Papers

Pointer analysis is a fundamental technique for compiler optimization and program analysis. Flow-sensitive pointer analysis provides high precision but is difficult to scale to large projects. Tailored for rapid iteration scenarios where software evolves continuously, we introduce IncSFS, the first incremental full-sparse flow-sensitive pointer analysis algorithm for C/C++ programs. IncSFS first transforms the value-flow graph into a constraint graph and performs strongly connected component detection to ensure precision. It then propagates increases and decreases in points-to sets in an interleaved manner, supporting code deletion and insertion within a single analysis pass. IncSFS is guaranteed to terminate and compute the least fixed point when the points-to relation remains object-acyclic during analysis. Experiments on six large-scale real-world projects show that IncSFS is precise and efficient, achieving average speedups of 9.60x over full flow-sensitive pointer analysis and 5.84x over the traditional reset-recompute approach. It also improves efficiency by 15.8% over state-of-the-art incremental pointer analysis algorithms that propagate points-to-set changes.

The intricate mixed-integer constraints in industrial load models not only pose challenges for their direct integration into economic dispatch or market clearing processes but also render current analytical dimension-reduction methods ineffective. We propose a novel data-driven dimension-reduction approach for industrial load modeling, which uses the optimal energy usage data from industrial loads to train a dimension-reduced model that best fits the original constraints. Our approach, implemented by the adjustable load fleet model, outperformed analytical methods across three industrial load datasets.

Tensor-valued prediction is fundamental to geometric deep learning, yet uncertainty quantification (UQ) for such outputs remains an open challenge. While E(3)-equivariant neural networks excel at point estimates, they lack rigorous confidence measures. We focus on symmetric rank-2 tensor prediction, where the target has six Kelvin–Mandel coordinates and full uncertainty is represented by a $6\times6$ covariance matrix. We introduce a framework for E(3)-equivariant UQ, modeling the full predictive distribution where both mean and covariance preserve rotational symmetry. Our approach decomposes the covariance into irreducible representations $\mathrm{Sym}^2(\rho_c) \cong 2\times(l=0) \oplus 2\times(l=2) \oplus 1\times(l=4)$. By mapping from the flat Lie algebra $\mathfrak{sym}(6)$ to the curved SPD manifold via matrix exponentiation, we strictly ensure positive-definite covariances while maintaining exact equivariance. Furthermore, we formulate a Log-Euclidean Equivariant Scoring Objective (LE-ESO)—a robust surrogate loss based on the Multivariate Laplace distribution—providing robustness to heavy-tailed errors and stable optimization. Validation on ModelNet40 inertia tensors and Materials Project dielectric tensors demonstrates that our method achieves competitive performance and provides physically consistent, symmetry-preserving uncertainty estimates with useful risk and OOD sensitivity.

Resistance training can be a high risk activity, and safe form is essential to avoiding injury. Laboratory-based movement analysis provides quantitive technique assessment, yet is not easily accessible. Markerless pose estimation infers body landmarks from images or video without physical markers and could offer a feasible alternative for technique assessment. We present a pose estimation framework to evaluate resistance-training technique from ordinary video footage. Using BlazePose, anatomical landmarks were extracted from squat, bench press, and deadlift videos and converted into joint-angle trajectories, with the squat serving as the primary case study. Trajectories were assessed against a defined reference repetition using root mean square error (RMSE). Results show that the framework recovers meaningful kinematic patterns for the squat and deadlift, enabling quantitative comparison between repetitions and identification of technique variability within a set. Performance depended strongly on camera orientation and visual occlusion, with non-sagittal views distorting 2D joint-angle estimates. The findings demonstrate that markerless pose estimation can support accessible biomechanical assessment outside laboratory environments.

