Papers

In the Shortest Common Superstring problem (SCS), one is given a finite set of strings and is asked to find a shortest string containing every input string as a substring. Its best known approximation ratio is $2.466$, whereas the currently strongest upper bound on the approximation guarantee of the maximum-overlap greedy algorithm is $3.396$ (Englert, Matsakis, and Veselý, 2023), though it is conjectured to be $2$. We improve both approximation guarantees: SCS admits a $\frac{7}{3}$ approximation and the approximation guarantee of the greedy algorithm is at most $3$. The main technical ingredient of our proof is a certain inequality for optimum cycle covers of an overlap graph associated with the input strings. Every previous improvement of greedy's worst-case guarantee and the two recent record guarantees for general SCS are driven by it. We improve this inequality by pushing it to its limit: for a particular coefficient of this inequality, we show a new upper bound and prove that it cannot be improved further.

Neural combinatorial optimization solvers generate many candidate solutions per instance and report the best one found, using the same sample budget for every instance regardless of difficulty. A companion study showed that reallocating a fixed budget toward harder instances can improve solution quality, but that the standard way of measuring this improvement is biased: deciding an allocation and evaluating it on the same data can manufacture an apparent gain even when none exists. This left open what property of a workload determines whether reallocation is worth doing, and whether a policy that spends part of the budget to decide how to allocate the rest still pays once that cost is counted. This paper answers both questions through pre-registered confirmatory experiments – analysis and verdict criteria fixed before data collection – across three independently trained solvers and two ways of constructing harder workloads on the traveling salesman problem. Within the workloads we study, the deciding property is how varied the instances within a workload are in difficulty, not how difficult the workload is on average: a uniformly easy or uniformly hard workload offers little room for reallocation, while a mixed workload offers substantial room. A budget-aware policy that pays for its own information about instance difficulty recovers most, though not all, of the improvement available when that information is assumed free. Every experiment was independently recomputed from its written specification, and every correction to an earlier version – including two that weakened the paper's own claims – is reported with the direction it moved the conclusion. The paper offers a specific empirical answer and a template for verifying that answer is not an artifact of how it was measured.

We study the problem of aggregating opinions from multiple black-box experts in noisy, conflict-prone settings where expert reliability varies across inputs. Static aggregation methods, such as majority voting, fail to capture this variability and often yield unreliable outcomes under disagreement. We propose a tractable, probabilistic-circuit-based fusion framework that dynamically combines expert responses using context-specific credibility estimates, enabling principled and reliable reasoning. The framework is agnostic to the underlying experts and does not require access to their internal representations or any retraining. We empirically validate our approach on multiple-choice question answering tasks using multiple LLMs as experts, comparing against individual models and static ensemble baselines. Our method consistently improves predictive performance and produces more reliable decisions under conflict, highlighting the effectiveness of context-aware credibility modeling for robust multi-expert fusion.

Global forecasting models pool many series and learn one shared function. Gradient-boosted trees are their most common form. We measure a failure of this design that has not, to our knowledge, been documented. Train a global tree on the individual series of a hierarchy, then ask it for the hierarchical aggregate. The aggregate sits far outside the model's training range, and the forecast collapses. The model under-predicts the total by 30-50x in our production deployment, and by up to 496x in a public M5 reconstruction. The mechanism is known: beyond its training range, a tree predicts a constant. It surfaces at the aggregate because the total dwarfs every training series. The cure is not new. Per-series scaling, the preprocessing step that Montero-Manso and Hyndman (2021) recommend, prevents the collapse. So do a weighted aggregate-level training row and seasonal differencing. Our contribution is the characterization. The collapse reproduces on five panels: a production business-to-business marketplace, a synthetic hierarchy, M5, Australian Tourism, and a public business-buyer panel. It holds on three tree libraries, is invariant across training seeds, and is statistically significant. Its onset is immediate and tracks a simple support bound: a scale gap of only 1.15x already costs a third of the total. No standard configuration change prevents it: pooling every hierarchy level into training fails at scale, and the one knob that fits linear models in the leaves softens it without curing it. Rolling the forecasts forward recursively separates the cures: the aggregate-row cure re-collapses, per-series scaling degrades but stays low, and only seasonal differencing keeps its one-step accuracy unchanged. We close with a three-step procedure for diagnosing and preventing the failure in deployed systems.

