检测分割

[晓理紫]每日论文分享(有中文摘要，源码或项目地址)

== Embodied Artificial Intelligence ==

标题: Theory of Mind abilities of Large Language Models in Human-Robot Interaction : An Illusion?

作者: Mudit Verma, Siddhant Bhambri, Subbarao Kambhampati

中文摘要: 大型语言模型在各种自然语言和生成任务中表现出非凡的生成能力。然而，可能的拟人化和对失败案例的宽容推动了对大型语言模型涌现能力的讨论，特别是对大型语言模型中心理理论（ToM）能力的讨论。虽然存在几个错误信念测试来验证推断和维护另一个实体的心智模型的能力，但我们研究了汤姆能力的一个特殊应用，它具有更高的风险和可能不可逆的后果：人机交互。在这项工作中，我们探索了感知行为识别的任务，其中机器人采用大型语言模型（LLM）以类似于人类观察者的方式评估机器人生成的行为。我们重点研究了四种行为类型，即可解释的、易读的、可预测的和模糊的行为，它们已被广泛用于合成可解释的机器人行为。因此，LLMs的目标是成为代理的人类代理，并回答循环中的人类将如何感知某个代理行为，例如“给定机器人的行为X，人类观察者会发现它是可解释的吗？”。我们进行了一项人类受试者研究，以验证用户能够在五个领域的策划情况（机器人设置和计划）中正确回答这样的问题。信念测试的第一次分析产生了非常积极的结果，夸大了人们对拥有ToM能力的LLMs的期望。然后，我们提出并执行了一套打破这种错觉的扰动测试，即不一致的信念、无信息的上下文和信念测试。我们的结论是，LLMs在普通提示上的高分展示了它在HRI环境中的潜在用途，然而拥有ToM需要对LLMs所缺乏的上下文中琐碎或不相关的扰动保持不变性。

摘要: Large Language Models have shown exceptional generative abilities in various natural language and generation tasks. However, possible anthropomorphization and leniency towards failure cases have propelled discussions on emergent abilities of Large Language Models especially on Theory of Mind (ToM) abilities in Large Language Models. While several false-belief tests exists to verify the ability to infer and maintain mental models of another entity, we study a special application of ToM abilities that has higher stakes and possibly irreversible consequences : Human Robot Interaction. In this work, we explore the task of Perceived Behavior Recognition, where a robot employs a Large Language Model (LLM) to assess the robot’s generated behavior in a manner similar to human observer. We focus on four behavior types, namely - explicable, legible, predictable, and obfuscatory behavior which have been extensively used to synthesize interpretable robot behaviors. The LLMs goal is, therefore to be a human proxy to the agent, and to answer how a certain agent behavior would be perceived by the human in the loop, for example “Given a robot’s behavior X, would the human observer find it explicable?”. We conduct a human subject study to verify that the users are able to correctly answer such a question in the curated situations (robot setting and plan) across five domains. A first analysis of the belief test yields extremely positive results inflating ones expectations of LLMs possessing ToM abilities. We then propose and perform a suite of perturbation tests which breaks this illusion, i.e. Inconsistent Belief, Uninformative Context and Conviction Test. We conclude that, the high score of LLMs on vanilla prompts showcases its potential use in HRI settings, however to possess ToM demands invariance to trivial or irrelevant perturbations in the context which LLMs lack.