In this work, we conduct an empirical study on self-supervised learning (SSL) for tabular data. Our experiments examine limited label regimes (that is, 1%, 5%, 10%, and 20% of the labels) and highlight limitations and opportunities in current tabular SSL. Contrary to what is expected, our mask-and-recover approach shows its largest improvements for datasets that have little to no recorded numeric missingness, whereas it more often leads to degradation on datasets that are characterized by native numeric missingness (post hoc split). Random Forest achieves an average AUC of 0.9015 across the 14 tasks for the 10% labels setting, whereas SSL shows a performance of ~0.8954 (RF: 1 draw; SSL: 3-draw mean). Moreover, the gain of SSL compared to training from scratch exhibits a high variance across tasks and is not significant (1 run; Wilcoxon p = 0.626, Holm-adjusted p = 1.000, for both the 5% and 10% settings). On the other hand, we observe that SSL-pretrained models achieve a higher average AUC when MNAR and MCAR missingness were injected at test-time. In comparison to models trained from scratch, our SSL-pretrained models show an improvement of +0.0245 (MCAR) and +0.0418 AUC (MNAR) under test-time missingness (single seed), with 11 (MCAR; plus one tie) and 8 (MNAR) positive results across the 14 tasks. However, after Holm-Bonferroni correction neither improvement is significant (Holm-adjusted p = 0.118 for MCAR and 0.518 for MNAR). Finally, when comparing mask-and-recover with three other approaches, namely, simplified VIME, SCARF and SubTab implementations using an identical encoder, we observe no significant difference in performance (Holm-adjusted p = 0.459 for VIME, 1.000 for SCARF, 1.000 for SubTab).

We study the shortest-path problem on graphs with positive real-valued edge weights. Given a source vertex $s$ and a target vertex $t$, the goal is to calculate the length of the shortest path from $s$ to $t$. We are particularly interested in instances that can be solved in sublinear time. Recently, Haeupler, Hladík, Rozhoň, Tarjan, and Tětek proved that (a version of) the bidirectional Dijkstra's algorithm is instance-optimal on positively weighted multigraphs, both directed and undirected, considering the number of vertices and edges queried by the algorithm. However, multigraphs are not the canonical setting for the shortest-path problem. The problem is typically formulated on simple graphs without loops and parallel edges. They therefore left as an open problem whether bidirectional Dijkstra remains instance-optimal on simple weighted graphs. We answer this question, but for simple graphs, the answer is more complex, depending on the setting. We show that bidirectional Dijkstra is still instance-optimal on simple undirected weighted graphs under the order-oblivious model, where incident edges are given in a random order. In contrast, under the order-dependent model, where incident edges have a given order, we show that bidirectional Dijkstra is not instance-optimal. For simple directed weighted graphs, we show that bidirectional Dijkstra is not instance-optimal under either the order-oblivious or the order-dependent model, being off by a factor of $\Theta(m/n)$ in both cases. We further show that no algorithm can have instance-optimality ratio $o(m/n)$ under the order-dependent model, or under the order-oblivious model when $m=O(n\sqrt{n})$. On the positive side, the above results imply that bidirectional Dijkstra is instance-optimal up to logarithmic factors on all sparse directed and undirected graphs satisfying $m/n=\log^{O(1)} n$.

Conventional audio language models (ALMs) have made significant progress in achieving alignment between auditory and textual representations, including recent explorations in spatial audio. However, in daily spatial scenarios, they still cannot effectively process multi-event audio sequences. Current approaches primarily rely on coarse-grained contrastive learning with global auditory and textual features, lacking the resolution to distinguish multiple sequential events. To overcome these limitations, we propose CoSTALA-a novel training paradigm that transitions from purely global alignment to fine-grained spatio-temporal reasoning. By constructing a multi-granularity hierarchical loss function system, we achieve explicit modeling of temporal dependencies, and successfully anchors individual acoustic events to preserve their semantic purity. Extensive experiments demonstrate that CoSTALA significantly establish a powerful new framework for spatio-temporal audio understanding.

AI-generated image quality assessment (AIGIQA) requires jointly reasoning about perceptual fidelity and prompt alignment, two quality dimensions that are often treated as independent in existing AIGIQA models. However, by re-examining human ratings, we uncover a previously overlooked phenomenon: the two dimensions are interdependent and exhibit both competitive and cooperative interactions during human rating. This observation suggests that a unified model should neither collapse the two dimensions nor rigidly separate them, but rather adaptively negotiate their interplay. Motivated by this insight, we introduce an interaction-aware learning framework that models perception-alignment relations through adversarial and collaborative inference pathways. Instead of designing a rigid dual-branch architecture, our method employs a gated interaction module that dynamically routes features according to the inferred relationship between the two dimensions. Task-aware prompts further modulate the gating behaviour, enabling the model to switch between competition and cooperation when necessary. Experiments across multiple AIGIQA benchmarks demonstrate that our approach not only achieves state-of-the-art accuracy but also yields interpretable interaction patterns, offering a more faithful approximation of human judgment. The codes are available at https://github.com/LQAMEI/ACL-IQA.

