Computer Science & Mathematics

We introduce the Literature and Linguistics Analysis Benchmark (LALA), comprising 236 open-ended questions written by experts in linguistics and literary studies, together with expert-provided ideal answers. By comparing models within families and at similar training scales using human and LLM judges, LALA identifies differences in linguistic competence associated with post-training, offering a framework for evaluating future models and guiding targeted improvements. We administered questions to 17 language models and evaluated their responses using both human expert assessment and two LLM evaluators, Sol and Qwen. Base models displayed little of the linguistic competence assessed by the benchmark, whereas their post-trained versions performed markedly better, suggesting that post-training plays a central role in making linguistic knowledge accessible and usable. Parameter count, by contrast, showed no statistically significant association with partial-credit performance, and competencies remained unevenly distributed across linguistic domains. We interpret this pattern as evidence that post-training adds linguistic competencies piecemeal, without producing a general linguistic intelligence capable of extending them beyond the learned policies that support them. Models performed best on logical relations between statements and selected discourse-level tasks, while even the strongest post-trained models continued to struggle with ambiguity, literary style, reference, and misleading assumptions. Human and Sol evaluations produced nearly identical model rankings in both the original and ReliableEval runs, supporting the replicability of the comparative results.

Inferring target position from WiFi channel state information (CSI) is challenging due to the noisy and heterogeneous nature of CSI-derived observations. Measurements such as angle, delay, and Doppler provide only imperfect information about target position and velocity, particularly in dynamic multipath environments where observation reliability varies. Motivated by this challenge, this paper proposes a physics-informed Bayesian state space framework for passive CSI-based tracking. We introduce a context-conditioned deep state space model (DSSM) that captures temporal motion and environmental dynamics, while adaptively modeling time-varying observation uncertainties. By incorporating geometric constraints, this framework enables heterogeneous CSI measurement modalities to jointly regularize the consistency and confidence of the tracking process. Experimental results demonstrate that the proposed framework yields smoother trajectory estimates with reduced temporal fluctuations and outperforms existing methods in tracking accuracy.

Time-sensitive neuromorphic architectures require predictable interconnects to preserve precise event timing. Statically scheduled time-division-multiplexed (TDM) circuit-switched networks-on-chip (NoCs) provide deterministic latency at the cost of linear configuration-memory scaling with schedule depth. Furthermore, jointly optimizing task placement, routing, and scheduling creates an intractable combinatorial search space. Existing deployment frameworks also ignore algorithmic axonal delays by forcing immediate spike injections that increase network contention. This paper proposes a hardware-software co-design to address these challenges. ChroNoC is a tableless TDM NoC that embeds precomputed full-grid broadcast trees into combinational logic as a fixed routing substrate. This fixed substrate reduces deployment to a spatial search via Delay-Aware Reservations and Spatial Annealing (DARSA). DARSA converts axonal delays into periodic scheduling slack to stagger network injections and reduce communication overhead. In a 130nm CMOS process, ChroNoC reduces per-hop energy consumption by more than 75% relative to the evaluated circuit- and packet-switched baselines. On UCI-HAR and N-MNIST workloads, DARSA reduces per-timestep communication cycles by approximately 60% and 45%, respectively, compared with immediate transmission, while also reducing provisioned communication memory. These results demonstrate how combining lightweight TDM switch design with delay-aware deployment improves both hardware efficiency and system-level communication performance while preserving neural event timing.

The simultaneous surveillance and containment (SSC) problem arises in multi-robot applications such as perimeter security and environmental monitoring, where robots must survey a region of interest and coordinate to enclose targets once they are detected. Surveillance favors spatial dispersion for coverage, whereas containment requires target-centric cohesion for enclosure. This conflict makes heuristic switching unstable, and the instability becomes more pronounced as larger robotic networks introduce more inter-agent and agent-target relations. End-to-end DRL reduces reliance on hand-crafted switching, but conventional single-stream policies often mix surveillance and containment cues in one shared feature space that could entangle dispersion-oriented and target-centric signals and weaken task transitions. In this paper, we propose the Dual-Stream Gated Network (DSGN), an architecture designed to address scale-amplified task-transition difficulties by separating surveillance and containment into specialized representation streams and coordinating them through a context-aware gating mechanism. A probabilistic prior further regularizes the gating process, reducing ambiguous intermediate decisions and supporting clearer behavioral transitions under large-scale interaction complexity. Evaluations show that DSGN maintains strong task efficacy and scalability under constant-density cross-scale settings. Additional ablation and phase-specific evaluations show that the proposed dynamic gating mechanism and probabilistic prior support effective task arbitration and reliable mode switching in both pure surveillance and task-switching scenarios. The source code is available at https://github.com/zuozysama/DSGN.