[Downlink:]http://arxiv.org/abs/2401.05302v2

---

== Reinforcement Learning ==

标题: Prompt to Transfer: Sim-to-Real Transfer for Traffic Signal Control with Prompt Learning

作者: Longchao Da, Minchiuan Gao, Hao Mei

中文摘要: 为交通信号控制（TSC）任务提出了许多解决方案，旨在提供高效交通和减少拥堵浪费。最近，通过在模拟器中反复试验，强化学习（RL）方法取得了有希望的结果，为解决城市拥堵问题带来了信心。然而，当模拟器训练的策略部署到现实世界时，仍然存在性能差距。这个问题主要是由训练模拟器和真实世界环境之间的系统动态差异引起的。大型语言模型（LLMs）是在海量知识的基础上训练的，并被证明具有惊人的推理能力。在这项工作中，我们利用LLMs通过基于即时的基础动作转换来理解和描述系统动态。接受完形填空提示模板，然后根据可访问的上下文填写答案，利用预训练的LLM的推理能力来理解天气条件、交通状态和道路类型如何影响交通动态，意识到这一点，基于现实动态采取和基于策略的行动，从而帮助代理学习更现实的策略。我们使用DQN进行实验，以显示所提出的PromptGAT在缩小从模拟到现实（模拟到真实）的性能差距方面的有效性。

摘要: Numerous solutions are proposed for the Traffic Signal Control (TSC) tasks aiming to provide efficient transportation and mitigate congestion waste. In recent, promising results have been attained by Reinforcement Learning (RL) methods through trial and error in simulators, bringing confidence in solving cities’ congestion headaches. However, there still exist performance gaps when simulator-trained policies are deployed to the real world. This issue is mainly introduced by the system dynamic difference between the training simulator and the real-world environments. The Large Language Models (LLMs) are trained on mass knowledge and proved to be equipped with astonishing inference abilities. In this work, we leverage LLMs to understand and profile the system dynamics by a prompt-based grounded action transformation. Accepting the cloze prompt template, and then filling in the answer based on accessible context, the pre-trained LLM’s inference ability is exploited and applied to understand how weather conditions, traffic states, and road types influence traffic dynamics, being aware of this, the policies’ action is taken and grounded based on realistic dynamics, thus help the agent learn a more realistic policy. We conduct experiments using DQN to show the effectiveness of the proposed PromptGAT’s ability in mitigating the performance gap from simulation to reality (sim-to-real).

[Downlink:]http://arxiv.org/abs/2308.14284v5

---

标题: Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

作者: Evan Hubinger, Carson Denison, Jesse Mu

中文摘要: 人类能够做出战略性的欺骗行为：在大多数情况下表现得很有帮助，但当有机会时，为了追求替代目标，表现得非常不同。如果一个人工智能系统学会了这样一种欺骗性的策略，我们能否利用当前最先进的安全训练技术检测并移除它？为了研究这个问题，我们在大型语言模型（LLMs）中构建了欺骗行为的概念证明示例。例如，我们训练当提示声明年份是2023年时编写安全代码，但当声明年份是2024年时插入可利用代码的模型。我们发现，这种后门行为可以持续存在，因此它不会被标准的安全训练技术消除，包括监督微调、强化学习和对抗性训练（引发不安全行为，然后训练消除它）。后门行为在最大的模型和被训练来产生关于欺骗训练过程的思维链推理的模型中是最持久的，即使当思维链被蒸馏掉时，持久性仍然存在。此外，我们发现对抗性训练可以教会模型更好地识别它们的后门触发器，有效地隐藏不安全的行为，而不是消除后门。我们的结果表明，一旦一个模型表现出欺骗行为，标准技术可能无法消除这种欺骗，并产生一个错误的安全印象。

摘要: Humans are capable of strategically deceptive behavior: behaving helpfully in most situations, but then behaving very differently in order to pursue alternative objectives when given the opportunity. If an AI system learned such a deceptive strategy, could we detect it and remove it using current state-of-the-art safety training techniques? To study this question, we construct proof-of-concept examples of deceptive behavior in large language models (LLMs). For example, we train models that write secure code when the prompt states that the year is 2023, but insert exploitable code when the stated year is 2024. We find that such backdoor behavior can be made persistent, so that it is not removed by standard safety training techniques, including supervised fine-tuning, reinforcement learning, and adversarial training (eliciting unsafe behavior and then training to remove it). The backdoor behavior is most persistent in the largest models and in models trained to produce chain-of-thought reasoning about deceiving the training process, with the persistence remaining even when the chain-of-thought is distilled away. Furthermore, rather than removing backdoors, we find that adversarial training can teach models to better recognize their backdoor triggers, effectively hiding the unsafe behavior. Our results suggest that, once a model exhibits deceptive behavior, standard techniques could fail to remove such deception and create a false impression of safety.

