42 accepted papers
Large language models (LLMs) are increasingly used in scientific domains. While they can produce reasoning-like content via methods such as chain-of-thought prompting, these outputs are typically unstructured and informal, obscuring whether models truly understand the fundamental reasoning paradigms
Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal reasoning tasks; however, their learning remains constrained by the limitations of human-annotated supervision. Recent self-rewarding approaches attempt to overcome this constraint by allowing models to act as their…
Training LLM agents for complex multi-turn decision-making tasks requires extensive exploration within their environment, with reinforcement learning (RL) as a natural way. However, the open-source community currently lacks a unified RL framework capable of training agents from scratch across divers…
Federated learning (FL) has emerged as a distributed machine learning paradigm due to its privacy-preserving advantages. Most FL studies assume offline labeled datasets are available at clients. In practice, however, client data often arrive without labels in a streaming manner, making label acquisi
Interpreting residual architectures as discretizations of Ordinary Differential Equations characterizes the forward pass as a continuous latent flow. Although this framework defines the mechanics of inference, conventional training paradigms primarily constrain the terminal state, leaving the interm…
Reinforcement fine-tuning (RFT) has shown promise for training LLM agents to perform multi-turn decision-making based on environment feedback. However, most existing evaluations remain largely in-domain—training and testing are conducted in the same environment or even on the same tasks. In real-wor…
Cognitive-functional dialogues, such as those for persuasion, consultation, and question-answering, are prevalent throughout human social interaction. The core difference between these dialogues and casual chat lies in their objective: to guide a person
Federated Graph Learning (FGL) has emerged as a principled framework for decentralized training of Graph Neural Networks (GNNs) while preserving data privacy. In subgraph-FL scenarios, however, structural noise arising from data collection and storage can damage the GNN message-passing scheme of cli
Autoregressive models are structurally misaligned with the inherently parallel nature of geospatial understanding, forcing a rigid sequential narrative onto scenes and fundamentally hindering the generation of structured and coherent outputs. We challenge this paradigm by reframing geospatial genera
Whole-body loco-manipulation for quadruped robots with arms remains a challenging problem, particularly in achieving multi-task control. To address this, we propose MLM, a reinforcement learning framework driven by both real-world and simulation data. It enables a six-DoF robotic arm–equipped quadru
Vision-language-action models exhibit an inherent trade-off in action chunk length (``horizon''): longer horizons improve global foresight but degrade fine-grained local control, while shorter ones yield the opposite. To mitigate the trade-off, we propose a $\textbf{mixture of horizons (MoH)}$ strat…
Automatically generating clear and accurate figures for research papers remains challenging, as it requires semantic understanding, precise structure, and visual aesthetics. Existing approaches struggle to balance fidelity and quality: large language model (LLM) code-based methods (e.g., SVG, Mermai
Open-vocabulary 3D object affordance grounding aims to identify functional regions of objects given arbitrary semantic descriptions. However, existing methods often rely on fixed training categories and geometric priors, lacking geometric invariance and analogical reasoning capabilities. Since there
To support long-term interaction in complex environments, LLM agents require memory systems that manage historical experiences. Existing approaches either retain full interaction histories via passive context extension, leading to substantial redundancy, or rely on iterative reasoning to filter nois…
The emergence of large vision-language models (VLMs) has significantly enhanced the efficiency and flexibility of geospatial interpretation. However, general-purpose VLMs remain suboptimal for remote sensing (RS) tasks. Existing geospatial VLMs typically adopt a unified modeling strategy and struggl
3D Gaussian splatting (3DGS) has become a vital tool for learning a radiance field from multiple posed images. Although 3DGS shows great advantages over NeRF in terms of rendering quality and efficiency, it remains a research challenge to further improve the efficiency of learning 3D Gaussians. To o
Reinforcement learning (RL) has recently become the core paradigm for aligning and strengthening large language models (LLMs). Yet, applying RL in off-policy settings—where stale data from past policies are used for training—improves sample efficiency, but remains challenging: policy entropy decline…
Vision-Language Models (VLMs) in remote sensing often fail at complex analytical tasks, a limitation stemming from their end-to-end training paradigm that bypasses crucial reasoning steps and leads to unverifiable outputs. To address this limitation, we introduce the Perceptually-Grounded Geospatial…
The quest for incremental unified multimodal anomaly detection seeks to empower a single model with the ability to systematically detect anomalies across all categories and support incremental learning to accommodate emerging objects/categories. Central to this pursuit is resolving the catastrophic
Learning to Defer operates by deferring AI-uncertain samples to humans, enabling the system to outperform either alone. Some works extend it by introducing human-AI combination to deferral. However, they remain limited to binary choices. Given that AI, humans, and human-AI combination are each indis
Recent studies indicate that human-AI collaboration performs better than either alone, particularly in medical diagnosis. Beyond collaboration methods that focus on assigning tasks to humans or AI, like deferral, combining human and AI decisions with their confidence scores is emerging as a promisin
