Science & Mathematics

Large language models (LLMs) can leverage semantic feature names alongside summary statistics for tabular feature selection, yet it remains unclear when such information improves subset selection relative to established statistical methods. We investigate this question within an 18-dataset framework comprising 10 datasets for Group Relative Policy Optimization (GRPO) training, 2 for checkpoint validation, and 6 held-out evaluation datasets. We compare five classical baselines, frozen Qwen2.5-7B and Llama-3.3-70B selectors, and two GRPO-adapted Qwen2.5-7B models, reporting classification ROC-AUC and regression R^2 separately and using dataset-wise ranks for cross-task comparison. Across five seeded evaluation partitions, classical selectors generally achieve higher predictive performance than the tested 7B LLM/RL selectors, including under a pre-specified ~33% feature budget with strict train-only classical fitting, while the 7B selectors retain compact subsets. A cardinality-matched semantic sensitivity experiment shows that, for Zero-shot Qwen-7B, informative feature names improve regression performance over anonymized and shuffled alternatives (+0.1964 and +0.1606 mean R^2), whereas aggregate classification differences are smaller (+0.0159 and +0.0084 ROC-AUC) and dataset-dependent. These results characterize when semantic information benefits LLM-based feature selection and show that classical selectors remain strong baselines for predictive tabular feature selection.

Airy beams are often proposed in array beamforming to bend beams around obstacles. We prove that, for a single-antenna receiver, conventional beamfocusing maximizes the received power regardless of how many antennas are blocked and, in expectation, even with phase errors in the channel state information. This also holds for multi-antenna receivers under a phase-separability condition that is satisfied in rank-one channels. Numerical results confirm that unmodified beamfocusing is optimal for physically small receivers (i.e., those that fit within the beamwidth) when an obstacle is present. For physically larger receivers, the optimal precoding deviates from beamfocusing to better illuminate the receiver's unobstructed parts, but will not become Airy beams. While Airy beams can be configured to perform well, their ability to follow a bent trajectory in front of an obstacle does not maximize the received power behind it.

Continuous flow language models offer a path to parallel generation with tools for control and acceleration unavailable to discrete diffusion, yet effective techniques for building them at the billion-parameter scale remain less established. We present Lafa (Large flow language model), a fast, high-performing flow language model scaled to 8B parameters in a data-efficient manner. Lafa is a blockwise flow model with causal attention on the prefix and bidirectional, asynchronous flow within each block. This design enables efficient training by adapting pretrained autoregressive (AR) models and fast inference via prefix caching and blockwise parallel decoding. We train Lafa at 0.5B, 7B, and 8B scales with fewer than 30B tokens, more efficient than previous diffusion models, and evaluate it on instruction following, reasoning, and coding, where the learned flow matches or surpasses the quality of AR and can be distilled into flow maps for one- and few-step generation. Notably, flow training preserves or even improves pretrained AR capabilities via a multi-token prediction effect, which we leverage in a self-speculative decoding scheme that verifies flow-based drafts autoregressively within a single model. This yields 2.54x the throughput of an AR model at comparable quality, outperforming multi-token prediction and discrete diffusion methods. Our findings show that continuous flows can serve as a scalable foundation for language modeling.

Visual anomaly detection aims to detect abnormal objects, while anomaly samples are often scarce during model training. Assembling multiple pretrained detectors can help by leveraging their transferable image scores and anomaly maps. However, most existing ensemble methods apply fixed fusion rules over saved predictions which struggle when pretrained detectors clearly disagree due to different training objectives, priors, and visual representations. To this end, we design AnomalyAgent, a post-training fusion framework for accurate visual anomaly detection. AnomalyAgent aligns heterogeneous scores and maps, summarizes score- and region-level disagreement, and treats fusion as a finite action-selection problem. It implements two types of actions: 1) score actions that determine how to fuse image-level detector confidence via consensus, robust aggregation, detector weighting, or score-map consistency; 2) map actions that fuse dense anomaly responses via consensus, detector weighting, geometric aggregation, or disagreement correction. A constrained vision-language model (VLM) is introduced to inspect high-conflict samples and select valid score action and map action to produce the final prediction. We evaluate AnomalyAgent on MVTec AD and VisA as primary benchmarks, together with seven additional public anomaly benchmarks and Real-IAD for large-scale score-level evaluation. Across these benchmarks, AnomalyAgent consistently improves over equal-weight fusion and achieves the best mean image AUROC and pixel AP among the compared fusion methods on MVTec AD and VisA, while demonstrating robust transfer across additional anomaly domains.