Recent years have seen a rapid growth in literature presenting new methods for simulating non-Clifford quantum circuits with classical hardware. These methods span a range of approaches, including stabiliser decomposition and tensor contraction techniques, varying in efficiency depending on circuit class, depth, non-Clifford gate count, and other metrics. A notable limitation of this literature is the lack of a standardised approach to benchmarking, with each new paper outlining its own specification, simulating its own set of circuits on the authors' own hardware. This paper seeks to address this issue by presenting a standardised and canonical benchmark suite and infrastructure for quantifying the efficiency of non-Clifford classical simulators, with a consistent dataset of circuits and providing consistent (virtual) hardware, thereby enabling a fair comparison of results.

While large language models (LLMs) possess vast zero-shot procedural knowledge, their tendency to produce homogenized logic often obscures the unique, idiosyncratic execution processes of individual human creators. In this paper, we investigate the computational discovery of procedural personas from unstructured data. To achieve this, we introduce ViralRecipesTrans, a new dataset of procedurally aligned execution flow graphs extracted from popular culinary video transcripts and explicitly mapped to specific creators. We formulate procedural stylometry as a graph learning and process discovery task, revealing a fundamental duality: while traditional lexical classifiers overfit via semantic leakage, discrete topological metrics successfully capture the rigid physical constraints of a creator's workflow. Building upon this characterization, we extend our framework into a novel generative task–predicting a creator's exact structural execution graph for unseen dishes. We expose a fundamental dichotomy in style generation between global macro-planning and local structural execution. Our results demonstrate that few-shot LLMs dominate semantic assignment but suffer from persistent macro-planning deficits, whereas our structured two-stage model achieves superior topological control via rigid Markovian priors. Together, an ensemble approach to procedural generation combines the strengths from both sides, dynamically synthesizing global semantic reasoning with localized topological footprints to automate the discovery and generation of personalized workflows.

Reliable multi-turn tool use requires an agent to preserve an evolving task state and ensure that each action remains consistent with it. However, direct function-calling and ReAct-style policies learn state tracking and action generation within the same autoregressive trajectory. This coupling creates state-action competition: the pressure to produce the next call can overwrite or ignore information accumulated earlier in the interaction. Inspired by Boyd's Observe-Orient-Decide-Act cycle, we introduce OODA-Tool, a typed closed-loop policy designed to mitigate this competition by separating state preservation from action realization. Rather than generating an action directly from the interaction history, OODA-Tool routes each decision through controller-checked intermediate states, ensuring that the final output remains grounded in the current task state. Specifically, Observe reconstructs the task state, Orient determines whether execution is warranted, Decide forms an admissible action structure, and Act realizes the external output. We evaluate OODA-Tool against direct function-calling and ReAct policies using Qwen3 models ranging from 0.6B to 14B across multi-turn, multi-tool, and incomplete-information settings. OODA-Tool consistently improves task success across model sizes, with larger gains on smaller models and on tasks whose actions depend strongly on information accumulated across turns and prior tool results. Controlled variants, stage-level ablations, and transfer evaluations further demonstrate the robustness of these improvements.

This chapter examines AI-media objects - creative and artistic outputs generated through generative artificial intelligence in the form of text-to-X tools - in relation to three avant-garde movements of the twentieth century: Dadaism, Surrealism, and Conceptual Art. Drawing on Lewis Carroll's Through the Looking-Glass as an early precursor to these three movements and to anti-rationalist aesthetics, and on three case studies in AI-generated conceptual architecture - Matias del Campo's "Deep House" and Hassan Ragab's "Post-Pharaonic Architecture" and "A State of Decay" - the chapter develops the concept of latent-surrealism. Latent-surrealism includes a set of aesthetic and methodological conditions inherent to creative AI-media objects. These include the use of readymade datasets reassembled through collage-like processes, the absurd as an aesthetic quality of machine hallucinations, and the decoupling of craft from artistic value. The chapter further argues that AI-media objects represent a shift in the conditions of creative production: where earlier computational tools required graphical interfaces and programming literacy, natural language now functions as the operative medium. This repositions the prompt (as language-based instructions) in the center of the creative process, in continuity with the legacy of Conceptual Art, and signals a linguistic turn across creative fields that make (use of) AI-media objects.