In this paper, we introduce BMMR, a large-scale bilingual, multimodal, multi-disciplinary reasoning dataset for the community to develop and evaluate large multimodal models (LMMs). BMMR comprises 100k university-level questions drawn from 300 UNESCO-defined subjects, spanning diverse formats—multip…
The cooperative driving technology of Connected and Autonomous Vehicles (CAVs) is crucial for improving the efficiency and safety of transportation systems. Learning-based methods, such as Multi-Agent Reinforcement Learning (MARL), have demonstrated strong capabilities in cooperative decision-making
The Latent Diffusion Model (LDM) has demonstrated strong capabilities in high-resolution image generation and has been widely employed for Pose-Guided Person Image Synthesis (PGPIS), yielding promising results. However, the compression process of LDM often results in the deterioration of details, pa…
Existing efforts to boost multimodal fusion of 3D anomaly detection (3D-AD) primarily concentrate on devising more effective multimodal fusion strategies. However, little attention was devoted to analyzing the role of multimodal fusion architecture (topology) design in contributing to 3D-AD. In this…
With advancements in deep learning, neural simulators have become increasingly important for improving the efficiency and effectiveness of simulating complex dynamical systems in various scientific and technological fields. This paper presents a novel neural simulator called Context-informed Polymor
Model performance in text-to-image (T2I) and image-to-image (I2I) generation often depends on multiple aspects, including quality, alignment, diversity, and robustness. However, models' complex trade-offs among these dimensions have been rarely explored due to (1) the lack of datasets that allow fin…
"Recently, the growing capabilities of deep generative models have underscored their potential in enhancing image classification accuracy. However, existing methods often demand the generation of a disproportionately large number of images compared to the original dataset, while having only marginal…
This paper introduces a novel bistable origami flexible gripper, which is based on a single-vertex and multi-crease (SVMC) origami structure that has sxeveral advantages, including a simple structure, low cost, and strong deformation capacity. This design addresses the drawbacks of slow response spe
Unsupervised Anomaly Detection (UAD) with incremental training is crucial in industrial manufacturing, as unpredictable defects make obtaining sufficient labeled data infeasible. However, continual learning methods primarily rely on supervised annotations, while the application in UAD is limited due…
We study the problem of $(\epsilon,\delta)$-certified machine unlearning for minimax models. Most of the existing works focus on unlearning from standard statistical learning models that have a single variable and their unlearning steps hinge on the direct Hessian-based conventional Newton update. W…
In the area of few-shot anomaly detection (FSAD), efficient visual feature plays an essential role in the memory bank $\mathcal{M}$-based methods. However, these methods do not account for the relationship between the visual feature and its rotated visual feature, drastically limiting the anomaly de…
High-precision point cloud anomaly detection is the gold standard for identifying the defects of advancing machining and precision manufacturing. Despite some methodological advances in this area, the scarcity of datasets and the lack of a systematic benchmark hinder its development. We introduce Re…
When talking about soft robots, softness is considered the most important feature, which brings dexterity and safety in interactive tasks with humans and environments. Such softness sometimes limits the real application of soft robots because load capability and rigidity are widely needed on many oc
Accurate segmentation of surgical instrument tip is an important task for enabling downstream applications in robotic surgery, such as surgical skill assessment, tool-tissue interaction and deformation modeling, as well as surgical autonomy. However, this task is very challenging due to the small si…
In this letter, we address the problem of forecasting the trajectory of an egocentric camera wearer (ego-person) in crowded spaces. The trajectory forecasting ability learned from the data of different camera wearers walking around in the real world can be transferred to assist visually impaired peo
Semantic segmentation of surgical instruments provides essential priors for autonomous surgery. This task is however challenging since the fine-structure of surgical instruments requires the accurate segmentation of detailed regions in images. As the visual guidance for autonomous surgery, the algor…
The miniature soft robots have many promising applications, including micro-manipulations, endoscopy, and microsurgery, etc. Nevertheless, it remains challenging to fabricate a miniatured robot device that is thin, flexible, and can perform multimodal locomotor mobility with sensory capacity. In thi
Recently, total variation (TV)-based regularization has become a standard technique for signal denoising. The reconstruction quality is generally sensitive to the value of regularization parameter. In this work, based on Cham-bolle's algorithm, we develop two data-driven optimization schemes based o…
Recently, the type of compound regularizers has become a popular choice for signal reconstruction. The estimation quality is generally sensitive to the values of multiple regularization parameters. In this work, based on BDF algorithm, we develop a data-driven optimization scheme based on minimizati…
For accurate signal reconstruction, proximal gradient methods generally require proper selection of regularization parameter. In this paper, we develop two data-driven optimization schemes, based on minimization of unbiased predictive risk estimate (UPRE). First, we propose a recursive UPRE to estim…
Sparsity-promoting regularization is often formulated as ℓv-penalized minimization (0 <; v ≤ 1), which can be efficiently solved by iteratively reweighted least squares (IRLS). The reconstruction quality is generally sensitive to the value of regularization parameter. In this work, for accurate reco…