Stochastic gradient descent (SGD) is the cornerstone of modern deep learning, yet its theoretical convergence properties remain poorly understood in non-convex settings. To address this gap, we study SGD dynamics in deep linear neural networks - simplified models that retain key non-convex characteristics while remaining analytically tractable. Building on recent results for gradient descent, we extend convergence guarantees to SGD under regularized squared loss. We establish almost sure boundedness of SGD iterates. Using the framework of asymptotic pseudotrajectories from stochastic approximation theory, our analysis establishes then, for both deterministic and adaptive step-size choices, convergence of SGD to the set of critical points under mild conditions (without relying on Lipschitz properties or whitened data). To our knowledge, this is the first work providing almost sure convergence guarantees for SGD in deep linear networks, bridging the gap between theoretical optimization and practical stochastic training.

Many scientific phenomena are best understood as outcomes of underlying optimization problems, yet automated scientific discovery has largely focused on recovering equations that describe observed input-output patterns. We introduce OptScientist, a novel automated framework for discovering compact symbolic optimization laws directly from data. OptScientist comprises (i) a model that infers symbolic optimization problems by first learning to represent data as an optimization problem and then extracting its symbolic form, and (ii) an LLM agentic system embedding the model in a discovery process that integrates theoretical priors from scientific literature, discovers candidate laws, abstracts their underlying mechanisms, and suggests new data to collect. We first evaluate OptScientist on OptFeynman, a novel benchmark of 99 physics-derived optimization problems, where OptScientist recovers ground-truth laws for over 80% of the problems and remains robust to noisy contextual information, while substantially outperforming machine-learning, symbolic, and LLM-based model-discovery baselines. On human risky choice and central-bank monetary policy, OptScientist discovers compact symbolic optimization laws that achieve held-out predictive performance competitive with leading baselines, while using few parameters and substantially fewer tokens than LLM baselines. Overall, OptScientist extends automated scientific discovery to uncovering optimization laws that generate observed phenomena, enabling automated theory building across the physical, biological, behavioral, and social sciences.

Online search activity is widely used as a digital trace of information seeking during epidemics, but its predictive value is often reduced to correlation, a selected lag, or a single forecast-error score. We examine eight pre-specified Google Trends-COVID-19 pairs in both directions using matched LightGBM forecasts over rolling 400-day windows and horizons of 1-14 days. Log-RMSE gain is combined with forecast-error distribution entropy (FEDE), which captures horizon-dependent changes in the signed error distribution. The resulting directional predictive signature comprises response class, RMSE effect, FEDE endpoint and early-slope responses, persistence across rolling regimes, and horizon profile. For screening candidate directions, we also introduce a relative persistence-weighted directional predictive response strength (DPRS) on a within-study 0-100 scale; 100 denotes the strongest observed response in the comparison set, not an absolute coupling maximum. The pairs show concordant, RMSE-only, FEDE-only, and mixed responses. "loss of smell" -> reported cases is accuracy-dominant, with mean log-RMSE gain 0.104 (about 9.9% equivalent RMSE reduction) and DPRS 100, whereas the reverse direction is FEDE-dominant (D_FEDE=0.048; DPRS=42). "covid test" -> positive rate is FEDE-dominant (D_FEDE=0.039; positive in 72% of windows; DPRS=98) despite negative RMSE gain, while "lockdown" -> stringency is the clearest concordant case (DPRS=78). World-level symptom/case pairs are less persistent. Thus, forecast accuracy and error-distribution dynamics are complementary, predictive effects can be direction- and metric-dependent, and forecast horizon is part of the relationship. Aggregated regime-level patterns should not be extrapolated to individual time points or interpreted as evidence of causality.

Operando and in-situ experiments extend powder X-ray diffraction (PXRD) to materials that evolve during experiments, producing series of hundreds or thousands of patterns. Such series are commonly refined one pattern at a time, so weak signals give noisy structural trajectories. Parametric refinement instead fits a shared model to all patterns, but with most existing programs, changing the model or fitting it efficiently remains challenging. Here, we introduce diffPXRD, a differentiable PXRD engine in PyTorch. diffPXRD calculates derivatives by automatic differentiation, so a shared model can be changed without new derivative code. Batched fitting is also more than seven times faster than serial fitting. In an operando cathode series, a shared lattice model can reduce lattice-parameter variability by 68.4% when 10% of the photon counts are retained. We also show shared models can describe the precipitation and dissolution of Li2S and reconstruct collective atomic displacements in Ca3Ti2O7.