End-to-end multimodal driving has progressed rapidly by fusing camera and LiDAR streams. Existing pipelines remain fragile under asymmetric sensor degradation, where either an entire modality or only a localized region is corrupted while other regions remain useful. The key difficulty is not simply to add an uncertainty head, but to obtain dense reliability supervision, calibrate this reliability against physical fault severity, and use it before unreliable features bias the planner. We propose Variance-Guided Spatial Attention Fusion (VG-SAF), in which dense heteroscedastic reliability estimates act as interpretable spatial gates. The framework couples three components. First, a physically grounded augmentor simulates representative camera and LiDAR failures and emits a continuous spatial mask, providing dense supervision without additional annotation. Second, modality-specific experts predict per-pixel reliability scales through cross-branch dense distillation in log space, enforcing a monotone severity-to-scale response. Third, calibrated reliability maps drive a hybrid attention mechanism that suppresses unreliable cells with a local spatial gate and arbitrates between modalities through a cross-modal trust softmax. A Laplace uncertainty head emits a systemic waypoint uncertainty scale that signals severe or combined sensor degradation, including severities outside the training ranges. On the CARLA Longest6 benchmark, VG-SAF consistently improves closed-loop robustness over the baselines across camera-only, LiDAR-only, and joint degradation regimes, as measured by driving score, route completion, and infraction score.

Transformer-based object trackers are renowned for their strong performance, yet dense token processing often leads to prohibitive computational cost, limiting real-time deployment on edge devices. While recent works explore token pruning to reduce computation, they often stop short of an end-to-end sparse pipeline, as early-layer token scores can be noisy without a motion prior, and many trackers ultimately fall back to dense reshaping to feed the dense prediction head that partially negates the savings. We introduce Motion-aware Sparse Tracker (MaST), a sparse tracking framework that makes sparsity effective from tokens to boxes. First, MaST injects a lightweight motion prior to refine cross-attention-based importance scores, enabling earlier and more stable token reduction in the search region. Second, we introduce a natively sparse prediction head that operates directly on the retained unstructured tokens with a score-first, regress-once design, eliminating dense padding/reshaping and reducing redundant computation. Extensive experiments on multiple benchmarks demonstrate that MaST establishes new state of the art among lightweight trackers, where MaST-tiny attains 63.8 AUC on LaSOT and 80.1 SUC on TrackingNet, surpassing the prior best AsymTrack-S by +1.0 AUC and +2.2 SUC while running at 152 FPS on Jetson Nano, nearly twice as fast as AsymTrack-S at 88 FPS. Code is available at https://github.com/TsingWei/MaST.

Self-supervised pretraining enables transferable representations for medical imaging, yet most CT encoders remain biased toward coarse semantic understanding, limiting their sensitivity to fine-grained anatomical structures such as vessels or small tumors. In this paper, we introduce Biased Masked Image Modeling (B-MIM), a modification of the iBOT objective that stochastically reduces global semantic alignment to prioritize local patch reconstruction. This bias encourages the encoder to capture high-frequency morphological details and structural continuity. We curate a multi-institutional CT abdominal dataset of 9,955 filtered studies from 17 public sources and pretrain a 3D Swin Transformer backbone using B-MIM. Across inter-dataset experiments on liver vessel segmentation, the proposed encoder improves topological fidelity (clDice) and achieves competitive Dice scores in tumor segmentation, compared to fully fine-tuned baselines, despite updating only a fraction of the parameters. Our results suggest that reducing global semantic pressure during pretraining enhances generalization to intricate anatomical structures.