As information technology advances, digital content has become widely adopted across diverse fields such as news broadcasting, entertainment, commerce, and forensic investiga?tion. However, the availability of sophisticated multimedia editing tools has significantly increased the risk of video and image forgery, raising serious concerns about content authenticity at both societal and individual levels.To address the growing need for robust and accurate detection methods, this study proposes a novel video forgery detection model that integrates both spatial and frequency-domain features. The model is built on a ResNet-LSTM framework enhanced by a Convolutional Block Attention Module (CBAM) for spatial feature extraction, and further incorporates Discrete Cosine Transform (DCT) to capture frequency domain information. Comprehensive experiments were conducted on several mainstream benchmark datasets, encompassing a wide range of forgery scenarios. The results demonstrate that the proposed model achieves superior performance in distinguishing between authentic and manipulated videos. Additional ablation and comparative studies confirm the contribution of each component in the architecture, offering deeper insight into the models capacity. Overall, the findings support the proposed approach as a promising solution for enhancing the reliability of video authenticity analysis under complex conditions.

Due to the rapid development of information technology, a huge and complex traffic network has been established across various sectors, including aviation, aerospace, vehicles, ships, electric power, and industry. However, because of the complexity and diversity of its structure, the complex traffic network is vulnerable to being attacked and faces serious security challenges. Therefore, this paper innovatively proposes a traffic network resilience recovery method based on resilience trend forecasting. In this paper, the risk value is introduced into the analysis of the network fault propagation process, and the Susceptible, Infectious, Recovered, Dead-Risk (SIRD-R) fault propagation model is established. The resilience model of traffic network, which encompasses real-time resilience and overall resilience, is constructed through the integration of network resilience bearing capacity and resilience recovery capacity. Ten, the resilience of complex traffic networks is forecasted by using long short-term memory networks, and the resilience recovery strategy of complex traffic networks based on forecasting is proposed. Finally, the effectiveness and scalability of the proposed method are demonstrated through experimental analysis conducted on a diverse range of complex traffic networks, affirming its applicability in real-world scenarios

Analytical studies of maritime wireless networks commonly assume a deterministic smooth sea surface, leaving unclear how realistic ocean conditions reshape network-level reliability. Unlike conventional intuition that sea roughness always deteriorates propagation, this work reveals a non-monotonic sea-state-induced performance transition caused by the competition between interference-null mitigation and coherent reflection loss. This paper develops a physically grounded, sea-state-aware stochastic geometry framework for maritime networks by incorporating sea surface roughness into propagation modeling. Specifically, we derive an effective reflection coefficient based on the classical Rayleigh roughness criterion, where the significant wave height explicitly characterizes the attenuation of the coherent specular reflection component caused by surface roughness. By integrating the proposed channel model into a stochastic geometry framework, we derive tractable expressions for uplink coverage probability under different sea states. Our analysis reveals a non-monotonic impact of sea roughness on network performance under the considered propagation model: moderate roughness can improve reliability-oriented coverage by mitigating destructive interference nulls, whereas stronger roughness attenuates coherent reflected energy and degrades high-SINR performance. Measurement comparisons support the underlying roughness-sensitive reflection mechanism, while rough-sea VHF results are interpreted as wavelength-specific model predictions rather than direct empirical validation.

A blind image denoising framework based on an infinite directional lower semi-frame (DLSF) is proposed for additive white Gaussian noise. The model employs scale-dependent directional analysis with resolvent regularization of the unbounded semi-frame operator. Noise variance is estimated directly in the DLSF domain by modeling the joint covariance of four directional difference channels and applying covariance whitening to obtain a chi-square statistic. A lower-tail moment estimator provides blind noise estimation without median absolute deviation. The estimated noise level is incorporated into channel-wise Wiener-type shrinkage and canonical-dual synthesis, followed by a data-consistent iterative reconstruction with automatic stopping. Experiments on three standard grayscale images at noise levels 15–30 yield a mean relative noise-estimation error of 3.28%, with average improvements of 7.45 dB in PSNR and 0.367 in SSIM. At 30/255 noise, the estimation error decreases to 1.73%, with a mean PSNR gain of 8.31 dB. Results demonstrate effective noise suppression and structural preservation, with the strongest performance on smooth and edge-dominated images.

