Papers

We present Pow3R-SLAM, a real-time RGB-D simultaneous localization and mapping (SLAM) system that uses Pow3R for tracking and mapping. Inspired by MASt3R-SLAM, a recent work on monocular SLAM using two-view 3D reconstruction priors, we extend the work to incorporate depth as a prior on the network's prediction, rather than as geometry to fuse. Where traditional RGB-D SLAM systems struggle with sparsity in the depth images, Pow3R utilizes the available depth to give a better-conditioned pointmap, while inferring the depths in empty regions from the two-view photometric, depth, and intrinsic data. Evaluated against MASt3R-SLAM following its protocol on 24 sequences from TUM, 7-Scenes, and Replica, Pow3R-SLAM runs 1.6x faster in wall time, has 15% lower mean trajectory error, a 3.1x lower unscaled error, and produces denser maps, with a 30% lower Chamfer distance. We also introduce a hybrid variant that runs 2.1x faster than MASt3R-SLAM at 25.3 frames per second (FPS), while maintaining improved tracking and mapping accuracy. Against ORB-SLAM3 in RGB-D mode, Pow3R-SLAM is more accurate on TUM, 7-Scenes, and ETH3D-SLAM, and completes every TUM sequence. While Pow3R-SLAM can struggle on a small set of self-similar scenes, its overall performance shows that adding depth as a prior for two-view 3D reconstruction SLAM can be beneficial. A project webpage is available at: https://ChrisKolios.github.io/Pow3R-SLAM , and code will be made open-source upon acceptance.

Approximate filtered vector search (FVS), a core operation in many data management tasks that combine structured data with vector embeddings, exhibits increased complexity due to the characteristics of filtering predicates. Each predicate is defined by selectivity (i.e., the fraction of vectors that satisfy the predicate) and correlation (i.e., the relationship between the filter and the vector space), which can significantly affect search difficulty even for the same query vector. This poses a key challenge for users aiming to integrate vector search with structured data, as efficient execution often requires extensive manual tuning of algorithm parameters. In this paper, we present VADER, the first approach that eliminates hyperparameter tuning by introducing declarative recall for approximate filtered vector search. With declarative recall, users specify a desired recall target, and VADER executes FVS queries to meet this target without requiring manual configuration. VADER achieves this by employing a filter-aware recall predictor that generalizes across varying selectivities and correlations without explicit tuning, and by performing early termination once the predicted recall reaches the user-defined target. Through extensive experimental evaluation, we show that VADER achieves near-optimal early termination, while providing significant speedups of up to 53% faster and improved result quality of 28% compared to the best-performing baseline.

We study generalized learning dynamics in multi-agent systems whose joint state evolves on a manifold and whose agents act through state-dependent, potentially nonholonomic vector fields. Under a bundle-splitting condition, we show that these dynamics admit an intrinsic representation as projected Riemannian gradients, giving rise to a class of nonholonomic gradient play dynamics. We characterize the local stability of its equilibria through an intrinsic linearization that explicitly captures the effects of the Riemannian connection and the nonholonomy of the actuation frame. The framework recovers classical gradient play on Euclidean spaces and manifolds as special cases while accommodating nonholonomic actuation. We then develop a geometric theory of deception under asymmetric information, whereby an agent exploits privileged knowledge of other agents' learning rules to manipulate the emerging equilibrium. We show that deception effectively tilts the Riemannian geometry perceived by the oblivious agents, distorting their projected gradient directions. We establish persistence of exponentially stable equilibria over an open set of deception parameters and derive an explicit first-order characterization of the resulting equilibrium displacement. Analytical and numerical examples illustrate the framework and its implications for equilibrium manipulation in games.