Recent advances in LLM-enhanced recommendation commonly align semantic representations with collaborative embeddings in a shared space, yet how alignment affects LLM-encoded information remains unclear. In this work, we revisit LLM-enhanced recommendation from a spectral perspective and show that collaborative and semantic signals benefit from different spectral parts. While collaborative representations are dominated by smooth low-frequency components due to user-item homophily, semantic embeddings contain useful non-principal singular components. Through component-wise evaluation and training-dynamics analysis, we find that alignment increasingly concentrates learned representations in dominant collaborative and principal semantic subspaces, reducing overlap with non-principal semantic components. Controlled comparisons show that non-principal components provide inconsistent gains under alignment but consistently improve performance through component-level decoupling, while full prediction-level decoupling achieves the best overall performance. These results indicate that alignment fails to effectively exploit complementary non-principal semantic information. Motivated by these findings, we propose UniSpecRec (Unifying Spectral Signals for Recommendation), which applies signal-specific spectral filtering while preserving collaborative and semantic representations in their respective spaces. UniSpecRec combines their predictions without cross-space alignment or additional trainable parameters. Extensive experiments demonstrate its effectiveness, efficiency, and generalizability.

Multi-agent LLM systems are increasingly deployed in real-world applications, where failures can be costly and difficult to localize. Despite growing efforts to automate failure attribution, diagnosing failed runs still largely relies on human engineers. Yet engineers rarely debug complex systems by reading raw logs end to end. Instead, observability tools organize traces around components, actions, and dependencies to support targeted navigation. We hypothesize that modern LLMs can benefit from the same paradigm. To test this hypothesis, we introduce Adaptive Influence Graphs (AIGs), a two-stage agentic framework that first transforms a failed trace into a structured graph and then navigates it to identify the critical error. Across multiple models, we show that richer trace representations consistently improve failure attribution, with adaptive graph construction and agent-directed traversal yielding the strongest results. AIGs establish a new state of the art on Who&When, the standard benchmark for multi-agent failure attribution. This affirms our hypothesis that attribution depends not only on the diagnosing model, but also on how the trace is represented and explored.

Language can be considered a design material in architecture, and in the context of text-to-X generative AI models becoming a common tool for architectural practice, looking more closely at language is more important now than in the past. After describing some of the important developments in linguistics starting from Wittgenstein, and including the work of Chomsky, Lakoff, conceptual and generative metaphors as proposed by Schön, this chapter connects them to contemporary architectural design and generative text-to-X tools. The chapter builds on the idea that three main forms of language intertwine in architectural design done using generative AI, namely (I) discourse (or natural language which can contain professional terminology specific to our field), (II) programming languages (which are artificial languages sitting at the basis of all computational systems), and (III) annotations (as language elements attached to pieces of data). It concludes by outlining a research agenda for connecting generative metaphors to generative AI: (a) conducting corpus linguistics studies on architectural texts (using quantitative tools such as topic modelling, and qualitative tools such as discourse analysis); (b) bringing communication theory and information studies closer to architectural research and (c) taking into account that different (natural) languages come with different affordances meaning generative and conceptual metaphors differ in relation to this.

Coding agents perform long-running tasks spanning dozens of model calls, tool uses, and code edits. As these runs unfold, users face a practical cost-quality trade-off: escalating to a stronger model when a cheaper one struggles, or downshifting once the hard reasoning is complete. Each switch requires the receiver to continue a non-native trajectory produced by another model. We study how this handoff affects quality and cost, and how varying the trajectory information inherited by the receiver changes the outcome. Using pairs of low-cost, low-capability (LC) and high-cost, high-capability (HC) models from the Claude and GPT families, we vary handoff direction, timing, and interface, comparing full-trajectory transfer, compaction, and trajectory removal while preserving the repository state. Across both model families, full-trajectory escalation recovers less than half of the LC-to-HC quality gap while incurring a substantial cost premium. We term this cost-quality penalty the handoff tax. By contrast, downshift offers a favorable cost-quality point. Interestingly, the preferred interface also reverses with direction: reducing LC-model trajectory information improves escalation quality, whereas removing the HC-model trajectory reduces downshift quality.

This work extends an earlier proposed universal random-coding ensemble for sample-wise lossy compression of individual sequences to the two-stage (successive-refinement) setting — an extension that is not quite straightforward, as explained in the sequel. The construction has three layers. First, a complete, self-contained direct/converse match in the language of type classes and typical sequences, a two-stage analogue of Rimoldi's classical characterization (applied in the superalphabet of blocks). Second, a realization of the same rates via a random-coding scheme built from the Lempel–Ziv (LZ78) complexity rather than the type-class uniform measure the first layer relies on, since an LZ-based ensemble is universal across source statistics and distortion measures, carries no block-length parameter in the codebook itself, and is directly implementable. Third, we give a fully type-free achievability construction (no joint types of $\ell$-vectors appear anywhere in its search criterion) and show it matches the same converse exactly, reaching every point of the achievable region: the same converse from the first layer already binds any code, type-free schemes included, so no separate argument is needed to certify the match. We also describe (without proof) how to extend the type-free construction to any fixed number $r>2$ of stages. Finally, we show this paper's achievable region is a superset of the one obtained by a companion paper, which is based on a finite-state modeling approach to a similar two-stage problem, extending a known single-stage domination result to two stages.