Safe indoor operation requires autonomous tracking of objects that are not directly visible to the robot. Optical non-line-of-sight (NLoS) methods depend on ambient lighting and scene geometry or on weak multi-bounce returns that need long acquisitions. Millimeter-wave (mmWave) radar offers an attractive alternative: walls act as reflectors for mmWave signals, and frequency-modulated continuous-wave (FMCW) radars enable fast, practical measurement of range, angle, and velocity. However, the low resolution of these radars leads to spatially ambiguous observations. The Doppler information, although particularly sensitive to target motion, remains underexplored in indoor NLoS localization: existing methods emphasize spatial representations and either collapse or discard the Doppler dimension before learning. To leverage the complete Range-Azimuth-Doppler (RAD) information, we acquire the first indoor NLoS radar dataset that pairs sequential RAD measurements with synchronized position and velocity ground truth under controlled variations in motion: 576 sequences totaling 720,856 frames across 12 trajectories of a mobile robot moving along varying speeds, starting positions, and material configurations. We then formulate NLoS tracking as sequential position-velocity estimation rather than frame-wise localization: a shared 3-D encoder processes the full RAD representation of consecutive frames, separate recurrent branches estimate position and velocity, and post-inference refinement combines the two into a motion-consistent trajectory. Our method achieves 8.95cm position and 1.75cm/s velocity errors on held-out trajectories and generalizes to unseen scenarios involving a moving human target. Together, our dataset and tracker establish a benchmark for tracking hidden moving targets with a low-cost radar.

In integrated sensing and communication (ISAC), when the sensed target is itself an eavesdropper, the illumination required for sensing coincides with the leakage that secrecy must suppress. This paper shows how a beyond-diagonal reconfigurable intelligent surface (BD-RIS) with movable elements decouples the two tasks in a monostatic setting, where the echo returns to the base station through the surface. We prove that a lossless reciprocal BD-RIS maps $K$ incident fields onto $K$ prescribed fields if and only if the two sets share the same Gram matrix and their bilinear Gram matrix is symmetric; unlike a diagonal RIS, it can thus steer a two-dimensional incident subspace onto any prescribed subspace. With a single stream, sensing incurs an unavoidable rate loss, whereas a closed-form zero-forcing design with a confidential and a Gaussian sensing stream nulls the data at the target and the sensing at the user, so that sensing costs only an SNR offset. Since all variables follow in closed form for any placement, the joint design reduces to a monotone search over the element positions, without alternating optimization. Simulations show that the dual-stream design loses only $1.1$ bit/s/Hz when the sensing requirement grows by $30$ dB, against $8.3$ bit/s/Hz for a single stream, while movement, interconnection, and the robust derivative-null design add up to $1.9$, $0.9$, and $2.6$ bit/s/Hz, respectively.

This study presents a two-layer optimization framework to determine the optimal sizing of PV, wind, and battery storage co-located with existing hydroelectric plants to support rapidly increasing data center loads while minimizing cost, grid reliance, and transmission upgrades. The outer layer uses Particle Swarm Optimization (PSO) to select capacity combinations, while the inner layer performs rolling-horizon dispatch simulations to model hourly energy balance among PV, wind, hydro, and BESS, considering operational constraints. Two case studies were analyzed. At The Dalles Dam (Oregon), favorable wind resources and land availability enabled deployment of 111.2 MW PV and 117.9 MW wind, with no BESS required. Hydro flexibility allowed the system to meet the load with minimal grid support. In contrast, Conowingo Dam (Maryland) faced land constraints and limited renewables to 130 MW PV, leading to higher grid dependence. Co-locating hydro with renewables significantly enhances reliability and cost effectiveness by leveraging hydro's dispatchability to balance daily and seasonal variable generation from PV + Wind, reducing the need for storage or costly grid imports. This analysis provides a framework to meet the 99.9% availability targets of this new load type on the grid with sustainable energy resources.