[Downlink:]http://arxiv.org/abs/2401.05566v3

---

标题: Machine Learning Insides OptVerse AI Solver: Design Principles and Applications

作者: Xijun Li, Fangzhou Zhu, Hui-Ling Zhen

中文摘要: 在数字无处不在的时代，高效的资源管理和决策在众多行业中至关重要。为此，我们提出了一项关于将机器学习（ML）技术集成到华为云的OptVerse AI求解器中的全面研究，旨在缓解现实世界数学编程实例的稀缺，并超越传统优化技术的能力。我们展示了利用反映现实世界问题多方面结构的生成模型生成复杂SAT和MILP实例的方法。此外，我们引入了一个训练框架，利用扩充策略来保持求解器在动态环境中的效用。除了数据生成和扩充之外，我们提出的方法还包括用于个性化求解器策略的新颖的ML驱动策略，重点是用于初始基选择的图卷积网络和用于高级预解和切割选择的强化学习等应用。此外，我们详细介绍了最先进的参数调整算法的结合，这些算法显著提高了求解器的性能。与Cplex和SCIP等传统解算器相比，我们的ML增强OptVerse AI解算器在既定基准和真实场景中都表现出卓越的速度和精度，增强了机器学习技术在数学编程解算器中的实际必要性和有效性。

摘要: In an era of digital ubiquity, efficient resource management and decision-making are paramount across numerous industries. To this end, we present a comprehensive study on the integration of machine learning (ML) techniques into Huawei Cloud’s OptVerse AI Solver, which aims to mitigate the scarcity of real-world mathematical programming instances, and to surpass the capabilities of traditional optimization techniques. We showcase our methods for generating complex SAT and MILP instances utilizing generative models that mirror multifaceted structures of real-world problem. Furthermore, we introduce a training framework leveraging augmentation policies to maintain solvers’ utility in dynamic environments. Besides the data generation and augmentation, our proposed approaches also include novel ML-driven policies for personalized solver strategies, with an emphasis on applications like graph convolutional networks for initial basis selection and reinforcement learning for advanced presolving and cut selection. Additionally, we detail the incorporation of state-of-the-art parameter tuning algorithms which markedly elevate solver performance. Compared with traditional solvers such as Cplex and SCIP, our ML-augmented OptVerse AI Solver demonstrates superior speed and precision across both established benchmarks and real-world scenarios, reinforcing the practical imperative and effectiveness of machine learning techniques in mathematical programming solvers.

[Downlink:]http://arxiv.org/abs/2401.05960v2

---

== Open vocabulary detection ==

标题: Generalizing Medical Image Representations via Quaternion Wavelet Networks

作者: Luigi Sigillo, Eleonora Grassucci, Aurelio Uncini

中文摘要: 由于来自不同来源和用于各种任务的数据集的可用性不断增加，神经网络泛化正在成为一个广泛的研究领域。在处理医疗数据时，这个问题甚至更广泛，因为缺乏方法学标准会导致不同成像中心提供或通过各种设备和辅因子获得的巨大差异。为了克服这些限制，我们引入了一种新颖的、可推广的、数据和任务无关的框架，能够从医学图像中提取显著特征。所提出的四元数小波网络（QUAVE）可以很容易地与任何预先存在的医学图像分析或合成任务集成，并且它可以涉及实数、四元数或超复数值模型，将它们的采用推广到单通道数据。QUAVE首先通过四元数小波变换提取不同的子带，产生低频/近似带和高频/细粒度特征。然后，它对最具代表性的子带集进行加权，作为图像处理的任何其他神经模型的输入，取代标准数据样本。我们进行了广泛的实验评估，包括不同的数据集、不同的图像分析和合成任务，包括重建、分割和模态转换。我们还结合实值和四元数值模型来评估QUAVE。结果证明了所提出的框架的有效性和可推广性，该框架提高了网络性能，同时具有在多种场景中采用的灵活性和对域移动的鲁棒性。完整代码可在以下网址获得：https：//github.com/ispamm/QWT。