Clustered federated learning (CFL) can suffer from cluster collapse. In particular, under unfavorable initialization, early client assignments concentrate heterogeneous groups on one cluster model, which is then trained toward a pooled compromise, while another model receives no updates. Since the current models determine client routing and routing determines subsequent model updates, we formulate clustered federated learning as control of an unknown dynamical system, with cluster models as the evolving state, client routing patterns as the control input, and routing interventions as state dependent controllers. In a two group linear model, we identify an initialization region in which greedy routing makes collapse self reinforcing. We then derive finite horizon error dynamics showing how routing interventions can reverse collapse by repeatedly training the starved model, primarily with updates from its compatible group. Motivated by this condition, each client uses blocked EXP3 to choose between greedy routing and interventions that redirect updates toward unused models. Experiments show that with simpler predictors and faster specialist formation, autonomous controller selection recovers usable specialists from unfavorable starts that trap standard CFL baselines. In transformed FEMNIST, the synchronized-entry configuration with persistent commitment recovers usable specialists in settings where the autonomous controller configuration often does not. Code and scripts for reproducing the experiments and figures are available at https://github.com/prsinha1/FedRecover.

Robots exploring unknown indoor environments must decide where to move while much of the environment remains unseen. Predicting the occupancy of unseen space beyond the current sensor range can improve these decisions, but recent approaches commonly rely on learned models that require offline training and repeated inference during exploration. We introduce the Mondrian-Anchored Prior for Learning-free Indoor Exploration (MAPLE), a training-free predictive approach that adapts the Mondrian process into an online generative prior conditioned on observed geometric cues in the current partial map. MAPLE generates multiple plausible map hypotheses and evaluates visibility separately within each to estimate frontier gain, preserving differences in predicted visibility across hypotheses. Across KTH floor-plans and real-SLAM environments at multiple sensor ranges, MAPLE consistently improves over non-predictive exploration and remains competitive with learned predictive approaches. Its lightweight procedural construction enables predictive exploration with limited computational resources. Experimental results further show that more accurate map prediction does not necessarily produce better exploration, highlighting the importance of how predictions are represented and used in exploration decisions.

Text-to-image personalization aims to equip pretrained diffusion models with the ability to learn user-specific concepts from only a few reference images. Existing methods, however, typically train each concept independently or assume that all concepts are available simultaneously. In practical applications, concepts often arrive sequentially, requiring models to continually learn new subjects, objects, or styles while minimizing interference with previously acquired concepts. We address this continual concept personalization setting by extending residual-space adaptation to the stochastic adaptation process of pretrained diffusion models. Our work achieves the highest concept fidelity among competing continual personalization methods while maintaining near-zero forgetting. Beyond these empirical gains, the learned residual-space dynamics reveal where adaptation capacity is concentrated within the diffusion model. Restricting adaptation to the identified layers by our method preserves nearly the same performance while reducing the number of trainable parameters by up to 72\%, demonstrating that residual-space dynamics can identify the layers that contribute most to continual concept personalization and directly guide more parameter-efficient adaptation.

Generative retrieval (GR) casts search and recommendation as autoregressive generation of discrete semantic identifiers (SIDs). SID design largely determines GR quality, but its discrete, interdependent components (SID construction, collision handling, decoding, and training) cannot be optimized end-to-end with gradients and remain largely hand-designed. We introduce a framework that improves these components through LLM-driven evolutionary program search, to our knowledge the first application of this approach to GR. A locally served, open-weight 31B model mutates executable GR components, so no proprietary LLM API is used, and a 1M-parameter proxy decoder provides fast feedback in place of repeatedly training the 1B target generator. Across 21 SID programs on NQ320K, proxy and target Recall@100 show strong rank correlation (Spearman $ρ= 0.90$). Guided by this proxy, evolution improves the seed GR on NQ320K, MSMARCO300K, and a multilingual e-commerce dataset under a fixed training recipe. On NQ320K, jointly evolving SID construction and collision handling improves Recall@1 from 55.8% to 59.3% and Recall@100 from 85.8% to 90.6%. A component study shows that proxy reliability depends on what is evolved: improvements transfer reliably for the semantic index but not for the generator's training objective or the full pipeline. LLM-driven evolutionary search is thus a practical tool for GR design, provided proxy reliability is established for the chosen search space.