Repository-level coding benchmarks have become the standard for evaluating coding agents, yet they inherently suffer from data leakage because they are built upon popular open-source repositories repeatedly used for training. Consequently, strong performance may reflect memorization of canonical repository cues rather than robust repository reasoning. We propose SchrodingerRepo (Schrödinger's Repository), an evaluation framework for testing coding agents under dynamically instantiated repository representations. Instead of repeatedly using a static representation of the test repository, SchrodingerRepo treats the test repository as an evaluation-time latent variable that is dynamically instantiated only when the agent enters the evaluation environment. The instantiated repository preserves the original executable behavior while eroding familiar cues such as naming conventions, file layouts, and implementation patterns through four transformation levels: problem statement reconstruction, namespace remapping, intra-file layout reordering, and functionality-preserving code rewriting. We evaluate popular LLMs on SWE-bench Verified and SWE-QA. Results show that removing familiar repository cues consistently degrades agent performance and substantially increases interaction costs across models. Further analysis reveals that the additional cost is primarily caused by increased difficulty in repository exploration and localization. These findings suggest that current coding agents may partially rely on memorized repository-side cues, highlighting the need for evaluation under dynamically instantiated repository representations.

Multimodal large language models (MLLMs) fre- quently generate text that is inconsistent with the input image. While object- and attribute-level hallucinations have received considerable attention, relational hallucinations (incorrect de- scriptions of spatial or interactive relationships between objects) remain largely unaddressed by existing post-hoc correction methods. We present RelCheck, a training-free post-hoc correction pipeline that augments object-level visual grounding with dual relational evidence: learned scene-graph triples from RelTR and deterministic spatial predicates from bounding-box geometry. These combine with a Woodpecker-style object claim layer to form a three-layer visual knowledge base, which a language model corrector uses to rewrite hallucinated text. Evaluated on LLaVA v1 13B, RelCheck achieves a total MME hallucination score of 630.0 versus 585.0 for a Woodpecker-style baseline, with the largest gain on the position subtask (+31.7 points, accuracy+ improving from 0.367 to 0.600). A four-configuration ablation confirms that both relational layers contribute independently (McNemar p = 0.025). These results show that structured relational evidence meaningfully improves post-hoc hallucination correction on the spatial reasoning subtasks where current MLLMs are most deficient.

A common characteristic of all popular multi-path routing algorithms in mobile ad-hoc networks, such as AOMDV, is that the end to end delay is reduced by utilization of parallel paths. The competition between the neighboring nodes for obtaining a common channel in those parallel paths is the reason for end to end delay increment. In fact, due to medium access mechanism in wireless networks, such as CSMA/CA, data transmissions even through two node-disjoint paths are not completely independent and each path will affect the other one. In this paper we have modified the AODV protocol which results in selection of zone-disjoint paths, to the extent feasible, and as a result we achieve less end to end delay. The efficiency of the proposed protocol has been evaluated on different scenarios and there has been a noticeable improvement in the packet delivery ratio and also in the reduction of end-to-end delay comparing to AOMDV.

In IEEE 802.15.4 standard for low-power low-range wireless communications, only one channel is employed for transmission which can result in increased energy consumption, high network delay and poor packet delivery ratio (PDR). In the subsequent IEEE 802.15.4-2015 standard, a Time-slotted Channel Hopping (TSCH) mechanism has been developed which allows for a periodic yet fixed frequency hopping pattern over 16 different channels. Unfortunately, however, most of these channels are susceptible to high-power coexisting Wi-Fi signal interference and to possibly some other ISM-band transmissions. This interference manifests itself in the form of the presence/absence of other devices with either or both static and dynamic channel selection policies. In order to isolate channels with undesirable conditions, blacklisting mechanisms are defined to adapt the channel hopping process. However, the existing solutions which form blacklists unrealistically assume that the statistical model of the external interference remains fixed, and do not vary over time. In this paper, we realistically assume that the impact of external interferes on 802.15.4 may generally follow a non-stationary pattern, and accordingly formulate the adaptive channel hopping problem as a Dynamic Multi-Armed Bernoulli Bandit (Dynamic MABB) process from the machine learning theory. We then propose an online learning algorithm with track-ability properties for computing an adaptive hopping policy. Simulations confirm that when the statistics of the external interference has a switching regime, the proposed solution outperforms the previous schemes in terms of both energy efficiency as well as two important KPIs for TSCH-based networks, i.e., PDR and latency.

