News & Research

arXiv:2605.02263v2 Announce Type: replace Abstract: Recent diffusion large language models (dLLMs) have demonstrated both effectiveness and efficiency in reasoning via a block-based semi-autoregressive generation paradigm. Despite their progress, the fixed-size block generations remain a critical bottleneck for effective and coherent reasoning. 1. From a global perspective, different reasoning tasks would correspond to different optimal decoding block sizes, which makes a ``one-size-fits-all'' assumption ineffective. 2. Even within a single reasoning task, the rigid block partitioning would break the logical flow and reduce reasoning coherence. Through empirical observations, we reveal that for block-wise entropy, incorrect reasoning exhibits a fluctuating and unsteady trend between blocks, whereas the correctly generated tasks follow a consistent descending trend. Therefore, this paper proposes b1, a novel post-training framework for dLLMs that learns dynamic-size reasoning blocks via

arXiv:2205.14090v2 Announce Type: replace-cross Abstract: Bayesian optimization (BO) has well-documented merits for optimizing black-box functions with an expensive evaluation cost. Such functions emerge in applications as diverse as hyperparameter tuning, drug discovery, and robotics. BO hinges on a Bayesian surrogate model to sequentially select query points so as to balance exploration with exploitation of the search space. Most existing works rely on a single Gaussian process (GP) based surrogate model, where the kernel function form is typically preselected using domain knowledge. To bypass such a design process, this paper leverages an ensemble (E) of GPs to adaptively select the surrogate model fit on-the-fly, yielding a GP mixture posterior with enhanced expressiveness for the sought function. Acquisition of the next evaluation input using this EGP-based function posterior is then enabled by Thompson sampling (TS) that requires no additional design parameters. To endow function

arXiv:2503.18893v2 Announce Type: replace-cross Abstract: Long-context Large Language Models (LLMs) enable powerful applications but incur high memory costs due to the key-value states (KV-Cache). Recent studies attempt to share KV-Cache across layers, but these approaches either require expensive pretraining or rely on per-token cross-layer cosine similarity that is often limited in practice. We show, via Centered Kernel Alignment (CKA), that the dominant singular vectors of KV-Cache are well aligned across layers. Motivated by this observation, we propose xKV, a post-training compression method that jointly factorizes grouped-layer KV-Cache into a shared low-rank subspace, substantially reducing KV-Cache memory. Across widely used LLMs, xKV achieves up to 8x KV-Cache compression while preserving accuracy on long-context tasks and in multi-turn settings. To further improve efficiency, we introduce Selective Reconstruction (SR) at decode time. Combined with SR, xKV achieves up to 4.23x

arXiv:2601.13433v3 Announce Type: replace-cross Abstract: Prior research demonstrates that performance of language models on reasoning tasks can be influenced by suggestions, hints and endorsements. However, the influence of endorsement source credibility remains underexplored. We investigate whether language models exhibit systematic bias based on the perceived expertise of the provider of the endorsement. Across 4 datasets spanning mathematical, legal, and medical reasoning, we evaluate 11 models using personas representing four expertise levels per domain. Our results reveal that models are increasingly susceptible to incorrect/misleading endorsements as source expertise increases, with higher-authority sources inducing not only accuracy degradation but also increased confidence in wrong answers. We also show that this authority bias is mechanistically encoded within the model and a model can be steered away from the bias, thereby improving its performance even when an expert gives

arXiv:2602.15894v2 Announce Type: replace-cross Abstract: In many large language model (LLM) alignment applications, users expect not only high-quality outputs but also substantial diversity. However, existing methods often face a fundamental trade-off between these objectives: approaches that improve output quality tend to reduce diversity, while methods that increase diversity often do so at the expense of quality. In this work, we propose Quality-constrained Entropy Maximization Policy Optimization (QEMPO), a novel framework that enhances the diversity of LLM outputs while explicitly preserving output quality. QEMPO is grounded in a strong theoretical foundation: we derive a closed-form analytical solution that provably maximizes entropy-a principled measure of diversity-subject to a quality constraint, with guarantees on optimality under the defined objective. Leveraging this solution, QEMPO naturally supports both online and offline training settings. Empirical results demonstrate