Modeling lifetime user behavior sequences is essential for capturing long-range preferences in industrial recommendation, yet sequences containing up to 100K interactions make target-aware ranking prohibitively expensive. Search-based methods that retrieve from raw sequences incur candidate-dependent costs that grow with sequence length, whereas latent compression methods reduce computation at the expense of item-level addressability. We propose ECHO (Efficient Cascade with Homogeneous Operators), a framework that balances efficiency and fine-grained retrieval by retaining a subset of original event records per user through a Distilled Selector (DS) and retrieving a smaller set of target-relevant records for each candidate through a Target-aware Fine-grained Retriever (TFR). DS is supervised by a candidate-aware full-history teacher via per-target distillation, enabling target-aware user-level selection without requiring full-history computation for every candidate. The retriever and target-attention scorer share the relevance parameters learned by the teacher, ensuring consistent relevance estimation throughout the cascade. At a sequence length of 100K, ECHO reduces online sequence-branch FLOPs by approximately 160 times compared with a representative end-to-end model. Three deployment optimizations, including common user inputs, compact record encoding, and key-side projection absorption, further reduce memory, storage, and data-movement overhead. Experiments on KuaiRand-1K and an industrial-scale short-video platform demonstrate relative GAUC improvements ranging from 0.113% to 1.520% over state-of-the-art baselines across multiple tasks. Furthermore, online A/B tests in two production scenarios, Kuaishou and Kuaishou Lite, show that ECHO improves app usage time by 0.240% and 0.373%, respectively.

Greenhouse supplemental lighting is essential for meeting crop Daily Light Integral (DLI) requirement. However, conventional always-on and rule-based strategies are inherently energy-inefficient since they ignore future sunlight availability. This paper reports the results of deploying various supplemental lighting control strategies, namely, always-on, always-off, heuristic, optimal prediction-based, and two peak-shift variants, simultaneously in a single greenhouse at the Hampton Roads AREC of Virginia Tech, to ensure that every treatment experiences identical conditions. %The shared infrastructure is built on an MQTT-based IoT platform with Photosynthetic Photon Flux Density (PPFD) sensing and PWM-driven LED control; The prediction-based and peak-shift strategies optimize LED schedules using a Markov chain-based solar forecast. Experimental results over two replicates %(20~February--15~April~2026, 72~plants per treatment) show that the prediction-based strategy meets the desired DLI target at 65\% cost reduction relative to the always-on baseline while achieving the highest dry-mass yield among the proposed adaptive strategies. The heuristic strategy incurs only a 6\% cost premium over prediction-based control, offering a competitive forecast-free alternative. The peak-shift variants achieve the lowest active-strategy costs by shifting LED demand into off-peak tariff windows. A yield-per-dollar efficiency metric is finally introduced to jointly capture energy cost and crop output showing that the prediction-based strategy ranks highest at 18.976~g/\$, more than double the always-on value of 9.391~g/\$. These findings can provide greenhouse operators with an evidence-based framework for selecting a lighting control approach that balances energy cost, crop quality, and system complexity.

ScreenCalib enables light detection and ranging (LiDAR)--camera extrinsic calibration using an ordinary programmable display, providing known visual markers and controlled geometry while reducing dedicated target preparation. The camera observes displayed AprilTags, whereas the LiDAR measures the physical screen surface and boundary, requiring indirect cross-modal association. Display geometry introduces association ambiguities, while indoor clutter complicates extraction. ScreenCalib addresses these challenges through tag-guided display extraction, geometric hypothesis generation and refinement, and surrounding scene edges for ambiguity resolution. The single-pose mode requires only one static placement. The joint multi-pose mode combines independent static observations through cross-pose consensus and jointly refines the shared extrinsics and tag scale, reducing sensitivity to tag-size measurement errors without requiring continuous motion or inter-pose trajectory estimation. Experiments in simulation and on three different sensor platforms cover mechanical and nonrepetitive LiDARs in clean and cluttered environments. Benchmark results show that ScreenCalib outperforms competitive calibration-board-based methods in translation accuracy. Joint multi-pose refinement further reduces mean rotation and translation errors across the three platforms and improves rendered-map alignment. Evaluation across readily available displays, including cell phones, tablets, laptops, computer monitors, and televisions, further examines ScreenCalib's versatility as a practical alternative to dedicated calibration targets. The source code is available at https://github.com/ZuhaoZou2026/ScreenCalib.