Papers

Stellar associations are difficult clustering targets because position, kinematics and extinction carry different physical units, different uncertainties and different discriminatory power. We extend the Extended Minimum Spanning Tree (XMST) framework without forcing these observables into a single metric. Three-dimensional spatial connectivity first defines candidate parent structures (XMST-S1), transverse kinematics then tests those parents for dynamically distinct populations (XMST-S3), and visual extinction provides a final reddening refinement (XMST-S2). Across 1000 Monte Carlo realisations, XMST-S1 produced a better purity–completeness balance than the eight predefined three-dimensional HDBSCAN configurations tested, although neither method reliably separated a deliberately superimposed pair using spatial information alone. Applying kinematics before reddening increased mean Jaccard similarity to 0.8354 and resolved the overlapping pair in 98.6 per cent of realisations. The S3 null experiment produced only five accepted subdivisions among 106,265 tested groups. Across 1000 paired reddening-uncertainty realisations, increasing uncertainty reduced the mean number of accepted S2 subdivisions from 4.224 to 1.982, while mean Jaccard similarity changed only from 0.8074 to 0.7993. Direct XMST-S3 uncertainty tests showed negligible practical degradation for transverse-velocity errors up to $0.5~\mathrm{km\,s^{-1}}$ per component, with larger errors progressively suppressing kinematic refinement. In a label-blind repartition of 2,551 stars from 56 published OB associations, median best-match Jaccard similarity increased from 0.6535 after spatial clustering to 0.8861 after complete refinement, with 22 associations recovered identically. The main result is that physically different observables can refine an existing spatial hierarchy without redefining its me

We propose endogenous interpretation: program, interpreter, machine, and derived execution language are not disjoint semantic objects but different parameterizations of one executable state-transition relation. Each realization carries an implicit constraint bias, the structural restrictions imposed by its instruction basis, state encoding, control transfers, and finite substrate. From this viewpoint the "semantic gap" metaphor is misleading for operational semantics: a semantics-preserving transformation cannot remove the computation required for execution; it redistributes it over another state space and transition granularity (semantic diffusion). Reverse engineering is then the recovery of a lower-complexity semantic contraction. We give a first quantitative instance. For finite-state realizations observed up to bisimulation, the contraction is canonical and computable in almost-linear time, yielding an abstraction-independent diffusion ratio; for deterministic realizations this extends to trace equivalence. For nondeterministic realizations observed up to traces, minimal contraction is

Equilibrium phase transitions are conventionally categorized into first-order and continuous phase transitions. Far from equilibrium, many new transitions emerge during the real-time evolution of quantum systems. One such transition is the temporal entanglement transition (TET) characterized by the nonanalyticity of the entanglement spectrum. So far, all TETs occur through a linear crossing of the leading Schmidt levels in different symmetry sectors, resembling a first-order transition in equilibrium. A natural question is whether a continuous TET, featured by entanglement spectrum touching, is possible. In a periodically driven transverse $J_1$-$J_2$ Ising chain, we show that two such TETs can merge into a critical touching, where the leading levels meet tangentially without exchanging, realizing the temporal analog of a continuous phase transition. Near the critical frequency, the temporal separation of the two TETs vanishes continuously and its derivative with respect to frequency diverges, a nonanalytic signature that is absent in a single first-order TET. This finite-frequency touching arises from the interplay between a weak symmetry-preserving perturbation of the product initial state and Floquet corrections. Further extending the frequency scan reveals a second critical touching at a higher frequency, and the two critical frequencies enclose a finite window with no TET. These features can be understood from a second-order Floquet Hamiltonian and persist across a broad range of coupling ratios, establishing the critical touching as a distinct form of TET.