Sharing a deep encoder does not, by itself, fix the central confound of task-comparison scores. We show that cross-evaluated heads on a frozen shared representation inherit the extrapolation confound of shallow exchange scores: pure input rotations with fixed labels inflate a deep exchange score from about 0 to 0.80, while representation-novelty scores are blind in the complementary direction (flat under label permutations that change the task completely). Transplanting a conditional two-discriminator discrepancy into the embedding space resolves both blind spots: the functional axis stays within +-0.001 under rotations and tracks label-permutation drift mass monotonically. Built into a mixture-of-heads lifecycle, the two-axis gate attains better decision quality with fewer heads than exchange or novelty triggers at a matched training budget. On generalized category discovery, the same chunk-level functional axis separates semantic novelty from photometric shift with AUROC 0.98-0.99 where per-input OOD scores (MSP, Energy, Mahalanobis, KNN) sit near chance for that distinction. All findings replicate across frozen ImageNet-21k ViT-B/16 and self-supervised DINOv2 backbones on CIFAR-100, and extend to residual adapter pools with recurrence, where a null-calibrated novelty trigger never fires on mechanism changes while the two-axis gate handles them with full recurrence reuse. We state explicitly the common-factoring condition under which embedding-space conclusions transfer to the original mechanism.

High-resolution surface solar radiation (SSR) is important for solar forecasting and grid operation. However, physically consistent reanalysis products are too coarse to resolve localized cloud-driven variability. In this paper, we study a multisource downscaling task that reconstructs high-resolution SolarCube SSR fields from coarse ERA5 radiative variables and co-registered satellite channels. The task is challenging because a single ERA5 grid cell may contain both sunlit and cloud-shadowed regions. As a result, the missing high-resolution correction can be spatially sharp and inherently ambiguous. One-stage predictors often oversmooth these structures. Post-hoc refinement also introduces a stage-wise mismatch: the generator is optimized independently, even though its output determines the refiner's initial state. We introduce SolarFlowRefiner, a refinement-aware flow-matching framework for SSR downscaling. A conditional FlowMatch generator first predicts a normalized correction to an upsampled ERA5 baseline. The refiner is then trained on prediction-conditioned states between the current FlowMatch output and the target residual. This exposes the refiner to the structured errors produced by the generator. The refinement objective is also backpropagated through the FlowMatch sampler, allowing generation and correction to be jointly optimized for the final reconstruction. Experiments on a day-blocked ERA5–SolarCube benchmark show consistent improvements over standalone generation and post-hoc refinement. More broadly, SolarFlowRefiner provides a general strategy for coupling generative predictors with iterative correctors.

Standard tokenizers used in large language models produce malformed text when applied to Brahmic scripts. They are a family of abugidas, writing systems whose consonants carry an inherent vowel that dependent marks can modify. They include Devanagari, Telugu, Tamil, Kannada, and others. The underlying issue is that these tokenizers violate orthographic constraints that do not arise in alphabetic scripts like English. We observe that while English orthography forms a semigroup (any two valid tokens can be freely concatenated), Brahmic orthography forms a partial semigroup: not every concatenation yields a valid string. We formalise this distinction in Agda, model valid Brahmic tokens as chains in a transition system, and derive a provably correct fixToken function that extends any candidate token to respect orthographic boundaries. We then show how this formal derivation translates into a practical patch for SentencePiece as well as a standalone Rust-based pre-tokenizer library, eliminating the observed errors across Indic scripts.

