Two Is Better Than One: Rotations Scale LoRAs

Scaling Low-Rank Adaptation (LoRA)-based Mixture-of-Experts (MoE) facilitates large language models (LLMs) to efficiently adapt to diverse tasks. However, traditional gating mechanisms that route inputs to the best experts may fundamentally hinder LLMs' scalability, leading to poor generalization and underfitting issues. We identify that the root cause lies in the restricted expressiveness of existing weighted-sum mechanisms, both within and outside the convex cone of LoRA representations. This motivates us to propose RadarGate, a novel geometrically inspired gating method that introduces rotational operations of LoRAs representations to boost the expressiveness and facilitate richer feature interactions among multiple LoRAs for scalable LLMs. Specifically, we first fuse each LoRA representation to other LoRAs using a learnable component and then feed the output to a rotation matrix. This matrix involves learnable parameters that define the relative angular relationship between LoRA representations. Such a simple yet effective mechanism provides an extra degree of freedom, facilitating the learning of cross-LoRA synergies and properly tracking the challenging poor generalization and underfitting issues as the number of LoRA grows. Extensive experiments on 6 public benchmarks across 21 tasks show the effectiveness of our RadarGate for scaling LoRAs. We also provide valuable insights, revealing that the rotations to each pair of representations are contrastive, encouraging closer alignment of semantically similar representations during geometrical transformation while pushing distance ones further apart. We will release our code to the community.

Not All Tokens Are What You Need In Thinking

Modern reasoning models, such as OpenAI's o1 and DeepSeek-R1, exhibit impressive problem-solving capabilities but suffer from critical inefficiencies: high inference latency, excessive computational resource consumption, and a tendency toward overthinking -- generating verbose chains of thought (CoT) laden with redundant tokens that contribute minimally to the final answer. To address these issues, we propose Conditional Token Selection (CTS), a token-level compression framework with a flexible and variable compression ratio that identifies and preserves only the most essential tokens in CoT. CTS evaluates each token's contribution to deriving correct answers using conditional importance scoring, then trains models on compressed CoT. Extensive experiments demonstrate that CTS effectively compresses long CoT while maintaining strong reasoning performance. Notably, on the GPQA benchmark, Qwen2.5-14B-Instruct trained with CTS achieves a 9.1% accuracy improvement with 13.2% fewer reasoning tokens (13% training token reduction). Further reducing training tokens by 42% incurs only a marginal 5% accuracy drop while yielding a 75.8% reduction in reasoning tokens, highlighting the prevalence of redundancy in existing CoT.

MORALISE: A Structured Benchmark for Moral Alignment in Visual Language Models

Warning: This paper contains examples of harmful language and images. Reader discretion is advised. Recently, vision-language models have demonstrated increasing influence in morally sensitive domains such as autonomous driving and medical analysis, owing to their powerful multimodal reasoning capabilities. As these models are deployed in high-stakes real-world applications, it is of paramount importance to ensure that their outputs align with human moral values and remain within moral boundaries. However, existing work on moral alignment either focuses solely on textual modalities or relies heavily on AI-generated images, leading to distributional biases and reduced realism. To overcome these limitations, we introduce MORALISE, a comprehensive benchmark for evaluating the moral alignment of vision-language models (VLMs) using diverse, expert-verified real-world data. We begin by proposing a comprehensive taxonomy of 13 moral topics grounded in Turiel's Domain Theory, spanning the personal, interpersonal, and societal moral domains encountered in everyday life. Built on this framework, we manually curate 2,481 high-quality image-text pairs, each annotated with two fine-grained labels: (1) topic annotation, identifying the violated moral topic(s), and (2) modality annotation, indicating whether the violation arises from the image or the text. For evaluation, we encompass two tasks, \textit{moral judgment} and \textit{moral norm attribution}, to assess models' awareness of moral violations and their reasoning ability on morally salient content. Extensive experiments on 19 popular open- and closed-source VLMs show that MORALISE poses a significant challenge, revealing persistent moral limitations in current state-of-the-art models. The full benchmark is publicly available at https://huggingface.co/datasets/Ze1025/MORALISE.

