Current agentic systems often hallucinate or drift during multi-turn reasoning because they rely on unstructured token-level generation. This paper proposes SGUR, a training-free framework that grounds multi-step inference in structured hypothesis spaces using partial graph alignment and active information gain. Evaluated on a sequential medical-diagnosis benchmark derived from PrimeKG, the results suggest that tighter structural discipline can partially offset the need for ever-larger language models.

This mixed-methods study examines how university students perceive their generative AI skills changing over time and whether those perceived trajectories connect to shifts in learning practices and academic outcomes. Using a retrospective survey of 313 students and interviews with 18 across disciplines, the authors identify heterogeneous developmental patterns—most commonly rapid early improvement followed by a plateau—and find that self-reported academic gains are linked to assignment-process adaptation rather than to a specific trajectory. The work highlights uneven growth across skill domains and signals that institutions should tailor support beyond generic tool training, though the cross-sectional design cannot establish actual longitudinal change.

Multimodal in-context learning remains weak because MLLMs often hallucinate and fail to ground textual outputs in the visual evidence provided in demonstrations.
This work argues that the root bottleneck is poor visual grounding rather than a lack of ICL-specific training data.
The authors propose ProCoRe, which uses proposition-level contrastive reinforcement learning on non-ICL image–caption pairs to sharpen fine-grained visual discrimination.
The result is stronger multimodal ICL without catastrophic forgetting, plus a new personalized-captioning benchmark that forces models to rely on demonstrations instead of lexical priors.

The paper tackles the closed-world problem in cybersecurity intrusion detection, where supervised models trained on benign and insider-threat labels fail when confronted with entirely novel botnet activation attacks at test time. The core idea is a hybrid fuzzy ensemble that couples a gradient-optimized Takagi-Sugeno-Kang (TSK) classifier for known suspicious behaviors with a Fuzzy C-Means (FCM) module that flags out-of-distribution (OoD) samples via a raw Min Distance metric rather than traditional membership scores. The work is relevant because it directly targets the operational reality that security operations centers must detect attack vectors never seen during training, and it attempts to do so with an interpretable, rule-based architecture rather than a black-box neural network.

Personalized mobile GUI agents must decide both how user preferences alter execution and when they apply along a trajectory. Existing benchmarks only report aggregate task success, obscuring where and why personalization fails. This paper introduces PrefDiag-Bench, a diagnostic benchmark with 318 end-to-end tasks and 1,664 stage-wise probes that categorize preferences into five execution types and pinpoint bottlenecks in evidence selection and trigger recognition. The result is a reproducible tool set for moving beyond coarse success rates toward targeted improvement of personalization capabilities.

AI agents that act as statistical analysts evolve through model upgrades, prompt edits, and new tools, but each change risks inflating Type I error. This paper introduces continual validation: before any update is deployed, it is frozen and tested on newly simulated null data using exact binomial confidence bounds with a decaying per-update risk budget. The result is a lifetime guarantee that no invalid version passes, which matters because classical fixed-procedure guarantees break under adaptive, open-ended evolution.

Steward is a multi-agent LLM trading system designed to act on live Web data while keeping its safety constraints intact as it learns. The paper introduces governed self-evolution: a deterministic gate and typed proposal system let the agent rewrite its own prompts, but an invariant core of risk rules, order arithmetic, and discipline-agent prompts is unreachable by any self-edit. In a five-month live paper-trading deployment the authors record every input, output, and grading event to evaluate whether the architecture survives real market noise, prompt injection, and its own reflection loop. The release of the full code, prompts, and replay harness makes this one of the most transparent live-agent studies to date.

Pairwise comparison data power modern preference learning, but disagreement in such data can mean persistent heterogeneity, random noise, or hidden non-binary responses. This paper develops a nonparametric latent semantic model with three querywise quantities—aggregate preference $\mu_q$, persistent disagreement $\tau_q$, and non-binary mass $\alpha_q$—and proves sharp identifiability frontiers. Anonymous binary feedback point-identifies only $\mu_q$, while same-annotator repeats identify $\tau_q$ via covariance and abstention identifies $\alpha_q$ directly; only their combination recovers the full semantic triple. The results imply that semantics-aware decisions like denoising versus personalization cannot be justified from anonymous forced-choice data alone.

SAM3RL replaces the hand-crafted FIFO memory rule in SAM-based visual trackers with a lightweight controller learned by reinforcement learning. It frames memory-bank eviction as a Markov Decision Process and trains a 3.2 M-parameter MLP with PPO on bounding-box LaSOT sequences to maximize tracking quality. The resulting policy sets a new state of the art on 10 of 12 benchmarks for both SAM 2 and SAM 3 backbones, while remaining fully online and running without extra feature extraction.

Offline RL promises to outperform the data but is notoriously unstable, whereas imitation learning is stable yet bounded by data quality. This paper introduces Advantage Guidance (AG), a policy extraction method that retains a simple supervised-learning objective—training a flow policy conditioned on advantage estimates—while correcting the failure mode where naïve conditioning is ignored by the network. At test time, AG uses a derived contrastive diffusion guidance to target the KL-regularized optimal policy $\pi_{\mathrm{KL}}^*(a \mid s) \propto \hat{\pi}(a \mid s)\exp(A(s,a)/\beta)$. On 26 long-horizon sparse-reward manipulation tasks, the method outperforms strong actor-critic, supervised-learning, and diffusion-guidance baselines.