This article investigates dynamic metasurfaces as an electronic realization of generalized pinching antennas (GPA). By jointly optimizing the locations and radiation coefficients of activated elements, the dynamic metasurface-enabled GPA (DM-GPA) architecture exploits additional design flexibility to enhance spatial reconfigurability beyond that attainable through antenna placement in the conventional pinching-antenna system (PASS). Based on this architecture, we study weighted sum rate (WSR) maximization for multiuser downlink transmission. 1) For two-user transmission, the sum-rate improvement of DM-GPA over conventional PASS is quantified through a lower bound that reveals when the benefit of channel orthogonalization achieved by adjusting the radiation coefficients outweighs the channel power loss caused by the Lorentzian constraint. Furthermore, when conventional PASS yields a rank-deficient channel, DM-GPA is shown to achieve spatial multiplexing and transmit diversity, with a strict sum-rate advantage at sufficiently high signal-to-noise ratios. 2) For general multiuser transmission, a weighted minimum mean-square error (WMMSE)-based alternating optimization algorithm is developed for joint pinching beamforming and digital precoder design. A sequential element-wise search is developed to optimize the locations of the activated elements, with a closed-form solution for the radiation coefficient at each candidate location. Numerical results are provided to demonstrate that: (i) DM-GPA achieves higher WSR than other PASS architectures and fixed-array benchmarks, (ii) increasing the number of activated elements further improves the WSR of DM-GPA, and (iii) radiation coefficient adjustment is particularly beneficial when candidate activation locations are limited.

Few-step distillation for autoregressive (AR) video generation commonly first initializes an AR student and then refines it with distribution matching distillation (DMD). This refinement process faces two challenges. First, although initialization already yields a reasonably capable student, DMD obtains the refinement direction indirectly by estimating the full student score before forming the teacher-student correction. Second, detached-history training prevents the generator from assigning credit to a current chunk for its influence on subsequent chunks. We introduce Leta, a few-step distillation framework that addresses these challenges with ScoreDelta and FutureCredit. ScoreDelta directly estimates the teacher-student correction from denoising residuals, while FutureCredit learns an auxiliary cross-chunk update signal from vector-Jacobian products (VJPs) of next-chunk teacher residuals. Leta combines these signals into a joint teacher-guided update field in a single forward pass during generator updates. Teacher evaluations and credit VJPs are used only for estimator training. For ScoreDelta, we establish population equivalence to score-difference estimation and a conditional approximation advantage under limited capacity in a fixed-feature linear model. In our controlled experiments, Leta improves over DMD by 0.9 points on VBench and 0.66 on VideoAlign. This advantage persists at a higher generator update frequency (a generator-to-estimator update ratio of G:E=1:1), where Leta improves over DMD by 1.6 on VBench and 0.35 on VideoAlign. Project page: https://scdzx.github.io/Leta/.

Open-vocabulary 3D occupancy prediction is important for embodied agents operating in complex environments whose semantic categories extend beyond fixed taxonomies. While recent studies have explored open-vocabulary occupancy in outdoor driving scenarios, indoor environments remain challenging due to dense occlusions, complex geometry, and fine-grained long-tailed semantics. To reduce the need for expensive voxel-level semantic annotations, we study a geometry-only supervision setting requiring only binary occupancy labels. We build upon Language-Embedded Gaussians as a unified representation coupling 3D geometry with language-aligned semantics. For geometry learning, we introduce an opacity-aware Poisson-based Gaussian-to-Occupancy formulation that enables stable volumetric aggregation under binary occupancy supervision. For semantic learning, we render Gaussian features into images and align them with a frozen open-vocabulary model, together with Progressive Temperature Decay to reduce feature mixing during Gaussian splatting. Building on this framework, we further introduce Instance-Level Semantic Consistency to alleviate fragmented object predictions, and Covisibility-Aware Support-View Selection to provide informative and less redundant multi-view semantic supervision. We also extend our framework to outdoor driving scenarios to evaluate its generality across substantially different environments. Extensive experiments on indoor and outdoor benchmarks demonstrate that our method achieves strong open-vocabulary occupancy performance, validating the effectiveness and generality of our framework. Code will be available at https://github.com/JuIvyy/LegoOcc.

Large language models (LLMs) generally lack deep causal reasoning capabilities in temporal reasoning. Inspired by document layout reordering, this paper proposes a paradigm shift from discrete narratives to causal topologies. It constructs two architectures: a training-free neuro-symbolic framework, temporal causal reasoning agent (TCR-Agent), which achieves interpretable causal deduction and counterfactual truncation through external causal graph instantiation and a symbolic intervention engine; and an end-to-end model, temporal causal reasoning former (TCR-Former), which internalizes these mechanisms into the Transformer attention space via a Causal-Topological Attention Mask and temporal span biases. Experiments on the TempoCausal benchmark demonstrate that TCR-Agent comprehensively surpasses existing training-free baselines. At the same time, TCR-Former achieves an accuracy 16.5% higher than that of state-of-the-art (SOTA) methods, outperforming large-scale models such as GPT-5.4 and DeepSeek-V3.2 with only approximately 8.6B parameters. Further analysis reveals a structure-reasoning disconnection phenomenon in existing methods: external symbolic constraints and chain-of-thought reasoning can improve reasoning steps, yet fail to enhance deep causal deductive capability. By internalizing temporal causal reasoning capabilities into the model, TCR-Former effectively bridges this gap, providing an efficient and reliable end-to-end foundation for complex reasoning scenarios.