Large language model agents are typically deployed with predefined configurations, although the required model capability, context, tools, permissions, memory, and computational resources can vary substantially across tasks. This study develops and evaluates AgentBetta, an adaptive AI Nano-Agent framework that represents these factors as an executable configuration and updates them through verification-driven diagnosis, selective expansion, and verification-based counterfactual contraction. The evaluation distinguishes controlled mechanism validation from external agent comparisons. On the AB-ConfigBench benchmark, AgentBetta achieved 91.38% verified success while reducing median context allocation from 64,000 to 8,000 context characters and median tool exposure from five tools to zero compared with the fully provisioned configuration. The configuration-deficiency diagnosis achieved a macro-F1 score of 0.819 with precision of 1.000 across the evaluated dimensions, and selective expansion avoided unnecessary changes to unrelated configuration dimensions. Post-success contraction preserved verification outcomes in 56.41% of evaluated one-dimension contraction probes, indicating that some successful configurations contained removable capability under the tested conditions. External evaluations indicate that adaptive configuration can improve the balance between verified task completion and capability exposure; however, the results vary across benchmarks and agent families. In particular, the cross-family replication did not reproduce the primary-backbone accuracy ordering, and specialized systems remained advantageous for certain task domains. These results support interpreting AgentBetta as a configuration-adaptation mechanism that regulates capability allocation and inference expenditure rather than as a universal replacement for specialized agent architectures.

In this work, we construct models compatible with ACT constraints on inflation in the framework of $f\left(Q, \tilde R\right)$ gravity, with $f(R)$ gravity coupled to a scalar field. The $f\left(Q, \tilde R\right)$ theory is equivalent to $f(Q, B)$ gravity models or $f(Q, C)$ gravity where $B$ or $C$ is the difference between $Q$ and the scalar curvature $\tilde R$ in Einstein's gravity, so, $B=Q-\tilde R$ or $C=\tilde R - Q$. Using appropriate reconstruction techniques compatible with $f\left(Q, \tilde R\right)$ theory, we propose two functional behaviors of the Hubble rate $H$ that satisfy the constraints, and we construct models that realize these behaviors of the Hubble rate $H$ using $f\left(Q, \tilde R\right)$ gravity and $f(R)$ gravity coupled with a scalar field. We also discuss the reheating stage after inflation. Although the scenario of $f(R)$ gravity coupled with a scalar field is well-known, since the model can be rewritten in the form of Einstein's gravity coupled with two scalar fields, the non-trivial problem occurs in the $f\left(Q, \tilde R\right)$ gravity case, because the number of dynamical degrees of freedom in $f(Q)$ gravity is not settled.

Let $R$ be a continuous von Neumann regular ring with no nonzero abelian idempotents. We prove that every element of $R$ is a single additive commutator. In particular, every operator affiliated with a type $\mathrm{II}_1$ von Neumann algebra is a commutator of operators affiliated with the same algebra. Our commutator theorem, together with a theorem of Jaikin-Zapirain and López-Álvarez, implies that, if $R$ is a nonzero unital algebra over a field $\mathbb{K}$ of characteristic zero, the quotient division ring of the first Weyl algebra over $\mathbb{K}$ embeds unitally into $R$. We also give a direct proof of this consequence.

Novel view synthesis is a key task for dynamic scene reconstruction, where high rendering speed is essential for applications such as virtual reality. Existing deformable Gaussian Splatting methods achieve high-fidelity dynamic scene modeling, but still face limitations in memory usage and rendering efficiency due to the large number of redundant Gaussians. To address these challenges, we propose Geometry-Aware Redundancy Optimization (GARO), a unified redundancy measurement framework in the adaptive density control stage of the traditional dynamic scene reconstruction pipeline. This framework first selects low-gradient candidates using an optimization activity assessment strategy, and then evaluates geometric complexity through low curvature analysis to further filter and prune redundant points, resulting in a compact and expressive Gaussian representation. Extensive experiments on synthetic and real-world datasets demonstrate that GARO achieves robust trade-offs between quality and speed, with PSNR remaining stable and rendering speed improved by 2x, validating the efficiency and effectiveness of GARO.