Population health stratification traditionally groups patients by clinical complexity, yet these descriptive frameworks offer little guidance on how to prioritize limited care resources. PROMETHEUS reformulates this task as a multi-treatment causal ranking problem: instead of estimating precisely calibrated treatment effects for every patient, it learns a global priority ordering over patient–profile opportunities by combining cross-fitted doubly robust pseudo-outcomes with a Siamese contrastive neural network. This matters because it separates clinical need from actionability, allowing health systems to identify which eligible patients should receive more intensive care first when capacity is constrained.

ActMOF addresses the challenge of optimizing metal-organic framework (MOF) syntheses when experimental data are sparse and outcomes noisy, by introducing an experimentally grounded benchmark for comparing active-learning policies. Starting from only 95 measured microwave syntheses of an aluminum-carboxylate MOF, the authors emulate a landscape of over six million encoded conditions and evaluate 39 policies—Gaussian processes, tree ensembles, and large language models—before transferring the best strategy to six wet-lab syntheses. The demonstration that all six campaigns yielded crystalline products within a 12-experiment budget makes the work relevant as both a reproducible testbed and a proof of concept for transferable, AI-guided reticular chemistry.

Looped Mixture-of-Experts models reuse parameters across depth, yet prior work treats each loop visit identically. This paper shows that loop visits learn strongly asymmetric residual scales—the first visit updates representations far less than the second—and proposes LatentLoop, which compresses the first pass into a latent sequence and reallocates saved FLOPs to model width. Extensive scaling experiments up to $7$B parameters show consistent validation and downstream gains, framing asymmetric compute allocation as a principled way to improve looped MoE efficiency.

Goal-conditioned RL in complex mazes suffers from a geometric mismatch: Euclidean embedding distances model straight-line proximity, yet walls can make actual route distances highly tree-like and deceptive. This paper introduces Hyperbolic CRL, a drop-in replacement of the critic's Euclidean distance with a hyperbolic geodesic distance on the hyperboloid. Its central finding is that this geometric alignment allows a depth-64 hyperbolic network to outperform a 1024-layer Euclidean baseline on hard humanoid maze tasks while training more than $12\times$ faster and using roughly $13\times$ fewer parameters.

Foundation models for dynamical systems can roll out unseen PDEs and ODEs, but it is unclear whether they genuinely learn in-context forecasting strategies or merely memorize. This paper proposes Test-Time Forecasting (TTF), a framework in which looped transformers use recurrent depth to perform gradient-descent iterations on a latent forecasting objective at inference time. A lightweight TTF model trained only for one-step prediction can generalize to harder, out-of-distribution system dynamics during autoregressive rollout, suggesting that recurrent depth offers a principled, parameter-efficient path to in-context learning for continuous dynamics.

Predicting which cellular region a mobile user will visit next lets operators cache content and allocate resources proactively, yet a predictor that never abstains commits to choices even when its evidence is weak.
This work applies MetaModelSelect, a lightweight post-hoc metamodel that scores the trustworthiness of a frozen RNN's top-1 next-location decision using only its output logits: a learned embedding of the predicted region, the four largest softmax probabilities, and a global concentration statistic.
By filtering predictions through this learned correctness score rather than retraining the backbone, the paper shows how to trade coverage for selective risk in a compact, deployable wrapper.

Traditional student models only use responses to standalone questions, ignoring the rich, fine-grained evidence embedded in tutoring dialogues. This paper proposes URF-QD, a framework that rewrites tutor-posed mathematical subtasks as self-contained questions and models them jointly with original items on a shared item response theory scale. The key novelty is a relative-order regularization that uses tutor pedagogical moves to constrain consecutive subtask difficulties, enabling dynamic tracking of student ability across both questions and dialogues. The authors show consistent improvements on two real-world math datasets, though the approach leans heavily on LLM annotations and modest sample sizes.

Batch black-box optimization requires selecting many parallel queries across only a few expensive rounds, yet conventional Bayesian optimization suffers from costly joint acquisition maximization and per-task surrogate refitting. This paper proposes Flow Matching for Thompson-Sampling Refinement (FMTR), an amortized approach that offline-learns a continuous, history-conditioned proposal distribution over query locations from which variable-size batches can be directly sampled at test time with fixed parameters. By marrying in-context adaptation to a generative flow model, FMTR aims to eliminate online acquisition optimization while preserving sample efficiency.

This paper studies whether learning an epsilon-optimal average-reward policy from a single continuing trajectory must pay an extra navigation cost compared with generative access. It develops an adaptive algorithm that learns a low-cost sampling allocation via inverse-count exploration and converts nonuniform samples into an all-state policy certificate using a transient-corrected quantile bound. For broad model classes and sufficiently high accuracy, it proves that the worst-case sample complexity matches the generative lower bound up to logarithmic factors, removing the extra diameter factor that plagued prior continuing-access bounds.

Vision-language models are moving onto battery-powered edge devices, yet their total-system energy footprint on NPU-equipped single-board computers has remained largely unmeasured. This paper presents a hardware-instrumented study of Qwen2-VL-2B-Instruct inference on a Raspberry Pi 5 with a Hailo-10H NPU and a Jetson Orin Nano Super, separating fixed request energy from per-token decode costs. It shows that NPU offloading cuts Pi energy per token by roughly two-thirds, that text generation dominates vision encoding at typical response lengths, and that the Pi-NPU and Orin-GPU stacks have comparable per-token energy with a crossover near 40 output tokens. The findings are practically useful for engineers choosing deployment stacks under power and thermal constraints.