arXiv:2605.00457v3 Announce Type: replace-cross Abstract: The coexistence of NR-U and Wi-Fi in unlicensed spectrum introduces a challenging coexistence management problem, where heterogeneous channel access mechanisms lead to a significant imbalance in spectrum utilization and degraded Wi-Fi performance. To address this challenge, we propose a policy-driven deep reinforcement learning (DRL) framework for adaptive transmission opportunity (TXOP) control, in which the coexistence process is formulated as a Markov decision process (MDP) and a deep Q-network (DQN) learns control policies through online interaction. A key contribution is the introduction of a policy layer via reward design, enabling explicit control of coexistence tradeoffs among fairness, throughput, and utility. Three policies, namely absolute fairness, moderate fairness, and utility-based fairness, are developed to achieve different operating points. Simulation results show that the proposed framework achieves a Jain fai

arXiv:2510.03559v3 Announce Type: replace-cross Abstract: UX professionals routinely conduct design reviews, yet privacy concerns are often overlooked, not only due to limited tools, but more fundamentally from low intrinsic motivation, driven by limited privacy knowledge, weak empathy for unexpectedly affected users, and low autonomy in identifying harms. We present PrivacyMotiv, an LLM-powered system that generates vulnerability-centered personas, persona journey stories, and traceable design diagnoses grounded in lo-fi user flows to support privacy-oriented UX design review. In a within-subjects study with professional UX practitioners (N=16), PrivacyMotiv significantly improved empathy, intrinsic motivation, and perceived usefulness, with participants identifying 59% more privacy issues and proposing 70% more redesign solutions compared to self-proposed methods. This work contributes empirical insight into motivational barriers in privacy-aware UX and a structured, narrative-driven

arXiv:2605.27461v1 Announce Type: new Abstract: Vision-Language-Action (VLA) policies have shown promising manipulation capabilities, yet their practical impact is often limited by the reliability demands of real-world deployment. We present a deployment study of an industrial packaging task at Siemens Factory (GWE, Erlangen, Germany), where a robot must pick a transparent accessory bag from a cluttered pile, insert it into the remaining cavity of a cardboard package, and ensure that the bag and its contents remain below the closing plane. Our goal is to understand the practical effort required to adapt a pretrained Pi0.5 policy to a single factory-floor task through iterative fine-tuning and deployment-driven refinement. The pipeline consists of repeated loops of data collection, curation, fine-tuning, evaluation, and targeted recovery data collection. We have accumulated 2535 episodes (10 hours) from the on-site factory settings. In this paper, we contribute an empirical account of a

arXiv:2605.27532v1 Announce Type: new Abstract: Emergent communication enables partially observant Autonomous Mobile Robots (AMRs) to coordinate effectively in decentralized multi-agent reinforcement learning (MARL) settings. However, existing approaches often struggle with unstable communication protocols, ungrounded message semantics, and interference between communication learning and policy optimization, leading to degraded coordination over time. We propose SCALE-COMM (Shared, Contrastively-Aligned Latent Embeddings for COMMunication), a self-supervised framework for learning compact, stable, and policy-relevant communication representations. SCALE-COMM decouples communication learning from policy optimization by training low-dimensional latent messages that capture task-relevant planning and traffic information, while enforcing consistency across agents and time. Across standard MARL benchmarks and a realistic warehouse coordination task, SCALE-COMM consistently outperforms exist

arXiv:2605.27533v1 Announce Type: new Abstract: Periods of heightened arousal or restlessness can interfere with children's ability to focus, self-regulation, and physically calm. Technologies that encourage embodied self-regulation through tactile interaction may provide a simple and accessible means of promoting calmness. This paper investigates how interaction with a pocket-sized tactile device influences physiological and behavioral markers of calmness in typically developing children. Building on prior work examining heart rate modulation, we present new findings on how tactile interaction affects full-body movement and postural stability. We employ a device that engages children through a hand-held rhythmic vibration-matching game, designed to focus attention and encourage stillness. Eighteen children participated in a within-subjects study that involved two conditions: with and without tactile interaction with a hand-held device, while having their heart rate and body movement r