As part of HVAC systems, residential heating provides foundational infrastructure for human habitation in cold weather. However, research on how residents interact with HVAC systems, particularly heating systems, remains fragmented across architecture, engineering, informatics, physiology, psychology, sociology, and design. Based on 541 studies from these fields, this review integrates interdisciplinary research on Heating in Human-HVAC Interaction in smart homes.The resulting synthesis is conceptualized through the Situated Interaction Dynamics of control and feedback between users and systems. User-initiated interactions involve monitoring past and present system performance and planning future operation, while system-initiated interactions rely on sensor networks to trigger automation or provide information enabling user action. These interaction dynamics connect Residents' Experience and Practices with Heating in HVAC System Mechanics. Residents' Experience and Practices include thermal comfort and energy management, where thermal comfort involves both individual physiological and psychological experiences of indoor climate and social practices shaped by norms, empathy, and negotiation among cohabitants. Heating in HVAC System Mechanics includes thermal conditions and energy performance. Thermal conditions concern the regulation of air temperature, mean radiant temperature, air velocity, and relative humidity, while energy performance concerns efficiency and environmental impact. This overview highlights four interdisciplinary tensions: sensed versus lived conditions, personalization versus negotiation, efficiency versus health, and automation versus agency. The resulting framework offers a conceptual lens to interpret heating interactions and design Human-HVAC Interaction that balances IEQ-driven healthy thermal conditions, affordability, and sustainability.

The increasing reliance on mobile phones has made phone-induced pedestrian distraction increasingly prevalent. Activities such as texting, watching videos, and making phone calls have become significant contributors to traffic accidents. Reliable detection of pedestrian distraction is essential for autonomous vehicles, as it improves situational awareness and enables timely risk assessment, thereby supporting safe motion planning and vehicle control. We propose a multimodal fusion Transformer (MFT) for detecting phone-induced pedestrian distraction. MFT jointly extracts skeletal dynamics from body pose keypoints and visual appearance features from pedestrian images, effectively leveraging the complementary information provided by the two modalities. A cross-modal attention module is proposed to capture inter-modal dependencies through multi-head cross-attention, facilitating effective fusion of complementary information across the two modalities. Then, a temporal attention fusion module, implemented with a Transformer encoder, is employed to capture temporal dependencies. MFT is trained and evaluated on a manually annotated dataset comprising 287 pedestrian instances with 20,741 images. Extensive experiments demonstrate that MFT attains an overall accuracy of 95%, exceeding the performance of six baseline approaches by 6%.

This paper develops a certified imitation-learning framework for approximating model predictive control (MPC) policies with feedforward neural controllers and validates it on autonomous-vehicle lateral control. An exact finite-horizon Q-loss is constructed by fixing the learner's first steering action in the expert MPC problem and re-optimizing the remaining horizon, thereby measuring its downstream optimal-control consequence rather than only pointwise action mismatch. The neural policy is represented as a linear fractional transformation (LFT) interconnection with activation nonlinearities described by sector integral quadratic constraints (IQCs). Combined with a quadratic Lyapunov condition, this representation yields a differentiable certification margin based on the largest eigenvalue of the Lyapunov-IQC matrix. The margin is enforced during training through a logarithmic barrier, while certified Dataset Aggregation (DAgger) and safe projection keep data-aggregation rollouts within the certified policy set. Experiments on a CAD-referenced autonomous-vehicle platform with AprilTag localization and real-time steering demonstrate the resulting closed-loop performance.

Overcoming the approximately 60~mV/dec Boltzmann limit of the subthreshold swing remains a central challenge for low-power nanoelectronics. Negative-capacitance field-effect transistors address it by amplifying the channel potential, but their effectiveness is constrained by capacitance matching and by the quantum capacitance of conventional channels. Here we model a negative-capacitance bandgap-change field-effect transistor (NC-BCFET), in which the ferroelectric gate stack instead amplifies the electric field that opens the bandgap of the channel, using experimentally proven materials: bilayer graphene as a channel with an electrically tunable bandgap and Al$_{0.55}$Sc$_{0.45}$N as the negative-capacitance ferroelectric. Our self-consistent framework couples a four-band tight-binding Hamiltonian with GW-corrected screening, a quasistatic Landau–Devonshire ferroelectric response, and ballistic Landauer–Büttiker transport. Negative capacitance amplifies the interlayer potential difference, compressing the switching window 37-fold relative to a dielectric-gated control. At 300~K, the optimized NC-BCFET achieves a subthreshold swing of 15~mV/dec, four times below the Boltzmann limit, improving to 2.5~mV/dec at 100~K. The room-temperature on/off ratio is limited to $\sim 10^{2}$ by the bandgap of bilayer graphene. The concept extends directly to materials with larger field-tunable bandgaps, where deep subthermionic switching and high on/off ratios can be achieved together at room temperature.

