Knowledge catalogue
arXiv:2610.05898v1 Announce Type: new Abstract: Real-world users often exhibit highly heterogeneous preferences over multiple objectives for LLM responses. A lightweight aligner can tailor these respo
arXiv:2610.06104v1 Announce Type: new Abstract: Multimodal imitation learning requires diverse executable futures under the same observation and consistent behavior across replanning cycles. We presen
arXiv:2610.04605v1 Announce Type: new Abstract: Many visual explanation methods in computer vision highlight pixel importance but struggle to link these low-level cues to semantically meaningful conce
arXiv:2610.04239v1 Announce Type: cross Abstract: Uncertainty quantification (UQ) for large language models (LLMs) aims to provide reliable measures of predictive confidence, yet current methods are o
arXiv:2610.06651v1 Announce Type: new Abstract: Methods for generalization in model-based reinforcement learning typically assume that an agent cannot recover the latent context governing the environm
arXiv:2610.04231v1 Announce Type: new Abstract: Humanoid whole-body controllers can now track a diverse set of dynamic motions, but they are typically trained offline and then frozen, so teaching such
arXiv:2512.11526v2 Announce Type: replace Abstract: Time series forecasting predicts future values from past data. In real-world settings, some anomalous events have lasting effects and influence the
arXiv:2610.05199v1 Announce Type: new Abstract: How strongly do the remaining diffusion-sampling steps amplify a perturbation at a late latent state? Standard measurements answer this question with ne
arXiv:2610.04851v1 Announce Type: new Abstract: Open-ended discovery requires learning from repeated attempts while continuing to explore directions whose value is not yet apparent. Search with a froz
arXiv:2610.05678v1 Announce Type: new Abstract: Most humanoid loco-manipulation controllers require human motion data to learn whole-body coordination and posture, leaving policies reliant on external
arXiv:2610.04700v1 Announce Type: new Abstract: Federated learning enables medical institutions to train a global model without sharing data, yet feature heterogeneity from diverse scanners or protoco
arXiv:2504.06470v2 Announce Type: replace-cross Abstract: Ensuring fairness is essential as machine learning increasingly informs consequential decisions. However, many fairness-aware methods focus on
arXiv:2603.19621v2 Announce Type: replace Abstract: Deep Reinforcement Learning (DRL) provides a general-purpose methodology for training inventory policies that can leverage big data and compute. How
arXiv:2610.05755v1 Announce Type: new Abstract: Diffusion and VLA policies for manipulation are often deployed through downstream impedance controllers. The stiffness and damping gains of these contro
arXiv:2511.04086v2 Announce Type: replace Abstract: With the rapid growth of graph-structured data in critical domains, unsupervised graph-level anomaly detection (UGAD) has become a pivotal task. UGA
arXiv:2610.04084v1 Announce Type: new Abstract: Generative AI can produce engineering artefacts, but generation alone does not determine whether or how those artefacts should enter safety-critical wor
arXiv:2610.05087v1 Announce Type: new Abstract: Contrastive Reinforcement Learning (CRL) learns representations that estimate goal reachability, yet its policy remains conditioned on raw goals and the
arXiv:2503.23835v2 Announce Type: replace Abstract: Grasp-based manipulation tasks are fundamental to robots interacting with their environments, yet gripper state ambiguity significantly reduces the
arXiv:2603.04359v2 Announce Type: replace Abstract: Quantization can drastically increase the efficiency of large language and vision models, but typically incurs an accuracy drop. Recently, function-
arXiv:2505.19058v2 Announce Type: replace Abstract: We propose a novel distributionally robust Q-learning algorithm for the non-tabular case accounting for continuous state spaces where the state tran
arXiv:2610.06078v1 Announce Type: new Abstract: This paper asks whether VLA generalization is grounded in a global understanding of objects' physical properties that enables policies to adapt their mo
arXiv:2610.06327v1 Announce Type: cross Abstract: Vision-based autonomous navigation for low-cost robots remains a fundamental challenge, primarily due to the significant gap between simulated trainin
arXiv:2610.05048v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense token-level supervision without a second model: one network acts as teacher with the reference solutio
arXiv:2610.04758v1 Announce Type: new Abstract: Machine-learning classification is now used across medicine, the social sciences, economics, education and engineering, very often by researchers who do
arXiv:2507.20335v2 Announce Type: replace Abstract: Educational language models should pursue three complementary objectives: helpfulness through responsible guidance, personalization to learner needs
arXiv:2607.09701v2 Announce Type: replace Abstract: The enduring vision of general-purpose robots serving humanity hinges fundamentally on policy steerability. However, prevailing paradigms of learnin
arXiv:2610.06235v1 Announce Type: cross Abstract: Instruction following is central to language-conditioned robot policies: language should determine what to do when the same scene permits multiple val
arXiv:2610.05733v1 Announce Type: new Abstract: Autonomous social navigation requires balancing efficiency, physical safety, and social compliance. Reinforcement Learning (RL) methods provide a viable
arXiv:2610.05292v1 Announce Type: new Abstract: What happens to information a pretrained model already encodes when post-training no longer rewards using it? The common language of representation comp
arXiv:2610.06269v1 Announce Type: cross Abstract: Open-ended scientific discovery often requires repeatedly proposing and evaluating candidate solutions. LLM-based systems can support this process by