This paper deals with the optimal state estimation problem for discrete-time Markov jump linear systems in the sense of linear minimum mean-square error where the noises corrupting the system are mode-dependent and need not be Gaussian. An orthogonal projection extension-based equation (OPEBE) is established, which can decompose a random vector into several parts where the random vector contains conditional expectation of another random vector. Using the OPEBE and some results obtained in this paper, an optimal linear filter for the system under consideration is proposed. The computational and storage load of the proposed filter does not increase with time. Also, the proposed filter is finite-dimensionally computable and has a recursive structure. Computer simulations are provided to verify the effectiveness and performance of the proposed optimal linear filter.

We give an $\tilde{O}(n/\varepsilon)$-space algorithm for the minimum cut problem in dynamic streams, resolving the space complexity of streaming min-cut up to polylogarithmic factors. Ding, Garces, Li, Lin, Nelson, Shah, and Woodruff (ITCS 2025) obtained the same $\tilde{O}(n/\varepsilon)$ space bound for insertion-only streams, while prior to our work the best known bound even for unweighted dynamic streams was $\tilde{O}(n/\varepsilon^2)$, due to Ahn, Guha, and McGregor (PODS 2012). Our main technical contribution is an $\tilde{O}(n/\varepsilon)$-space for-each cut sketch in dynamic streams, which may be of independent interest. The construction applies to weighted graphs under general $\pmΔ$ updates, a stronger model than the standard dynamic-stream setting in which weighted edges are inserted and deleted in their entirety. Despite this stronger guarantee, our algorithm is surprisingly simple and relies on very little machinery. We also obtain applications to minimum $\ell_p$-norm $k$-partitioning for every fixed $k$, weighted correlation-clustering evaluation for two clusters, and for-each cut sketching in balanced directed graphs.

Pull requests written by coding agents can pass a project's test suite while omitting a guard or a validation step. To find the missing condition, a reviewer currently relies on static analysis of the final code. An agent-generated change, however, is accompanied by a development history that a human commit does not provide. The agent's trajectory records each edit together with the agent's stated assumptions, and structural differencing shows how each changed function evolved. We present TRACER, a workflow that uses this development history as evidence for security review. TRACER tracks changes at the symbol level, treats the agent's recorded rationale as claims to be verified against the current code, accepts a repair only after assessment and functional validation, and attaches to each accepted repair an executable check that later changes must satisfy. On the SusVibes benchmark, symbol-level tracking places the vulnerable symbol in the review scope for 94.6% of agent patches, compared with 13.4% for the primary scanner warning. With two investigator models, the configuration that includes the agent's rationale yields the largest number of verified security repairs while preserving functionality, and adding AST history to it does not increase that number. A funnel analysis identifies where the remaining candidates are lost. In a study of structural-biology software, the combination of repair memory and a boundary check allows a subsequent feature request to succeed in nine of ten tasks without reintroducing the repaired vulnerability, whereas functional feedback alone succeeds in four and reintroduces the vulnerability in three.

Comet C/2025 K1 (ATLAS) underwent major fragmentation after its 2025 October 8 perihelion passage at q ~ 0.33 au. We present four epochs of Johnson-Cousins BVR imaging obtained with the Mount Laguna Observatory 40-inch reflector on 2025 November 26, November 29, December 5, and December 8, sampling the subsequent evolution of the integrated fragment-coma complex. The images were homogenized to common circular two-dimensional effective point-spread functions before morphology was measured along the projected fragment axis. We evaluated 324 profile-extraction configurations and removed solutions whose localized peak landed on the search-window boundary. The fragment-axis surface-brightness enhancement relocates inward between November 26 and 29 and outward between December 5 and 8; the sign of both changes is recovered in every valid configuration and at all tested common target PSFs from 5.3 to 6.0 arcsec. Calibrated BVR photometry using Gaia DR3 synthetic Johnson-Kron-Cousins standards independently shows a repeatable November 26-29 change in broadband spectral shape: B-V decreases while V-R increases. The curvature index S'_(V-R) - S'_(B-V) changes by +3.47 (+1.50/-1.65) percent per 1000 Angstrom, with 99.1 percent of Monte Carlo realizations positive, while the full B-R reflectivity gradient remains approximately constant. Repeating the color comparison at a fixed projected radius of 2205 km produces nearly identical changes. The combined results are consistent with continued restructuring of the unresolved post-fragmentation inner coma, while broadband imaging alone cannot uniquely separate dust-color evolution, molecular emission, and changing contributions from unresolved fragments.