Task-oriented grasping (TOG) requires robots to grasp functional parts of objects (e.g., the handle of a mug for pouring), yet these affordance regions are frequently occluded in cluttered scenes. Active perception via next-best-view (NBV) planning can resolve such occlusions by moving the camera for more informative observations. However, existing NBV methods typically optimize viewpoints for grasping the target object as a whole without distinguishing which part is task-relevant. A naive adaptation, fully scanning the target object before predicting the affordance, wastes most of the viewpoint budget on task-irrelevant surfaces (e.g., the mug body for pouring). To address this, we propose ATAP, an Affordance-Targeted Active Perception framework that shifts viewpoint planning from exhaustive target scanning to targeted affordance verification. ATAP hypothesizes the occluded target geometry via a generative shape prior and predicts the affordance distribution over the imagined complete surface. In cluttered scenes, severe occlusion can make the location of the hidden affordance ambiguous, leaving multiple locations plausible given the partial observation. ATAP therefore introduces an uncertainty-aware viewpoint planner that jointly optimizes expected entropy reduction over these competing hypotheses and expected affordance verification gain from real observations. This process iterates until the affordance is sufficiently verified for grasp execution. Experiments in simulation and real-world cluttered scenes show that ATAP substantially improves the functional grasp success rate over fixed-view TOG baselines, and outperforms reconstruction-based active perception with over 57% fewer NBV steps.

Let $M$ be a complete, non-compact Riemannian manifold. We prove that its Riesz transform is of weak type $(1,1)$, with constant $2$ for real-valued functions. Consequently, it is bounded on $L^p(M)$ for $1<p\leq2$, with constants depending only on $p$, which proves the Coulhon–Duong conjecture. The proof uses an obstacle decomposition for positive self-adjoint operators with sub-Markovian semigroups. Applying this decomposition to the shifted square-root Laplacian and using locality of the Sobolev differential yields the endpoint estimate without geometric or heat kernel assumptions. We also obtain the corresponding result for Dirichlet spaces admitting a local Hilbertian differential calculus.

In this paper, we study a finite-data inverse nodal optimization problem for radial Schrödinger operators in an arbitrary fixed angular-momentum sector. The analysis is built directly at the Friedrichs endpoint and in the physical weighted space $L_d^p$, so that the singular radial geometry is retained rather than replaced by a regular one-dimensional model. The main purpose of this paper is to provide a singular Friedrichs finite-data variational framework valid in every angular-momentum sector, thereby extending the existing finite-data variational theories concerning either regular one-dimensional operators or the radial sector $\ell=0$. By means of a Volterra representation of the Friedrichs branch, we prove weak continuity and continuous Fréchet differentiability of nodal radii, exact realization of compatible same-mode nodal data, existence of optimal potentials, and finite-codimensional constraint geometry. The same framework also incorporates mixed angular-momentum and spectral–nodal observations through finite-dimensional transversality. Remarkably, a global uniqueness theorem is established for inward displacements of the unique interior node of the second mode in the radial sector \(\ell=0\). For a constant reference potential and \(p>(d+2)/2\), every such displacement admits a unique global optimizer. Unlike local inverse-mapping or one-dimensional integrability arguments, the proof first selects the admissible critical sign globally and then reduces every minimizer to a scalar mass-balance equation between a focusing ball branch and a logistic annulus branch. The strict opposite monotonicity of the two weighted masses makes the balance parameter unique, providing a global rigidity mechanism over the entire inward-displacement regime.