arXiv:2610.03951v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as classifiers, yet they operate as opaque systems whose decisions are difficult to interpret, which
arXiv:2610.04140v1 Announce Type: new Abstract: Mixture-of-experts (MoE) language models send each token to a few experts, so each expert is trained on a different part of the data, and this part chan
arXiv:2610.04805v1 Announce Type: cross Abstract: Three-dimensional perception is critical for robotic manipulation, particularly for high-precision tasks, as recovering metric depth and precise 3D ob
arXiv:2610.06484v1 Announce Type: new Abstract: Graph neural networks (GNNs) are the standard tool for node representation learning and are increasingly used in high-stakes settings. Their message-pas
arXiv:2610.05968v1 Announce Type: new Abstract: We study last-iterate convergence in unknown two-player zero-sum discounted Markov games with bandit feedback. The players learn independently along a s
arXiv:2512.05134v2 Announce Type: replace Abstract: Diffusion transformers require repeated denoiser evaluations, making image and video generation computationally expensive. We propose a training-fre
arXiv:2610.05869v1 Announce Type: cross Abstract: This paper studies online quantile regression for large-scale and streaming data using Stochastic SubGradient Descent (SSGD) with constant learning ra
arXiv:2610.04283v1 Announce Type: new Abstract: Activation steering exploits interpretable directions in the residual stream to enable inference-time manipulation of model behavior. Composing steering
arXiv:2610.06105v1 Announce Type: new Abstract: On-policy distillation (OPD) provides dense teacher supervision on student-generated trajectories, but generating and evaluating long rollouts incurs su
arXiv:2610.05483v1 Announce Type: new Abstract: World--action models (WAMs) promise a unified model that predicts action-conditioned futures, generates feasible actions, and supports planning in imagi
arXiv:2510.14242v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) often produce inconsistent answers when faced with different phrasings of the same prompt. In this paper, we prop
arXiv:2610.04767v1 Announce Type: cross Abstract: Latent world models enable robots to plan by predicting the consequences of actions. Planning long tasks with control-rate actions requires many predi
arXiv:2606.14049v2 Announce Type: replace-cross Abstract: We present FoleyGenEx, a unified video-to-audio (VTA) framework integrating multi-modal control, frame-level temporal alignment, and fine-grai
arXiv:2610.04607v1 Announce Type: cross Abstract: Robots need to anticipate how their actions will change the world, since manipulation success hinges on the resulting contacts and object motions. How
arXiv:2610.04754v1 Announce Type: new Abstract: Even a few action perturbations can substantially degrade the performance of a deployed decision policy. Certifying the resulting return loss is challen
arXiv:2610.04372v1 Announce Type: cross Abstract: Transferring visual representations pretrained on human videos to robot manipulation requires learning reliable correspondences between human and robo
arXiv:2610.05734v1 Announce Type: new Abstract: Online control of execution speed is essential for deploying robot policies in real-world scenarios, as robots may need to speed up under time constrain
arXiv:2503.17589v3 Announce Type: replace Abstract: Kinematic motion planners are among the most widely used control approaches in robotic applications. By modeling the robot as a first-order system,
arXiv:2610.04569v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in safety-critical settings, yet their black-box nature makes it difficult to provide formal gu
arXiv:2610.06280v1 Announce Type: new Abstract: World action models (WAMs) have emerged as a promising paradigm for robotic manipulation. They act by first predicting how a task should be performed an
arXiv:2610.05563v1 Announce Type: cross Abstract: Small language model (SLM) agents need safety controls that track consequences across tool calls with little monitoring overhead. A private read, for
arXiv:2610.06290v1 Announce Type: cross Abstract: GAMBIT is an opening chess move in which a player sacrifices a piece, typically a pawn, to gain a positional advantage later in the game. Analogously,
arXiv:2610.04980v1 Announce Type: new Abstract: Preference alignment has become a standard practice for text-to-image diffusion models. Direct Preference Optimization (DPO) simplifies this process by
arXiv:2610.04910v1 Announce Type: new Abstract: Collision avoidance among convex bodies is a fundamental problem in robotics. Control Barrier Functions (CBFs) provide a practical framework for real-ti
arXiv:2610.04827v1 Announce Type: new Abstract: Time-series foundation models (TSFMs) enable zero-shot forecasting through large-scale cross-domain pretraining, while retrieval augmentation further im
arXiv:2610.05450v1 Announce Type: new Abstract: Preference-based alignment methods such as reinforcement learning from human feedback (RLHF) and Nash learning from human feedback (NLHF) aggregate pair
arXiv:2604.00385v2 Announce Type: replace Abstract: Type 1 diabetes (T1D) management requires continuous adjustment of insulin and lifestyle behaviors to maintain blood glucose within a safe target ra
arXiv:2610.06503v1 Announce Type: new Abstract: Text-to-image generative models can produce highly realistic imagery but also raise concerns about harmful misuse. While safety mechanisms exist, system
arXiv:2610.05855v1 Announce Type: new Abstract: A bipedal robot cannot deviate from its path to avoid an obstacle without disturbing its balance, and this coupling is most severe on underactuated plat
arXiv:2610.05446v1 Announce Type: new Abstract: Off-policy hierarchical reinforcement learning must estimate the values of high-level decisions while the low-level policy changes. HIRO adapts replay d