arXiv:2605.27539v1 Announce Type: new Abstract: Many children experience challenges in emotional regulation and social interaction, which can limit their participation in everyday activities and therapeutic programs. For socially assistive robots to be effective in this context, it is essential that children remain consistently and meaningfully engaged. We explore engagement strategies for a tactile robot designed to support children suffering from anxiety disorders through daily interactions. The robot delivers either synthetic emotional feedback or point rewards to encourage user participation. We evaluated these strategies through two studies: a preference assessment with 16 school children aged 6-8 years, and a behavioral study with 14 university students aged 20-27 years in naturalistic environments. The study with school children indicated a preference for emotional engagement over points-based approaches. The follow up study with university students across a full day of interact

arXiv:2605.27699v1 Announce Type: new Abstract: Sampling-based motion planners offer a practical and scalable approach to kinodynamic motion planning, notably for high-dimensional, underactuated, or non-holonomic systems. However, these planners are typically used offline, requiring execution to begin only after the trajectory has been computed. In addition, the planned trajectory may not be accurately tracked in the presence of motion uncertainty, leading to deviations from the nominal solution. In this work, these limitations were addressed within a unified framework, \method, an asymptotically-optimal meta-planner framework that improves both path quality and tracking performance during execution. In addition to the main execution thread, this framework comprises a replanning method that continuously explores the state space and refines the trajectory during execution, and an optimization process that refines future control inputs to reduce tracking error. Together, these components

arXiv:2605.27724v1 Announce Type: new Abstract: Imitation learning is a promising approach for training humanoid robots to both walk and manipulate, but it requires a large number of demonstrations, which are time-intensive and difficult to collect via teleoperation. Existing data-generation algorithms can automatically synthesize demonstrations for manipulators, but they are ineffective on humanoids because their high-dimensional composite action spaces involve arms, legs, and torsos. We present HumanoidMimicGen, a method for generating humanoid legged loco-manipulation data. Our method adapts contact-rich whole-body skills from a handful of source demonstrations to new states, generalizing across changes in object pose. By interleaving these single- and dual-arm skills with whole-body locomotion and manipulation planning, the method generates stable, collision-free data across diverse scenes and layouts. To evaluate our approach, we introduce a new simulated loco-manipulation benchma

arXiv:2605.27917v1 Announce Type: new Abstract: Uncrewed Aerial Systems (UASs) have become a growing threat to the security of critical infrastructure, exploiting spatiotemporal gaps in sensor perimeters to infiltrate restricted airspace undetected. We formulate this interaction as a two-player zero-sum differential game between an adversarial UAS and a heterogeneous sensor network of directional and omnidirectional sensors. Unlike earlier game-theoretic approaches that restrict the defender to discrete placement graphs or fixed configurations, we introduce a continuous sensor redeployment technique in which each sensor slides freely along the convex building boundaries. This is enforced via a log-sum-exp smooth approximation that preserves differentiability at polygon vertices, enabling optimization with gradient-based methods. The attacker's best response is computed via a two-step approach combining STP-RRT* for feasible trajectory initialization and nonlinear programming for detect

arXiv:2605.28048v1 Announce Type: new Abstract: Sequence-based visual place recognition (VPR) for SLAM and robot relocalization must decide whether the retrieved top-1 candidate is safe to accept. Conformal prediction is a natural framework for this accept/reject decision, but its finite-sample guarantees rely on exchangeability between calibration and deployment (test) data, which is violated under cross-condition deployment. We introduce SAFEVPR, a non-trainable verification-and-calibration pipeline for safe cross-condition sequence VPR. SAFEVPR replaces the standard backbone cosine similarity with a mutual-nearest-neighbour (MNN) patch-matching score computed from frozen DINOv2 ViT features, and replaces flat Learn-Then-Test calibration with Mondrian conformal LTT, fitting separate Bonferroni-corrected thresholds across score bins. Under exchangeability, these thresholds would provide finite-sample false-discovery-rate (FDR) control; under condition shift, we evaluate empirical vali