The cosmic evolution of galactic bars offers key insights into the dynamical history of disk galaxies. However, studies of high-redshift bars are severely hampered by observational effects such as resolution degradation and surface brightness dimming. To quantify these biases and establish a robust methodology of bar detection and measurement for high-redshift galaxies, we develop an automated pipeline integrating isophotal ellipse fitting and Fourier decomposition. We benchmark its performance using a comprehensive set of mock observations generated by artificially redshifting high-quality images of local barred galaxies to match the background and resolution of HST ($z \approx 0.5-3$) and JWST ($z \approx 1-6$) observations. We find that applying standard bar detection criteria derived from local galaxies to high-redshift data results in a severe underestimation of the bar fraction, missing $>50%$ of bars at $z>0.5$ in HST imaging and a similar fraction at $z>2$ in JWST imaging. Using adaptive, redshift-dependent criteria optimized via mock data achieves recovery rates above $60%$ out to $z \approx 3$ for HST and $z \approx 6$ for JWST. The ellipse fitting method significantly outperforms Fourier decomposition in measurement reliability. While ellipse fitting systematically underestimates bar length by $\sim 4%$–$36%$ due to surface brightness dimming, this bias is consistent and correctable ($\sigma \approx 0.88$–$2.22$); in contrast, the Fourier method is prone to catastrophic failures ($\sigma \approx 2.40$–$6.14$) in the regime of low signal-to-noise ratio. We propose a standardized workflow that prioritizes mock-derived adaptive criteria and ellipse fitting, providing quantitative correction factors and empirical relations that enable future studies to recover intrinsic bar statistics from raw observations.

NLO simulations are essential for LHC physics analyses but are expensive, as they are not only slow but also lead to negative weights, which imply the need to simulate much larger samples of events. Folding is a powerful technique to reduce negative weights but is itself expensive. In this paper I propose “vectorized folding” as a new idea to speed up these calculations using SIMD and GPUs, and I present its CUDACPP-based implementation for MG5aMC in MadVfold, including its extension for unfolded NLO event generation. Preliminary results show overall speedups around 6x to 9x with folding and 3x without it. This work is based on a test-centric, LLM-assisted software development process.

We construct closed embedded ${\lambda}$-hypersurfaces in ${\mathbb{R}^{n+1}}$ of topological type ${S^1\times M}$ for any ${\lambda \ge 0}$, where ${M\subset\mathbb{S}^n}$ is an isoparametric hypersurface with equal principal curvature multiplicities. This extends previous results of Angenent, McGrath, and Riedler (for ${\lambda = 0}$) as well as Cheng-Wei and Ross (for ${\lambda > 0}$), and in particular completes the missing case of Riedler's construction.

Agentic systems increasingly fulfill user requests through multi-step tool workflows over files, services, and external resources. In these workflows, isolated per-call checks can miss a workflow-level failure: each call may be locally admissible, but the sequence can exceed the authorization boundary established for the session. We identify this failure mode "authorization drift." To address this problem, we present Runtime Authorization Consistency Checking (RAC), a lightweight guard at the controller-side tool-call boundary. RAC treats authorization as runtime state carried by accepted workflow steps. For each pending action, it reconstructs a trusted authorization event from controller-observed metadata and admits the call only when it remains no more permissive than the basis inherited through accepted lineage. Rejected steps are excluded from lineage, so later continuations can draw support only from accepted workflow history. Our evaluation shows that RAC reduces missed authorization drift across both controlled and planner-generated workflows. On the 1,248-workflow TraceBench suite, RAC has no missed-block cases, while the strongest Static+History baseline misses 509 of 1,008 oracle-BLOCK workflows. On a high-confidence observable subset of blind LLM-generated plans, RAC reaches 92.8% block recall, compared with 68.8% for the strongest baseline. In the real MCP filesystem planner replay, RAC stops nine unsafe continuations before server execution, with sub-millisecond p99 checking latency.

We introduce various classical, foundational results on the regularity theory of harmonic maps, with focuses on the energy minimising harmonic maps. Topics include inner and outer variations, the monotonicity identity, the $\varepsilon$-regularity theorem, blow-ups and tangent maps, and the dimension estimate and stratification of singular sets. We also discuss the analogous theory for stationary harmonic maps and mention some significant recent developments.

Three dimensional Earth-Moon transfers usually consume more fuel than planar ones, but are needed depending on the latitude of the launch site and the desired inclination of the final lunar orbit. A systematic search of the optimal 3D transfers represents a true challenge due the dimension of the parameter set to span. In this paper, we propose a new method to design 3D transfers using a tangential velocity constraint for the burns, modeled using the overdetermined constraints technique based on the Theory of Functional Connections for an optimal search of solutions in a numerically efficient way. The developed method is applied here for a systematic study of optimal 3D Earth-Moon transfers, but could be used to design any type of 3D maneuvers.