Enhancing Knowledge Graph Completion with GNN Distillation and Probabilistic Interaction Modeling

Knowledge graphs (KGs) serve as fundamental structures for organizing interconnected data across diverse domains. However, most KGs remain incomplete, limiting their effectiveness in downstream applications. Knowledge graph completion (KGC) aims to address this issue by inferring missing links, but existing methods face critical challenges: deep graph neural networks (GNNs) suffer from over-smoothing, while embedding-based models fail to capture abstract relational features. This study aims to overcome these limitations by proposing a unified framework that integrates GNN distillation and abstract probabilistic interaction modeling (APIM). GNN distillation approach introduces an iterative message-feature filtering process to mitigate over-smoothing, preserving the discriminative power of node representations. APIM module complements this by learning structured, abstract interaction patterns through probabilistic signatures and transition matrices, allowing for a richer, more flexible representation of entity and relation interactions. We apply these methods to GNN-based models and the APIM to embedding-based KGC models, conducting extensive evaluations on the widely used WN18RR and FB15K-237 datasets. Our results demonstrate significant performance gains over baseline models, showcasing the effectiveness of the proposed techniques. The findings highlight the importance of both controlling information propagation and leveraging structured probabilistic modeling, offering new avenues for advancing knowledge graph completion. And our codes are available at https://anonymous.4open.science/r/APIM_and_GNN-Distillation-461C.

Flying through cluttered and dynamic environments with LiDAR

Navigating unmanned aerial vehicles (UAVs) through cluttered and dynamic environments remains a significant challenge, particularly when dealing with fast-moving or sudden-appearing obstacles. This paper introduces a complete LiDAR-based system designed to enable UAVs to avoid various moving obstacles in complex environments. Benefiting the high computational efficiency of perception and planning, the system can operate in real time using onboard computing resources with low latency. For dynamic environment perception, we have integrated our previous work, M-detector, into the system. M-detector ensures that moving objects of different sizes, colors, and types are reliably detected. For dynamic environment planning, we incorporate dynamic object predictions into the integrated planning and control (IPC) framework, namely DynIPC. This integration allows the UAV to utilize predictions about dynamic obstacles to effectively evade them. We validate our proposed system through both simulations and real-world experiments. In simulation tests, our system outperforms state-of-the-art baselines across several metrics, including success rate, time consumption, average flight time, and maximum velocity. In real-world trials, our system successfully navigates through forests, avoiding moving obstacles along its path.

Efficient Swept Volume-Based Trajectory Generation for Arbitrary-Shaped Ground Robot Navigation

Navigating an arbitrary-shaped ground robot safely in cluttered environments remains a challenging problem. The existing trajectory planners that account for the robot's physical geometry severely suffer from the intractable runtime. To achieve both computational efficiency and Continuous Collision Avoidance (CCA) of arbitrary-shaped ground robot planning, we proposed a novel coarse-to-fine navigation framework that significantly accelerates planning. In the first stage, a sampling-based method selectively generates distinct topological paths that guarantee a minimum inflated margin. In the second stage, a geometry-aware front-end strategy is designed to discretize these topologies into full-state robot motion sequences while concurrently partitioning the paths into SE(2) sub-problems and simpler R2 sub-problems for back-end optimization. In the final stage, an SVSDF-based optimizer generates trajectories tailored to these sub-problems and seamlessly splices them into a continuous final motion plan. Extensive benchmark comparisons show that the proposed method is one to several orders of magnitude faster than the cutting-edge methods in runtime while maintaining a high planning success rate and ensuring CCA.

Reflection on Data Storytelling Tools in the Generative AI Era from the Human-AI Collaboration Perspective

Human-AI collaborative tools attract attentions from the data storytelling community to lower the barrier of expertise and streamline the workflow. The recent advance in large-scale generative AI techniques, e.g., large language models (LLMs) and text-to-image models, has the potential to enhance data storytelling with their power in visual and narration generation. After two years since these techniques were publicly available, it is important to reflect our progress of applying them and have an outlook for future opportunities. To achieve the goal, we compare the collaboration patterns of the latest tools with those of earlier ones using a dedicated framework for understanding human-AI collaboration in data storytelling. Through comparison, we identify persistent collaboration patterns, e.g., human-creator + AI-assistant, and emerging ones, e.g., AI-creator + human-reviewer. The benefits of these AI techniques and other implications to human-AI collaboration are also revealed. We further propose future directions to hopefully ignite innovations.