Despite significant progress in general visual question answering and cross-modal understanding, multimodal large language models still face a pronounced gap in evaluation for complex reasoning within the mechanical engineering domain. Existing benchmarks predominantly focus on rudimentary tasks such as drawing recognition, CAD interpretation, or single-chart querying, falling short of assessing whether models can integrate multiple images, textual conditions, physical principles, and engineering constraints to perform multi-step reasoning when confronted with authentic, intricate mechanical problems. To address this, we introduce MechReason, a benchmark derived from real mechanical engineering papers, comprising 12k question-answer pairs with explicit reasoning-chain annotations and 21k visual materials spanning nine evidence types, including statistical charts, parameter tables, engineering drawings, microscopic images, simulation images, system architectures, real mechanical scene photos, CAD model images and manufacturing flowcharts. MechReason covers eight task types across four reasoning dimensions: explanation, prediction, design, and diagnosis. We devise a four-stage construction pipeline: we first extract core engineering claims and decompose their supporting evidence into premises, reasoning processes, conclusions, and corroborative evidence; we then generate shortcut-preventing questions by masking posterior verification information; finally, we apply multimodal quality validation to ensure task quality and multi-hop nature. Extensive experimental results demonstrate that MechReason is highly challenging, with even the most advanced models achieving only 62.89% accuracy.

Large language model (LLM)-based autoregressive speech enhancement (SE) produces natural speech using learned clean-speech priors, but may hallucinate content unsupported by the input. Deterministic SE better preserves observation-coupled evidence, yet often retains residual noise or local distortion. We propose an evidence-grounded generative SE framework that uses a deterministic estimate as imperfect evidence. A Whisper-guided DPRNN produces an enhanced waveform, which is blended with the observation and tokenized into a discrete evidence sequence. The evidence conditions an autoregressive clean-speech token generator and is reused during decoding through Code-Space Grounding (CSG), which penalizes candidates according to their Hamming distance in the factorized finite-scalar-quantized (FSQ) space. Because the appropriate grounding strength depends on acoustic difficulty, we introduce SNR-Conditioned CSG (SNR-CSG), which maps a calibrated residual-SNR estimate to an utterance-level strength and constructs an adaptive grounded anchor. Although grounding improves content fidelity, the anchor may retain local acoustic defects inherited from the evidence. Since such defects are predominantly local in the FSQ space, nearby tokens may provide better acoustic realizations without large departures from the observation-supported trajectory. We therefore propose Grounded Neighborhood Refinement with LLM ranking (GNR-LLM). It performs one additional teacher-forced pass conditioned on the grounded-anchor history, intersects the LLM top-$K$ candidates with a local FSQ Hamming neighborhood. Experiments on in-domain, controlled-SNR, and DNS no-reverb conditions show that SNR-CSG provides robust automatic grounding, while GNR-LLM substantially improves low-SNR perceptual quality without sacrificing content fidelity.

We present a projected sensitivity study for hadronic mono-$Z$ dark-matter production using CMS Run~2015D HTMHT open data corresponding to 2.256382381~\invfb, from which 1{,}439{,}523 events satisfy the hadronic mono-$Z$ selection. Backgrounds are modelled with a conditional flow-matching continuous normalizing flow trained on the selected HTMHT events and evaluated on a held-out validation split reweighted to the full selected population. To mitigate artifacts from missing-object features and avoid in-sample scoring bias we apply sentinel imputation for undefined angular features, persist the train/validation split indices, and enforce a minimum reported background yield of 20 events when selecting the working point. A signal-side offline trigger proxy is applied to the simulated signal before scoring. Under this procedure the baseline analysis yields expected significances of 2.89$\sigma$, 7.62$\sigma$, and 7.41$\sigma$ for three simplified-model benchmarks. An ablation study that removes the detailed extra-jet kinematics reduces the expected significance by 53–71%, indicating that extra-jet topology carries substantial discriminating power in the hadronic mono-$Z$ channel. These results are projected sensitivities (no unblinding performed); the limitations and reproducibility of the study are discussed in Sections limitations and reproducibility.