Failures of high-resolution MLLMs are commonly attributed to a visual problem, motivating zooming, cropping, and related visual interventions to recover fine-grained evidence or suppress interference. Yet recent studies suggest that relevant visual evidence is already encoded in intermediate representations, indicating that visual-side improvements alone insufficient. This raises a natural question: does the remaining bottleneck lie in the text that guides visual search? We identify a previously overlooked linguistic bottleneck: questions formulated for answering do not necessarily specify the visual evidence required for localization. To address this mismatch, we introduce EviSpec, a training-free compiler that derives complementary evidence specifications while preserving the original question for final reasoning. We further validate it through matched-control experiments that isolate the roles of evidence specification and localization. With the search budget fixed, structured evidence specifications yield an 8.6% relative gain over generic requests. With evidence geometry matched, the evidence localized by EviSpec yields a 14.8% relative gain over random evidence. Together, these controls isolate the benefit of specifying what evidence to seek rather than merely expanding visual access. Across all five MLLMs, EviSpec consistently improves upon the corresponding baseline on each of the three benchmarks, yielding average relative gains of 10.4%, 8.8%, and 12.4% on V\textsuperscript{*}Bench, HR-Bench-4K, and HR-Bench-8K, respectively. Beyond high-resolution reasoning, EviSpec also achieves state-of-the-art performance on VQA and hallucination-focused benchmarks.

We study Borel-Tukey morphisms between almost-homogeneity relations. For finite $r,s\ge1$ and $m,n\ge2$, we give the complete one-stage classification: $\mathbf{Hom}^r_m\to_{\mathrm{BT}}\mathbf{Hom}^s_n$ holds exactly when $r>s$, or when $r=s$ and $m\ge n$. Our main result is that binary triple homogeneity requires exactly three sequential binary pair stages: $\mathbf{Hom}^2_2;\mathbf{Hom}^2_2;\mathbf{Hom}^2_2\longrightarrow_{\mathrm{BT}}\mathbf{Hom}^3_2$ and $\mathbf{Hom}^2_2;\mathbf{Hom}^2_2\not\longrightarrow_{\mathrm{BT}}\mathbf{Hom}^3_2$. We also construct Borel morphisms $\mathbf{Hom}^2_2;\mathbf{Hom}^2_3\longrightarrow_{\mathrm{BT}}\mathbf{Hom}^3_2$ and $\mathbf{Hom}^2_2;\mathbf{Hom}^2_2\longrightarrow_{\mathrm{BT}}\mathbf{Hom}^2_k$ for every finite positive $k$.

We present a physics–informed neural solver for the generalized Falkner–Skan boundary-value problem that combines an exact boundary–admissible trial representation with adaptive regional residual balancing (ARRB). The three prescribed boundary conditions are embedded analytically, eliminating boundary-condition penalty terms while allowing the finite–domain streamfunction value and wall shear to be determined by the governing equation. The residual is divided into wall, middle, and tail regions, and a physical global residual is reconstructed from regional mean-squared errors using region-length fractions. Safeguarded exponential–moving–average inverse–gradient coefficients adaptively balance the regional contributions during Adam optimization. A deterministic 800-point residual monitor is used for checkpoint selection, followed by two deterministic L–BFGS stages minimizing the physical global residual. The method is tested on the Blasius, favourable-pressure–gradient Falkner–Skan, and Pohlhausen cases. For \((\beta_0,\beta_1)=(0.75,0.50)\), the predicted wall shear is \(f''(0)=0.8997161394\), compared with the finite-domain reference \(0.8997168085\), giving an absolute error of \(6.691\times10^{-7}\). The residual MSE is \(6.357\times10^{-9}\), while the relative \(L_2\) errors in \(f'\) and \(f''\) are \(2.848\times10^{-6}\) and \(4.188\times10^{-5}\). A matched single–seed ablation shows that ARRB reduces residual MSE by 52.81% relative to global–residual training and by 50.24% relative to equal-regional weighting. Wall–shear errors are reduced by 77.30% and 72.37%, respectively. These results support ARRB as an accuracy-oriented residual-conditioning strategy, while multi–seed experiments remain necessary to quantify optimization variability.