Efficient and Distributed Large-Scale Point Cloud Bundle Adjustment via Majorization-Minimization

Point cloud bundle adjustment is critical in large-scale point cloud mapping. However, it is both computationally and memory intensive, with its complexity growing cubically as the number of scan poses increases. This paper presents BALM3.0, an efficient and distributed large-scale point cloud bundle adjustment method. The proposed method employs the majorization-minimization algorithm to decouple the scan poses in the bundle adjustment process, thus performing the point cloud bundle adjustment on large-scale data with improved computational efficiency. The key difficulty of applying majorization-minimization on bundle adjustment is to identify the proper surrogate cost function. In this paper, the proposed surrogate cost function is based on the point-to-plane distance. The primary advantages of decoupling the scan poses via a majorization-minimization algorithm stem from two key aspects. First, the decoupling of scan poses reduces the optimization time complexity from cubic to linear, significantly enhancing the computational efficiency of the bundle adjustment process in large-scale environments. Second, it lays the theoretical foundation for distributed bundle adjustment. By distributing both data and computation across multiple devices, this approach helps overcome the limitations posed by large memory and computational requirements, which may be difficult for a single device to handle. The proposed method is extensively evaluated in both simulated and real-world environments. The results demonstrate that the proposed method achieves the same optimal residual with comparable accuracy while offering up to 704 times faster optimization speed and reducing memory usage to 1/8. Furthermore, this paper also presented and implemented a distributed bundle adjustment framework and successfully optimized large-scale data (21,436 poses with 70 GB point clouds) with four consumer-level laptops.

Self-supervised Spatial-Temporal Learner for Precipitation Nowcasting

Nowcasting, the short-term prediction of weather, is essential for making timely and weather-dependent decisions. Specifically, precipitation nowcasting aims to predict precipitation at a local level within a 6-hour time frame. This task can be framed as a spatial-temporal sequence forecasting problem, where deep learning methods have been particularly effective. However, despite advancements in self-supervised learning, most successful methods for nowcasting remain fully supervised. Self-supervised learning is advantageous for pretraining models to learn representations without requiring extensive labeled data. In this work, we leverage the benefits of self-supervised learning and integrate it with spatial-temporal learning to develop a novel model, SpaT-SparK. SpaT-SparK comprises a CNN-based encoder-decoder structure pretrained with a masked image modeling (MIM) task and a translation network that captures temporal relationships among past and future precipitation maps in downstream tasks. We conducted experiments on the NL-50 dataset to evaluate the performance of SpaT-SparK. The results demonstrate that SpaT-SparK outperforms existing baseline supervised models, such as SmaAt-UNet, providing more accurate nowcasting predictions.

Autonomous Tail-Sitter Flights in Unknown Environments

Trajectory generation for fully autonomous flights of tail-sitter unmanned aerial vehicles (UAVs) presents substantial challenges due to their highly nonlinear aerodynamics. In this paper, we introduce, to the best of our knowledge, the world's first fully autonomous tail-sitter UAV capable of high-speed navigation in unknown, cluttered environments. The UAV autonomy is enabled by cutting-edge technologies including LiDAR-based sensing, differential-flatness-based trajectory planning and control with purely onboard computation. In particular, we propose an optimization-based tail-sitter trajectory planning framework that generates high-speed, collision-free, and dynamically-feasible trajectories. To efficiently and reliably solve this nonlinear, constrained \textcolor{black}{problem}, we develop an efficient feasibility-assured solver, EFOPT, tailored for the online planning of tail-sitter UAVs. We conduct extensive simulation studies to benchmark EFOPT's superiority in planning tasks against conventional NLP solvers. We also demonstrate exhaustive experiments of aggressive autonomous flights with speeds up to 15m/s in various real-world environments, including indoor laboratories, underground parking lots, and outdoor parks. A video demonstration is available at https://youtu.be/OvqhlB2h3k8, and the EFOPT solver is open-sourced at https://github.com/hku-mars/EFOPT.