MLLMs are increasingly deployed in user-facing applications, yet they inherit backdoor risks from the pipelines used to construct them: triggers may reside in images, texts, or both. Existing model-level backdoor removal methods, largely designed for conventional classifiers, show limited effectiveness on MLLMs, while MLLM-specific defenses mainly operate at inference time, filtering suspicious inputs without removing the backdoor embedded in the model. To address this gap and eliminate latent backdoors from MLLMs at their source, we present RACER, a model-level repair framework motivated by a key observation: backdoors induce abnormal layer-to-layer evolution in internal representations, which we term the layer-wise inconsistency anomaly. Importantly, this anomaly is modality-dependent, concentrating primarily in the token region encoding the trigger features that the backdoor model actually relies on. RACER therefore decomposes the fused representation into visual and textual token regions, normalizes their layer-wise inconsistency separately, and recomposes them using modality-aware weights over a deep-layer window, yielding a region-aware inconsistency objective that better captures localized backdoor-induced anomalies. Through a min-max optimization, this objective drives worst-case perturbation synthesis and adversarial fine-tuning against the resulting perturbation to repair the model, suppressing the deep representational directional shifts on which backdoor behaviors rely. RACER requires only 100 clean samples and no knowledge of the trigger, attack objective, or even whether the input model contains a backdoor. Evaluations on three open-source MLLMs across 36 backdoor settings spanning image, text, and multimodal triggers show that RACER reduces the average ASR to 1.1%, reaching 0% in 32 settings, while preserving clean-task utility on both backdoor and clean models.

As neuromorphic systems scale beyond a single core, inter-core event communication can become a dominant contributor to memory footprint, latency, and energy consumption. Biological neural systems address a similar scaling challenge through small-world organization, combining dense local connectivity with sparse long-range projections. In this work, we compare two recent multicore neuromorphic systems implemented in the same 22-nm FDSOI technology and explicitly optimized for such connectivity. The first, NeoCorAl, uses an asynchronous packet-switched tree with hierarchical multicast, whereas the second, MOSAIC, employs an RRAM-based, circuit-switched two-dimensional mesh that performs routing in memory. We examine the resulting trade-offs in routing flexibility, hop count, memory requirements, multicast efficiency, and scalability. We further study how the relative efficiency of tree- and mesh-based routing depends on communication locality in spatially-embedded, random, and layered networks. Finally, we discuss routing-aware training as a means of jointly optimizing neural connectivity, task performance, and hardware mappability.

To reduce the hallucination risk caused by outcome-driven rewards in large language models trained through reinforcement learning with verifiable rewards, existing mitigation approaches introduce process-level factual supervision. However, due to coarse-grained aggregation of factual signals and the lack of reliability assessment for these signals, they create a mismatch between fact verification and policy updates. We term this noisy factual credit assignment and decompose it into two aspects: credit localization ambiguity and credit reliability ambiguity. To address these issues, we propose FARCA (Fact-Aligned Reliability-Aware Credit Assignment), a policy optimization framework that transforms factual supervision into localized, reliability-weighted token-level training signals. FARCA achieves fine-grained credit localization by aligning the granularity of fact verification with that of policy updates. It further introduces counterfactual evidence attribution, which uses the dependence of a factual judgment on key evidence as an empirical proxy for verification reliability to compute reliability weights. These weights modulate factual rewards and local policy advantages, reducing the influence of potentially unreliable signals on policy optimization. Experiments across different models and multiple factual reasoning benchmarks show that FARCA significantly improves model factuality while preserving general reasoning capabilities.