Perception for embodied agents is video-based, often multi-view (ego, exo, or both), and inherently continual, with simultaneous task and viewpoint shifts. Yet continual learning (CL) remains dominated by exo-only recognition tasks, obscuring behavior under these real-world coupled shifts. We introduce Continual Ego, E}xo, and Ego-Exo Learning (CE$^4$L), a unified multi-view CL benchmark spanning four representative tasks: cross-view referenced skill assessment, temporal action segmentation, cross-view association, and action anticipation & planning. CE$^4$L highlights challenges largely absent in prior CL benchmarks, including cross-view correspondence, view-dependent asynchrony, and heterogeneous semantic objectives. To this end, we propose Video Incremental Subspace-routed Task Adapters (VISTA), a parameter-efficient baseline method that stores task-specific updates in lightweight adapters and performs training-free routing via residual distance to task-specific whitened subspaces estimated from second-order statistics. Extensive experiments demonstrate the significantly varied efficacy of representative CL methods across CE$^4$L settings, while VISTA is consistently competitive and achieves state-of-the-art overall performance. Our source code for benchmarks and methods is available at https://github.com/AnAppleCore/CE4L .

We present Elevator-VIGS, a visual-inertial 3D Gaussian Splatting SLAM system that keeps tracking and mapping through elevator rides. Inside a moving elevator, the two sensors are in conflict. The camera sees only the robot's motion relative to the elevator, while the IMU senses that motion plus the elevator's motion relative to the world. This conflict is challenging for existing visual-inertial estimators. If vision dominates, the estimator tracks only the robot's motion within the elevator and misses the elevator's rise, and if the conflict remains, the estimator diverges. We observe that the conflict comes from forcing both observations into a single coordinate frame. We instead estimate the robot's pose in the elevator's coordinate frame, and the elevator's motion relative to the world as a per-keyframe transport state, the elevator's rise and vertical velocity, within dense visual-inertial bundle adjustment. Elevator-VIGS detects rides zero-shot with a vision-language model and a depth network, and constrains the transport state at the departure and the arrival. We record real-world and simulated elevator sequences. On these sequences, Elevator-VIGS achieves state-of-the-art tracking and rendering performance. On four elevator-free public benchmarks it keeps the state-of-the-art performance of VIGS-SLAM. Project page: https://ruizhou-cn.github.io/elevator-vigs/.

Large language model (LLM) agents increasingly execute multi-step workflows through tool use and interaction with users and environments. However, current agent evaluations are largely English-centric, limiting our understanding of agent capabilities in multilingual settings. We introduce BabelFlow, a benchmark-general agentic workflow that adapts existing agent benchmarks to new languages by analyzing runtime dependencies, coordinating structure-preserving translation, and combining multi-layer verification with human review to preserve task and evaluation semantics. Using BabelFlow, we construct BabelArena, a task-aligned benchmark comprising 16,146 instances derived from 702 canonical tasks across four benchmark families, 13 domains, and 23 languages. Experiments with five frontier models show that no single model dominates across benchmark families and that cross-language disparities extend well beyond task success. Lower-resource languages exhibit distinct failure patterns, with larger shares of tool-use and control-flow errors rather than answer-quality errors alone, pointing to gaps in reliable task execution across the resource levels of these languages. On the same tasks, agents in low-resource languages also consume substantially more tokens than in English (up to roughly twice the input) without proportional increases in interaction length, and language consistency degrades further on tasks requiring structured output, where switches are directed overwhelmingly toward English. We believe BabelArena provides a foundation for advancing research on reliable and efficient multilingual agents.

Triangle counting is one of the most fundamental problems in network analysis. Given the massive sizes of real-world graphs, there is a long history of small-space streaming algorithms providing accurate estimates for this problem. However, most of the results focus on estimating the total triangle count or the number of triangles incident to individual nodes. In practice, one often wants fine-grained information to understand how triangles are distributed, as captured by clustering coefficients. In particular, a standard network analysis task requires computing the binned degree-wise clustering coefficient distribution, which provides a rich and informative summary of the structure of the graph. In this work we present BOLIDE, the first efficient and practical algorithm for estimating binned degree-wise clustering coefficients in streaming. Our algorithm makes a single pass over the edge stream, and is allowed to store only a small fraction of the total number of edges. BOLIDE carefully combines different sampling strategies to efficiently gather degree and triangle information across sets of nodes. As a result, our algorithm provably approximates the binned degree-wise clustering coefficients, and provides guarantees on the amount of memory used. Our experimental evaluation shows that BOLIDE accurately estimates clustering coefficient distributions while efficiently processing large datasets with billions of edges and triangles.