A composite structural index summarises a network in one number, and for a triangle-based index it is spectrally redundant: Tr(A^3) is the third moment of the adjacency spectrum. The non-redundant content sits one level down, in diag(A^3), which depends on eigenvectors and is not spectrally determined. A corollary in the theory paper predicted that the global scalar should tie sharpened spectral baselines rather than beat them, while the node-wise attribution should do better where the number of structural epicentres is unknown. We construct Omega-N by localizing each of the four factors. The direct localization is badly conditioned; two corrections from published practice fix it, a configuration-null excess per factor and a personalized-PageRank neighbourhood at several scales, giving ten interpretable features per node, with no attributes, training or embeddings. Against a recursive feature engine at five levels of recursion, Omega-N wins on three and ties on two of the six in-domain evaluations, the sixth a declared null where every arm returns chance, with ten features against its 28 to 252 before pruning. Two statistics from the graph and labels, not from performance, partition the eight benchmarks without error, and the two they exclude are the two on which it loses. The strongest application is drug-target prioritisation on protein interaction networks: +0.032 to +0.103 AUPRC over a six-feature centrality battery and +0.084 to +0.208 over the four-feature one, across three constructions, replicated on an independent AP-MS network and label source (degree-matched: +0.0723 on STRING, +0.0560 on BioPlex, p=0.00195). Adding Omega-N to centralities plus Node2Vec changes nothing. The claim is narrow and it is the point: ten named features, computed without training, match or beat hand-crafted centralities and a recursive engine, and do not touch learned representations.

Distant speaker diarization remains challenging due to adverse acoustic conditions, varying numbers of speakers and overlapping speech. While data-driven approaches have shown strong performance, model-driven methods offer a compelling alternative by leveraging spatial information from multichannel recordings. This paper is motivated to propose a Bayesian diarization model for a model-driven method called neural FCASA to enhance its robustness. Specifically, we propose a beta prior over speaker activity and hence a variational lower bound objective that can be seen as a regularized continuous speaker activity score in place of the original cross-entropy loss to train the diarization model. Our experiments show significant improvements in terms of Diarization Error Rate by at least 3% (16% relatively) and Jaccard Error Rate by at least 4% (20% relatively) on the AMI dataset compared to the baseline.

Test-time scaling (TTS) improves language model reasoning by allocating additional test-time compute, but prior work mainly studies closed-ended tasks such as math and coding. We study TTS for automated equation discovery, an open-ended setting where models search over candidate equations and rely on observed datapoints for feedback. We formulate LLM-driven equation discovery as an iterative search process that unifies Best-of-N, sequential refinement, tree search, and evolution-style methods under a common compute-allocation view. To isolate allocation effects from prompt engineering and other heuristics, we compare minimal parallel controllers under fixed budgets. On LLM-SRBench equation-discovery tasks, we find that search width is the dominant allocation parameter: the best width in our sweep generally increases with the compute budget, while the population–branching split and controller choice matter less. Appropriate width selection also improves wall-clock efficiency by increasing parallelism. These results suggest that, given an informative verifier, controlling exploration and exploitation is central to scaling LLM-based equation discovery.

Phylogenetic networks model evolutionary histories that involve reticulate events, but their structural complexity makes them difficult to interpret. Extracting their simple substructures both clarifies the evolutionary pathways and quantifies the complexity of the networks themselves. For a given rooted almost-binary phylogenetic network, the Level Minimization problem asks for a spanning subgraph that has the same root and leaf-set and whose level is minimum, i.e., which is as close to a tree as possible. Networks for which the minimum level is zero are known as tree-based networks and can be recognized in linear time. However, Level Minimization is NP-hard in general. State-of-the-art algorithms rely on exhaustive searches of the solution spaces and hence apply only to networks of limited size. In this paper, we propose two methods for Level Minimization using integer linear programming: an exact formulation for finding such a subgraph of level at most one, and a heuristic formulation for the general case. Computational experiments confirmed the practicality of both formulations. An application to ancestral recombination graphs suggests that the minimum level provides an alternative measure of the topological complexity of an inferred network.

Anonymous Multi-Agent Path Finding (AMAPF) admits polynomial-time network-flow algorithms for several objectives, including makespan, total distance, and sum-of-costs (SoC) when agents disappear upon reaching goals. We show that standard goal-staying AMAPF is fundamentally different. We first formulate SoC minimization by augmenting the standard time-expanded flow model with goal-settlement constraints and show that the resulting linear programming relaxation is non-integral. We then prove that minimizing SoC in goal-staying AMAPF is NP-hard via a reduction from 3-SAT. Together with the polynomial-time result for the disappearing variant, this establishes a sharp complexity boundary determined by whether completed agents remain at their goals.