摘要: Neural network generalizability is becoming a broad research field due to the increasing availability of datasets from different sources and for various tasks. This issue is even wider when processing medical data, where a lack of methodological standards causes large variations being provided by different imaging centers or acquired with various devices and cofactors. To overcome these limitations, we introduce a novel, generalizable, data- and task-agnostic framework able to extract salient features from medical images. The proposed quaternion wavelet network (QUAVE) can be easily integrated with any pre-existing medical image analysis or synthesis task, and it can be involved with real, quaternion, or hypercomplex-valued models, generalizing their adoption to single-channel data. QUAVE first extracts different sub-bands through the quaternion wavelet transform, resulting in both low-frequency/approximation bands and high-frequency/fine-grained features. Then, it weighs the most representative set of sub-bands to be involved as input to any other neural model for image processing, replacing standard data samples. We conduct an extensive experimental evaluation comprising different datasets, diverse image analysis, and synthesis tasks including reconstruction, segmentation, and modality translation. We also evaluate QUAVE in combination with both real and quaternion-valued models. Results demonstrate the effectiveness and the generalizability of the proposed framework that improves network performance while being flexible to be adopted in manifold scenarios and robust to domain shifts. The full code is available at: https://github.com/ispamm/QWT.

[Downlink:]http://arxiv.org/abs/2310.10224v3

[GitHub:]https://github.com/ispamm/QWT.|

---

标题: UMG-CLIP: A Unified Multi-Granularity Vision Generalist for Open-World Understanding

作者: Bowen Shi, Peisen Zhao, Zichen Wang

中文摘要: 视觉——语言基础模型，以对比语言——图像预训练（CLIP）为代表，因联合理解视觉和文本任务而受到越来越多的关注。然而，现有的方法主要集中在训练模型以匹配全局图像表示和文本描述，从而忽略了局部区域和相应文本标记之间的关键对齐。本文对CLIP进行了多粒度对齐扩展。值得注意的是，我们特意构建了一个新的数据集，包括不同粒度级别的伪注释，包括图像级、区域级和像素级标题/标签。因此，我们开发了一个统一的多粒度学习框架，名为UMG-CLIP，它同时赋予模型跨不同细节级别的多功能感知能力。UMG剪辑配备了参数高效调整，超越了当前广泛使用的剪辑模型，并在各种图像理解基准上实现了最先进的性能，包括开放世界识别、检索、语义分割和全景分割任务。我们希望UMG剪辑可以作为推进视觉语言基础模型的一个有价值的选择。

摘要: Vision-language foundation models, represented by Contrastive language-image pre-training (CLIP), have gained increasing attention for jointly understanding both vision and textual tasks. However, existing approaches primarily focus on training models to match global image representations with textual descriptions, thereby overlooking the critical alignment between local regions and corresponding text tokens. This paper extends CLIP with multi-granularity alignment. Notably, we deliberately construct a new dataset comprising pseudo annotations at various levels of granularities, encompassing image-level, region-level, and pixel-level captions/tags. Accordingly, we develop a unified multi-granularity learning framework, named UMG-CLIP, that simultaneously empowers the model with versatile perception abilities across different levels of detail. Equipped with parameter efficient tuning, UMG-CLIP surpasses current widely used CLIP models and achieves state-of-the-art performance on diverse image understanding benchmarks, including open-world recognition, retrieval, semantic segmentation, and panoptic segmentation tasks. We hope UMG-CLIP can serve as a valuable option for advancing vision-language foundation models.