End-to-end autonomous driving maps current observations directly to future trajectories, yet those trajectories must remain valid as the scene evolves. Future state modeling aims to address this temporal mismatch, but general representations often contain information unrelated to ego planning and affect trajectory generation only through auxiliary supervision, static conditioning, or proposal evaluation. We propose FeasibleFlow, a one-step end-to-end generative framework that jointly transports a configuration-space feasibility field and multimodal ego trajectories. Our Asymmetric Joint MeanFlow uses the pathwise Jacobian-vector product in the MeanFlow identity to incorporate field evolution into trajectory transport. Because safety feedback is sparser than progress feedback, we further introduce the Anchor-relative ranker (ARR) and Pareto-ReinFlow to balance safety and progress in candidate selection and generation, respectively. Experiments on the NAVSIM benchmark demonstrate the strong performance of FeasibleFlow and validate both the joint transport of feasibility and trajectories and the proposed safety-first mechanisms.

Existing wearable exoskeleton architectures are typically constrained by a single mechanical output modality, providing either joint torque around an anatomical joint or linear traction along a limb-training-oriented direction, which limits adaptability to diverse training scenarios. This letter presents a cable-driven switchable actuator (CDSA) that can rapidly switch between torque and tension modes while centralizing all sensing and actuation components at the proximal drive unit. A Coupled Movable Pulley Mechanism (CMPM) provides tension amplification at the distal end-effector, while a bidirectional Cable-Driven Ratchet Mechanism (CDRM) enables mode switching and preload regulation. To eliminate the need for distal instrumentation, multi-source proximal sensors are integrated with a data-driven fusion model to estimate distal output forces. An adaptive dual-mode force control strategy based on iterative learning control (ILC) is further developed. Platform experiments demonstrate transmission efficiencies of $(92.4 \pm 2.0)\%$ and $(96.5 \pm 3.3)\%$ in the torque and tension modes, respectively, along with a tension amplification ratio of $2.77 \pm 0.10$ under tension mode. Tracking tests on simulated knee-joint gait trajectories and short-stroke tension profiles yield stable control, with RMSEs of $(4.52 \pm 0.51)\%$ and $(3.15 \pm 0.19)\%$ of the uncontrolled peak value, respectively. Finally, seated human-coupled experiments validate the system's controllable force generation in both joint-torque and linear-traction application modes.

Robots operating in human-centered environments are typically designed to execute explicit instructions, and most robot-learning datasets likewise pair observations with task instructions or low-level actions. Although recent work has begun to explore proactive embodied assistance, existing resources target different settings and action levels, leaving real-world human-centered multimodal decision-making underexplored. We formulate this problem as Proactive Robot Action Reasoning (ProRobo), an upstream cognitive decision problem in which a robot must determine which action to take based on multimodal human and environmental cues without explicit action instructions. To support ProRobo, we introduce ProAction, a real-world multimodal dataset containing 10K samples of visual observations, audio signals, and text inputs across 12 daily-life scenarios in five common scenes. To construct cognitively grounded high-level action supervision, we develop a two-stage human-in-the-loop pipeline that combines appraisal-guided candidate generation with Affective Theory-of-Mind-guided human refinement, explicitly incorporating contextual judgment about human states, urgency, feasibility, and potential risk into action annotation. Based on this supervision, we benchmark representative Multimodal Large Language Models (MLLMs) and introduce MMC2Act, a reference model that implicitly learns the mapping from multimodal observations to cognitively grounded high-level actions. Experiments across modality settings, subject-disjoint generalization, cross-dataset transfer, and human evaluation show that general-purpose MLLMs struggle with proactively reasoning high-level actions from multimodal cues, whereas training on ProAction substantially improves performance.