Behavior-cloned Vision-Language-Action (VLA) driving policies struggle with rare rule-governed maneuvers at signalized intersections. Braking and launching examples contribute little to averaged trajectory loss, while fused representations lack explicit supervision for the governing traffic-light and stop-line state. We present RedLight-VLA, a training objective that uses expert futures and automatically generated perception targets without additional manual rule annotation. First, trajectory-derived behavioral reweighting (BR) emphasizes rare deceleration and acceleration using rotation-invariant longitudinal dynamics and a scale-preserving reduction that exactly recovers the baseline when disabled. Second, parallel auxiliary (AUX) heads ground traffic-light and stop-line state in continuous post-fusion rule tokens, without autoregressive language generation or changes to the trajectory decoder. We evaluate on a curated set of 20 s sequences with a 5 s prediction horizon. Controlled variants share the same backbone, training data, decoder, and evaluation population. Against an otherwise identical VLA baseline, RedLight-VLA reduces red-light stop-line overshoot from 7.3% to6.8%, reduces stop-line velocity error by 12.7%, and improves 3 s trafficlight-sliced ADE/FDE from 0.274/0.964 m to 0.247/0.897 m. Green-light false stops increase from 3.2% to 3.9%; however, combining BR with AUX supervision mitigates the larger increase observed for AUX alone (4.0%). The combined model also improves non-traffic-light ADE/FDE from 0.268/0.956 m to 0.241/0.876 m and outperforms either mechanism alone on all four sliced displacement measures.

Graphical user interfaces have emerged as an important environment for evaluating autonomous AI agents on multimodal interactive tasks. Existing benchmarks such as AndroidWorld and MobileWorld provide strong foundations for mobile agent evaluation, but their application coverage and task design do not yet fully capture the diversity and complexity of realistic mobile use. We present GMA, a benchmark for evaluating general mobile assistants in challenging real-world scenarios. GMA introduces seven applications based on open-source projects, spanning domains such as lifestyle sharing and travel planning, and 300 tasks across four difficulty tiers, from atomic actions to complex multi-step workflows. We evaluate eight frontier models and find that performance declines substantially as task complexity increases, with current agents remaining far from reliably handling realistic user requirements. We further conduct controlled ablation studies of agent harness choices, including context retention and explicit state tracking, under a shared environment, model setting, and task taxonomy. Results show that appropriate harness design can meaningfully improve performance, particularly on demanding workflows, while the effectiveness of specific designs can vary across foundation models. Overall, GMA complements existing benchmarks by expanding application coverage and task complexity, providing a challenging testbed for evaluating mobile agents and studying how harness design supports reliable execution in complex mobile workflows.

Large language models (LLMs) are increasingly used for verbal interaction in social robots, yet prompt design in human-robot interaction (HRI) remains underspecified. As a result, robots may present hallucinated capabilities, unclear behavioural boundaries, and misleading personas. This paper develops a framework for prompt design in LLM-based robots and introduces a structured prompt template comprising eight functional components through which robot behaviour can be specified, bounded, and adapted. The framework is grounded in a review of prior LLM-based HRI work and complemented by survey and discussion data from HRI experts gathered at the Robo-Identity workshop at IEEE RO-MAN 2025 (N=27). The qualitative findings highlight limited legibility of robot personality, the need for user adaptation, and strong ethical concerns about safety, deception, and governance. Based on these findings, we present prompting guidelines accompanied by proof-of-concept template as a structured design and reporting aid for HRI research. We argue that prompt design should be treated as a socio-technical problem rather than a minor implementation detail, requiring explicit capability boundaries, transparent behavioural assumptions, and context-sensitive safeguards to support reliable and interpretable HRI.

This paper describes the development of software drivers that enable precise timing applications with the low-cost HackRF One software-defined radio (SDR). The original firmware was extended with timing functions for synchronized sampling, time-scale control, and clock-frequency adjustment. An initial implementation based on a Python wrapper was later redesigned as a SoapySDR-compatible driver. Using an AI-assisted development workflow, the new driver was generated from the existing HackRF One and SoapySDR codebases and became operational after only limited debugging. The approach reduced development effort by more than an order of magnitude while improving software maintainability.