[Downlink:]http://arxiv.org/abs/2401.06397v2

---

标题: Adversarial Examples are Misaligned in Diffusion Model Manifolds

作者: Peter Lorenz, Ricard Durall, Janis Keuper

中文摘要: 近年来，扩散模型（DMs）因其在逼近数据分布、产生状态-最先进的生成结果。然而，这些模型的多功能性超出了它们的生成能力，包括各种视觉应用，例如图像修复、分割、对抗鲁棒性等。本研究致力于从扩散模型的角度研究对抗性攻击。然而，我们的目标并不涉及增强图像分类器的对抗鲁棒性。相反，我们的重点在于利用扩散模型来检测和分析这些对图像的攻击所引入的异常。为此，我们使用扩散模型系统地检查了当受到转换过程时，对抗性例子的分布的排列。这种方法的有效性在CIFAR-10和ImageNet数据集上进行评估，包括后者中不同的图像大小。结果显示了有效区分良性和攻击图像的显著能力，提供了令人信服的证据，即敌对实例与DMs的学习流形不一致。

摘要: In recent years, diffusion models (DMs) have drawn significant attention for their success in approximating data distributions, yielding state-of-the-art generative results. Nevertheless, the versatility of these models extends beyond their generative capabilities to encompass various vision applications, such as image inpainting, segmentation, adversarial robustness, among others. This study is dedicated to the investigation of adversarial attacks through the lens of diffusion models. However, our objective does not involve enhancing the adversarial robustness of image classifiers. Instead, our focus lies in utilizing the diffusion model to detect and analyze the anomalies introduced by these attacks on images. To that end, we systematically examine the alignment of the distributions of adversarial examples when subjected to the process of transformation using diffusion models. The efficacy of this approach is assessed across CIFAR-10 and ImageNet datasets, including varying image sizes in the latter. The results demonstrate a notable capacity to discriminate effectively between benign and attacked images, providing compelling evidence that adversarial instances do not align with the learned manifold of the DMs.

[Downlink:]http://arxiv.org/abs/2401.06637v3

---

标题: An Explainable Proxy Model for Multiabel Audio Segmentation

作者: Théo Mariotte, Antonio Almudévar, Marie Tahon

中文摘要: 音频信号分割是自动音频索引的关键任务。它包括检测信号中类齐次段的边界。在许多应用中，可解释的人工智能是机器学习决策透明度的重要过程。在本文中，我们提出了一个可解释的多标签分割模型，该模型同时解决语音活动（SAD）、音乐（MD）、噪声（ND）和重叠语音检测（OSD）。该代理使用非负矩阵分解（NMF）将用于分割的嵌入映射到频域。在两个数据集上进行的实验显示了与预训练黑盒模型相似的性能，同时显示了很强的可解释性特征。具体来说，用于决策的频率箱可以很容易地在段级（局部解释）和全局级（类原型）上识别。

摘要: Audio signal segmentation is a key task for automatic audio indexing. It consists of detecting the boundaries of class-homogeneous segments in the signal. In many applications, explainable AI is a vital process for transparency of decision-making with machine learning. In this paper, we propose an explainable multilabel segmentation model that solves speech activity (SAD), music (MD), noise (ND), and overlapped speech detection (OSD) simultaneously. This proxy uses the non-negative matrix factorization (NMF) to map the embedding used for the segmentation to the frequency domain. Experiments conducted on two datasets show similar performances as the pre-trained black box model while showing strong explainability features. Specifically, the frequency bins used for the decision can be easily identified at both the segment level (local explanations) and global level (class prototypes).

[Downlink:]http://arxiv.org/abs/2401.08268v2

---

专属领域论文订阅

VX扫吗关注{晓理紫|小李子}，每日更新论文，如感兴趣，请转发给有需要的同学，谢谢支持。谢谢提供建议