Mixed precision is a promising approach for reducing the computational cost and energy consumption of Computation Fluid Dynamics (CFD) simulations, but its effectiveness depends strongly on where precision is reduced within the full simulation pipeline. In this work, we study Taylor-Green vortex case using Neko, a matrix-free CFD solver based on the spectral element method (SEM). Profiling shows that the fluid time step is not dominated by Krylov convergence alone: the velocity and pressure solvers require only a small number of iterations per step, while a substantial fraction of runtime is spent in other SEM operators and solver components. Motivated by this structure, we propose a three-level hierarchical mixed-precision control model. Two groups of configurations are evaluated in environments bounded by 64-bit floating-point (fp64) and 32-bit floating-point (fp32) precision, respectively. The fp64-bounded group identifies accuracy sensitive components and shows that SEM-focused fp32 computation is a promising direction for future optimization. The fp32-bounded group provides the main practical benefit. For the high Reynolds number case studied, selected configurations reduce both time- and energy-to-solution by about 34% relative to the fp64 baseline, while improving robustness compared with global fp32. Targeted fp16 kernel overrides are also explored, showing potential for selected operations but increased sensitivity in gradient-based quantities such as enstrophy. Overall, these results indicate that mixed-precision for matrix-free SEM-based CFD should be treated as a simulation-level control problem rather than solely as a Krylov-solver optimization.

Clustering is one of the most fundamental tools in data analysis, allowing large datasets to be summarized by a small number of representative points. Given a metric space $(\mathcal{X}, \mathbb{d})$ and a set $S$ of $n$ points in this space, the continuous $k$-median clustering problem asks to find a set $C$ of $k$ points that minimizes the objective function $\sum_{s\in S} \mathbb{d}(s,C)$. When $\mathcal{X} = \Sigma^\ell$ is the set of strings of length $\ell$ and $\mathbb{d}$ is the Hamming distance, the continuous $k$-median clustering problem is known to be W[1]-hard when parameterized by $k$. In this work, we present the first $(1+\varepsilon)$-approximation algorithm for this problem with FPT runtime $2^{\mathrm{poly}(\varepsilon^{-1},k)} \cdot n\ell \mathrm{polylog} \; n$. An additional feature of the algorithm is that it can be implemented in streaming, requiring only $\tilde{O}_\varepsilon(\ell k + k^2)$ space. As an auxiliary tool of independent interest, we show the first streaming algorithm for computing an $\varepsilon$-coreset for continuous $k$-median clustering under the Hamming

Synthetic image generation is a promising strategy to address data scarcity and the underrepresentation of clinically important phenotypes in medical imaging, yet generating images that faithfully reflect meaningful patient characteristics remains challenging. In this work, we investigate metadata-conditioned cardiac magnetic resonance (CMR) synthesis using a pretrained latent diffusion model, encoding structured clinical metadata and slice position as textual prompts to guide CMR generation. To improve metadata adherence and address the imbalance of clinical attributes, we integrate three strategies: Metadata-Free Classifier-Free Guidance (CFG), Contrastive Batching, and Inverse-Frequency Sampling. The framework was fine-tuned and evaluated on 59,058 short-axis CMR from the UK Biobank using paired image similarity, distributional fidelity, and subgroup-level analyses. The combined approach achieved a Fréchet Inception Distance (FID) of 37.47, improving by 57.04% over the same model fine-tuned without these strategies and by 28.68% over a previous text-conditioned CMR diffusion baseline requiring cardiac geometry as additional input, while relying solely on patient metadata. This distributional gain, driven mainly by Metadata-Free CFG, came with a modest reduction in paired similarity, suggesting that the model prioritizes population-level realism over exact image reproduction. Subgroup analyses demonstrated improved alignment across demographic and acquisition-related metadata, with disease-specific conditioning being the most challenging task. These findings demonstrate the potential of generative foundation models for clinically meaningful CMR synthesis while highlighting the need for more effective metadata-aware conditioning strategies. Our code is available at https://github.com/rodriguezmarc/conditional-cmr.