Large Language Models (LLMs) offer new opportunities for automated code refactoring. However, generated changes must reduce targeted quality problems without introducing new issues or altering behaviour-relevant code structures. We introduce REFINE (Refactoring with Evidence-aware Flow for Integrated ageNtic Execution), a tool-agnostic, evidence-aware multi-agent approach for generating Java file-level refactoring candidates. REFINE combines static-analysis-guided smell identification, smell-informed planning, LLM-based transformation, automated re-analysis, preservation checks, and structured reporting. We evaluate REFINE on 450 Java files from 15 open-source systems, producing 1,350 model-pass outputs using OpenAI GPT-5.5, Google Gemini 3.1 Pro Preview, and Anthropic Claude Opus 4.8. REFINE reduces detected code smells by 68.26%, 72.79%, and 68.49% across the three configurations, respectively, with the strongest reductions observed for major smells. A matched 150-file direct-prompt baseline shows that REFINE achieves a higher median code-smell reduction with smaller edits and fewer public-method removals. However, broader quality improvements are inconsistent, and preservation checks reveal residual risks, including assert/fail-call changes and public-method removal. Therefore, REFINE outputs should be treated as refactoring candidates requiring compilation, testing, dependency analysis, and human review before adoption in repository- or system-level settings.

A pipeline promoting an AI system publishes records claiming the thing evaluated is the thing deployed and that the evidence licensed the transition. We measure, from public material only, whether those records can express that claim and whether it holds where declared. First, a two-class documentation survey of 47 delivery platforms (20 CI/CD, 27 model-serving/agent) under one fixed three-label protocol, graded twice (second pass blind), every consulted page pinned by content hash and date. Across 188 double-graded cells we found no platform whose default record emits a content-addressed identity of the behavioral tuple (model version, instructions, tool definitions, retrieval and runtime configuration); the blind pass grades that column default on zero of 47. Immutable nominal versioning is meanwhile arriving as the agent platforms' default answer (16 of 27): version integers behind mutable pointers, a layer the artifact supply chain already found insufficient. Second, an instrument computes realized assurance depth from a pipeline's published exhaust alone and compares it with the declared depth. Applied to a frozen two-stratum frame of 30 public repositories graded twice from a hashed archive (second pass blind; cell-level agreement 23 and 19 of 30, both passes independently finding the same five full realizations), the sharpest result is a verifiability hole: seven of the 15 repositories chosen for adopting attestation tooling publish source-only releases, so the binding their workflows declare cannot be checked where declared. Where checkable it mostly checks out: five of seven measurable adopters realize the binding end to end; both shortfalls fall at identity binding. Together the results locate the field's records structurally short of justifiability, the one rung that can refuse a transition. The survey carries an expiry clock; we state what would falsify each finding.

Local optimization does not necessarily yield end-to-end optimization in layered browser architectures. Teams often treat Web Workers, WebGL, and WebAssembly as interchangeable ways to "make it faster," yet each lever targets a different layer. We present a controlled architectural decomposition–using DOM-sourced particle pipelines as a concrete workload–of five particle pipelines (P1-P5), with an additional non-particle CSS-layer baseline (P0), that isolates thread placement, renderer choice, and simulation backend. Using a reproducible harness we measure interactive pacing, high-load end-to-end stress, and a simulation-only microbenchmark, plus same-host cross-browser / dual-GPU-class replication (Firefox 153+Intel UHD; Chrome 138+NVIDIA NVK) and an independent second-host slice on Google Colab (Chrome 150, Tesla T4, n=5). Four results stand out. (1) Worker offload improves interactive pacing on paper-primary Chrome (approx. 144 vs approx. 52 FPS; Cliff's delta=1, n=10). (2) AssemblyScript speeds the Chrome update kernel by about 1.5-1.6x on the primary host and approx. 1.85x on Colab T4 (sim-only). (3) Under approx. 250k WebGL particles the Chrome sim-only WASM win need not raise product FPS (P5<=P4 on primary/Firefox; P5 approx. P4 on Colab T4); in-worker CPU phase timers show simulate still dominates the accounted budget, so the non-translation is not a simple "draw dominates CPU" story. (4) Renderer ranking and absolute margins vary across browser/GPU/host configurations. The contribution is bottleneck-aware architectural measurement: identify the dominant layer and test whether a layer win propagates to user-visible FPS.