Pretrained Bayesian Non-parametric Knowledge Prior in Robotic Long-Horizon Reinforcement Learning

Reinforcement learning (RL) methods typically learn new tasks from scratch, often disregarding prior knowledge that could accelerate the learning process. While some methods incorporate previously learned skills, they usually rely on a fixed structure, such as a single Gaussian distribution, to define skill priors. This rigid assumption can restrict the diversity and flexibility of skills, particularly in complex, long-horizon tasks. In this work, we introduce a method that models potential primitive skill motions as having non-parametric properties with an unknown number of underlying features. We utilize a Bayesian non-parametric model, specifically Dirichlet Process Mixtures, enhanced with birth and merge heuristics, to pre-train a skill prior that effectively captures the diverse nature of skills. Additionally, the learned skills are explicitly trackable within the prior space, enhancing interpretability and control. By integrating this flexible skill prior into an RL framework, our approach surpasses existing methods in long-horizon manipulation tasks, enabling more efficient skill transfer and task success in complex environments. Our findings show that a richer, non-parametric representation of skill priors significantly improves both the learning and execution of challenging robotic tasks. All data, code, and videos are available at https://ghiara.github.io/HELIOS/.

SHAPE : Self-Improved Visual Preference Alignment by Iteratively Generating Holistic Winner

Large Visual Language Models (LVLMs) increasingly rely on preference alignment to ensure reliability, which steers the model behavior via preference fine-tuning on preference data structured as ``image - winner text - loser text'' triplets. However, existing approaches often suffer from limited diversity and high costs associated with human-annotated preference data, hindering LVLMs from fully achieving their intended alignment capabilities. We present \projectname, a self-supervised framework capable of transforming the already abundant supervised text-image pairs into holistic preference triplets for more effective and cheaper LVLM alignment, eliminating the need for human preference annotations. Our approach facilitates LVLMs in progressively enhancing alignment capabilities through iterative self-improvement. The key design rationale is to devise preference triplets where the winner text consistently improves in holisticness and outperforms the loser response in quality, thereby pushing the model to ``strive to the utmost'' of alignment performance through preference fine-tuning. For each given text-image pair, SHAPE introduces multiple visual augmentations and pairs them with a summarized text to serve as the winner response, while designating the original text as the loser response. Experiments across \textbf{12} benchmarks on various model architectures and sizes, including LLaVA and DeepSeek-VL, show that SHAPE achieves significant gains, for example, achieving +11.3\% on MMVet (comprehensive evaluation), +1.4\% on MMBench (general VQA), and +8.0\% on POPE (hallucination robustness) over baselines in 7B models. Notably, qualitative analyses confirm enhanced attention to visual details and better alignment with human preferences for holistic descriptions.

SecPE: Secure Prompt Ensembling for Private and Robust Large Language Models

With the growing popularity of LLMs among the general public users, privacy-preserving and adversarial robustness have become two pressing demands for LLM-based services, which have largely been pursued separately but rarely jointly. In this paper, to the best of our knowledge, we are among the first attempts towards robust and private LLM inference by tightly integrating two disconnected fields: private inference and prompt ensembling. The former protects users' privacy by encrypting inference data transmitted and processed by LLMs, while the latter enhances adversarial robustness by yielding an aggregated output from multiple prompted LLM responses. Although widely recognized as effective individually, private inference for prompt ensembling together entails new challenges that render the naive combination of existing techniques inefficient. To overcome the hurdles, we propose SecPE, which designs efficient fully homomorphic encryption (FHE) counterparts for the core algorithmic building blocks of prompt ensembling. We conduct extensive experiments on 8 tasks to evaluate the accuracy, robustness, and efficiency of SecPE. The results show that SecPE maintains high clean accuracy and offers better robustness at the expense of merely $2.5\%$ efficiency overhead compared to baseline private inference methods, indicating a satisfactory ``accuracy-robustness-efficiency'' tradeoff. For the efficiency of the encrypted Argmax operation that incurs major slowdown for prompt ensembling, SecPE is 35.4x faster than the state-of-the-art peers, which can be of independent interest beyond this work.

Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense

Large Language Models (LLMs) have showcased remarkable capabilities across various domains. Accompanying the evolving capabilities and expanding deployment scenarios of LLMs, their deployment challenges escalate due to their sheer scale and the advanced yet complex activation designs prevalent in notable model series, such as Llama, Gemma, and Mistral. These challenges have become particularly pronounced in resource-constrained deployment scenarios, where mitigating inference efficiency bottlenecks is imperative. Among various recent efforts, activation approximation has emerged as a promising avenue for pursuing inference efficiency, sometimes considered indispensable in applications such as private inference. Despite achieving substantial speedups with minimal impact on utility, even appearing sound and practical for real-world deployment, the safety implications of activation approximations remain unclear. In this work, we fill this critical gap in LLM safety by conducting the first systematic safety evaluation of activation approximations. Our safety vetting spans seven sota techniques across three popular categories, revealing consistent safety degradation across ten safety-aligned LLMs.

SEGMN: A Structure-Enhanced Graph Matching Network for Graph Similarity Learning

Graph similarity computation (GSC) aims to quantify the similarity score between two graphs. Although recent GSC methods based on graph neural networks (GNNs) take advantage of intra-graph structures in message passing, few of them fully utilize the structures presented by edges to boost the representation of their connected nodes. Moreover, previous cross-graph node embedding matching lacks the perception of the overall structure of the graph pair, due to the fact that the node representations from GNNs are confined to the intra-graph structure, causing the unreasonable similarity score. Intuitively, the cross-graph structure represented in the assignment graph is helpful to rectify the inappropriate matching. Therefore, we propose a structure-enhanced graph matching network (SEGMN). Equipped with a dual embedding learning module and a structure perception matching module, SEGMN achieves structure enhancement in both embedding learning and cross-graph matching. The dual embedding learning module incorporates adjacent edge representation into each node to achieve a structure-enhanced representation. The structure perception matching module achieves cross-graph structure enhancement through assignment graph convolution. The similarity score of each cross-graph node pair can be rectified by aggregating messages from structurally relevant node pairs. Experimental results on benchmark datasets demonstrate that SEGMN outperforms the state-of-the-art GSC methods in the GED regression task, and the structure perception matching module is plug-and-play, which can further improve the performance of the baselines by up to 25%.

Learning Task Planning from Multi-Modal Demonstration for Multi-Stage Contact-Rich Manipulation

Large Language Models (LLMs) have gained popularity in task planning for long-horizon manipulation tasks. To enhance the validity of LLM-generated plans, visual demonstrations and online videos have been widely employed to guide the planning process. However, for manipulation tasks involving subtle movements but rich contact interactions, visual perception alone may be insufficient for the LLM to fully interpret the demonstration. Additionally, visual data provides limited information on force-related parameters and conditions, which are crucial for effective execution on real robots. In this paper, we introduce an in-context learning framework that incorporates tactile and force-torque information from human demonstrations to enhance LLMs' ability to generate plans for new task scenarios. We propose a bootstrapped reasoning pipeline that sequentially integrates each modality into a comprehensive task plan. This task plan is then used as a reference for planning in new task configurations. Real-world experiments on two different sequential manipulation tasks demonstrate the effectiveness of our framework in improving LLMs' understanding of multi-modal demonstrations and enhancing the overall planning performance.