The prediction of student engagement from the online tutoring videos is difficult because engagement is a multidimensional construct comprising distinct behavioral, emotional, and cognitive states. A reliable prediction requires bringing together different types of behavioral signals as well as expressive cues. Through our analysis of the CASED dataset, it is clear that engagement prediction gets even harder due to the high inter-person variability as well as the subjectivity of the engagement annotation. To tackle these challenges, we develop a multimodal framework that integrates the implicit spatiotemporal features extracted from pretrained video, audio, and image encoders along with structured behavioral modalities like head pose, gaze, facial action units, emotion, and wavelet-based audio features. We integrate these modalities via a Perceiver IO latent bottleneck. Moreover, student and instructor personalities are modeled as variational posteriors over learnable embeddings to enable partial pooling across participants. We employ evidential regression and spectral-normalized Gaussian process classification heads for uncertainty-aware prediction to further improve robustness and calibration. Benchmark on the CASED challenge test set shows that all participating methods converge near random-chance performance, revealing the difficulty of the dataset. In this highly ambiguous regime, our framework achieves competitive performance while uniquely offering well-calibrated uncertainty metrics, demonstrating that reliable risk-quantification is an essential prerequisite for deploying engagement models in real-world educational tools.

Inference-time decoding methods improve LLM reasoning by exploring multiple candidate trajectories, yet treat each trajectory as atomic: either retaining it whole or discarding it irreversibly. This wastes computation on partially promising candidates whose high-quality prefixes are abandoned alongside degraded suffixes. We introduce Selective Regenerative Decoding (SRD), which routes each candidate to discard, keep, or refine only the degraded portion of the suffix while preserving the useful prefix of borderline candidates, without requiring a larger target model. Under mild assumptions, SRD achieves a provable 1.28-to-1.36-fold gain in sample efficiency over rejection sampling with strictly higher expected trajectory quality, with the gain growing as the candidate pool grows. Across MATH500, GPQA Diamond, HotpotQA, and AlpacaEval with multiple generation-reward model pairs, SRD matches Best-of-N accuracy with substantially fewer generated tokens and outperforms speculative rejection in low-compute regimes. By enabling segment-level intervention rather than whole-trajectory selection, SRD opens a previously underexplored region of the accuracy-compute tradeoff for inference-time reasoning.

Metric bases of graphs have been widely studied since their introduction in the 1970's by Slater and, independently, by Harary and Melter. In this paper, we concentrate on the existence of vertices in a graph $G$ that belong to all metric bases of $G$. We call these basis forced vertices, and denote the number of them by $\mathrm{bf}(G)$. We show that $\mathrm{bf}(G)\le 2/3(n-k-1)$ for any connected nontrivial graph $G$ of order $n$ having $k$ vertices in each metric basis. In addition, we show that this bound can be attained. Furthermore, the previous result implies the bound $\mathrm{bf}(G)\le 2/5(n-1)$ formulated in terms of the order $n$ of the graph for any nontrivial connected graph $G$. This result answers a question posed by Bagheri et al. in 2016. Moreover, we provide a complete realization of the parameters $n$, $\dim(G)$ and $\mathrm{bf}(G) \ge 1$ within the previous bounds. We consider some extremal cases related to basis forced vertices in a graph, in particular, we give a full characterization of the graphs with $\mathrm{bf}(G) = 2$ and $\dim(G) = n-4$.

Activation steering can change behaviour without establishing that the effect is specific to the intended concept. We introduce SteerCheck, a preregistered attribution audit that matches off-target KL and separates mean, protected-tail, polarity, transfer, and semantic claims. Exact replay of 960 Qwen3-14B interventions reveals complementary limits of common controls: isotropic directions occupy a narrow near-orthogonal region, whereas sign-randomized same-construction directions often retain substantial target alignment. Effect is strongly associated with signed cosine within the sign-randomized family ($\rho=.94$); $25.3\%$ of its draws exceed cosine $.5$, and every draw exceeding the observed mean effect has cosine above $.80$. This alignment leakage does not by itself invalidate a conditional randomization test; it limits what the comparator can distinguish and motivates reporting exchangeability assumptions, a construction diagnostic $A$, and the empirical cosine distribution. The primary Qwen complete gate remains negative because the protected tail fails all families. On independent data, continuous margin transfers only in Qwen and accuracy transfers in no selected cell. Prospectively registered language controls pass the complete gate in Qwen and DeepSeek, while a passing DeepSeek detox comparator rules out categorical separation; all nominal passes are sensitive to $\Gamma=1.10$. Frozen three-rater open-generation evaluation supports factual correction in DeepSeek but not Qwen; the automatic judge fails calibration (macro-F1 $.562$), so null-wide semantic results remain descriptive. SteerCheck makes these conditional and mixed conclusions auditable.