Real-time Contact State Estimation in Shape Control of Deformable Linear Objects under Small Environmental Constraints

Controlling the shape of deformable linear objects using robots and constraints provided by environmental fixtures has diverse industrial applications. In order to establish robust contacts with these fixtures, accurate estimation of the contact state is essential for preventing and rectifying potential anomalies. However, this task is challenging due to the small sizes of fixtures, the requirement for real-time performances, and the infinite degrees of freedom of the deformable linear objects. In this paper, we propose a real-time approach for estimating both contact establishment and subsequent changes by leveraging the dependency between the applied and detected contact force on the deformable linear objects. We seamlessly integrate this method into the robot control loop and achieve an adaptive shape control framework which avoids, detects and corrects anomalies automatically. Real-world experiments validate the robustness and effectiveness of our contact estimation approach across various scenarios, significantly increasing the success rate of shape control processes.

Flexible Informed Trees (FIT*): Adaptive Batch-Size Approach for Informed Sampling-Based Planner

In modern approaches to path planning and robot motion planning, anytime almost-surely asymptotically optimal planners dominate the benchmark of sample-based planners. A notable example is Batch Informed Trees (BIT*), where planners iteratively determine paths to groups of vertices within the exploration area. However, maintaining a consistent batch size is crucial for initial pathfinding and optimal performance, relying on effective task allocation. This paper introduces Flexible Informed Tree (FIT*), a novel planner integrating an adaptive batch-size method to enhance task scheduling in various environments. FIT* employs a flexible approach in adjusting batch sizes dynamically based on the inherent complexity of the planning domain and the current n-dimensional hyperellipsoid of the system. By constantly optimizing batch sizes, FIT* achieves improved computational efficiency and scalability while maintaining solution quality. This adaptive batch-size method significantly enhances the planner's ability to handle diverse and evolving problem domains. FIT* outperforms existing single-query, sampling-based planners on the tested problems in R^2 to R^8, and was demonstrated in real-world environments with KI-Fabrik/DARKO-Project Europe.

Contact-aware Shaping and Maintenance of Deformable Linear Objects With Fixtures

Studying the manipulation of deformable linear objects has significant practical applications in industry, including car manufacturing, textile production, and electronics automation. However, deformable linear object manipulation poses a significant challenge in developing planning and control algorithms, due to the precise and continuous control required to effectively manipulate the deformable nature of these objects. In this paper, we propose a new framework to control and maintain the shape of deformable linear objects with two robot manipulators utilizing environmental contacts. The framework is composed of a shape planning algorithm which automatically generates appropriate positions to place fixtures, and an object-centered skill engine which includes task and motion planning to control the motion and force of both robots based on the object status. The status of the deformable linear object is estimated online utilizing visual as well as force information. The framework manages to handle a cable routing task in real-world experiments with two Panda robots and especially achieves contact-aware and flexible clip fixing with challenging fixtures.

Comparative Study on Semi-supervised Learning Applied for Anomaly Detection in Hydraulic Condition Monitoring System

Condition-based maintenance is becoming increasingly important in hydraulic systems. However, anomaly detection for these systems remains challenging, especially since that anomalous data is scarce and labeling such data is tedious and even dangerous. Therefore, it is advisable to make use of unsupervised or semi-supervised methods, especially for semi-supervised learning which utilizes unsupervised learning as a feature extraction mechanism to aid the supervised part when only a small number of labels are available. This study systematically compares semi-supervised learning methods applied for anomaly detection in hydraulic condition monitoring systems. Firstly, thorough data analysis and feature learning were carried out to understand the open-sourced hydraulic condition monitoring dataset. Then, various methods were implemented and evaluated including traditional stand-alone semi-supervised learning models (e.g., one-class SVM, Robust Covariance), ensemble models (e.g., Isolation Forest), and deep neural network based models (e.g., autoencoder, Hierarchical Extreme Learning Machine (HELM)). Typically, this study customized and implemented an extreme learning machine based semi-supervised HELM model and verified its superiority over other semi-supervised methods. Extensive experiments show that the customized HELM model obtained state-of-the-art performance with the highest accuracy (99.5%), the lowest false positive rate (0.015), and the best F1-score (0.985) beating other semi-supervised methods.