Generative models for function-valued data, such as time series and solutions of partial differential equations, must learn distributions over infinite-dimensional spaces. Functional Flow Matching (FFM) extends Flow Matching to this setting, learning a velocity field whose flow transports a Gaussian prior to the data distribution, but it inherits the independent endpoint pairing of standard Flow Matching: in each batch, prior and data samples are matched arbitrarily, so the conditional bridge must traverse both the shared global structure of the dataset and instance-specific residuals. In function space this is harder to fix than in finite dimensions, since optimal transport (OT) on function spaces is delicate to formulate and a flat Euclidean surrogate ignores the geometry that distinguishes function-valued data. We propose kernel Functional Flow Matching (kFFM), which replaces the independent pairing by entropic OT under a kernel-induced cost, the coupling underlying the Hilbert Sinkhorn Divergence (HSD), leaving the FFM neural-operator architecture unchanged. We prove that the kernel cost and the HSD objective are uniformly bounded and well-posed on Banach ambient spaces, derive an error decomposition against quadratic-cost OT on compact metric spaces that isolates an irreducible kernel-cost mismatch term, and prove a discretization-invariance bound whose rate is governed by Sobolev regularity. Empirically, kFFM improves distributional matching over FFM, diffusion, adversarial, and finite-dimensional OT baselines on time-series and PDE benchmarks, with significant paired-seed gains over FFM and improvements that persist under non-kernel and physics-based diagnostics, including a turbulent Navier-Stokes benchmark. Bounded kernel costs already outperform raw $L^2$ Sinkhorn, and function-space-aware kernels (signature, Sobolev RBF) give further gains on rough or path-valued data.

We propose a QCQP-representable IMU pre-integration factor that enables certifiable estimation with pre-integrated inertial measurements. To the best of our knowledge, this is the first work to directly incorporate IMU pre-integration into certifiable estimation. Inertial sensing is a common and reliable modality in robotics, and incorporating it broadens the practical scope of certifiable estimation. The main challenges are obtaining the required algebraic structure and a sufficiently tight convex relaxation. Standard IMU pre-integration relies on the exponential map, which does not admit an exact polynomial representation. Moreover, obtaining a QCQP formulation requires auxiliary lifting variables, for which the standard semidefinite programming (SDP) relaxation can be loose. We address these issues by deriving an IMU pre-integration factor based on the Cayley map and an explicit set of redundant constraints that tighten the resulting relaxation. To validate the proposed factor, we apply it to certifiable GNSS-IMU smoothing and evaluate it on synthetic and real-world data. The results show that the proposed formulation yields tight relaxations and solves the resulting estimation problems to verified global optimality.

Egocentric human demonstrations offer an accessible source of task experience, but differences in body scale and controller response, together with missing robot states, limit their value as humanoid training supervision. We present EgoAlign, a data-construction framework that converts these demonstrations into action and state supervision compatible with a general-purpose, continuous whole-body controller, without collecting physical-robot demonstrations. Using the target-robot model and simulator, EgoAlign guides demonstration collection through execution feedback. It preserves locomotion references for visually guided periodic stepping while adapting upper-body interaction geometry through scale alignment and controller-in-the-loop refinement. A final causal replay reconstructs the corresponding robot states and motion-token labels for training with the human observations. We assess the resulting supervision by fine-tuning a vision–language–action model solely on adapted human demonstrations and deploying it zero-shot on a physical humanoid. The resulting policies perform long-range object relocation, navigation to unseen goal positions, and independently evaluated foot interaction. Refinement improves simulated hand alignment and physical pickup success over kinematic alignment alone, while human collection reduces on-site acquisition time relative to teleoperation. https://lambdahumanoid.github.io/EgoAlign/

We reveal an unexpected connection between the parallel Kac's walk and the Kipnis-Marchioro-Presutti (KMP) process. The twirling channel induced by the parallel Kac's walk on the symmetric subspace is exactly encoded by a classical Markov chain on partitions, which lifts to a parallel KMP process on complete graphs. This correspondence reduces the analysis of the twirling channel to the mixing of the parallel KMP process. We prove that $O(\log d+\log(1/\varepsilon))$ repetitions suffice to approximate Haar twirling on the symmetric subspace of $(\mathbb C^d)^{\otimes t}$ to error $\varepsilon$, uniformly in the number of copies $t$. For the standard KMP process on general graphs, we prove a mixing-time analogue of Aldous's conjecture: at fixed accuracy, the mixing time of the $t$-particle process is at most a constant times the single-particle mixing time multiplied by the logarithm of the number of vertices, uniformly in $t$. As an application, we improve the total variation mixing-time bound for coordinate hit-and-run on the $n$-dimensional standard simplex from $\widetilde O(n^3)$ (Kook and Vempala, 2026) to $\widetilde O(n)$, while removing the dependence on the initial distribution. Our main technical contribution is conditional product structure for both parallel and standard KMP processes. Conditioned on suitable auxiliary randomness, the labeled particles evolve independently. Combining this structure with an exact coupling yields mixing bounds uniform in the number of particles for both unlabeled KMP models. These bounds are sharp up to logarithmic factors and imply rapid convergence of the parallel Kac twirling channel on the symmetric subspace.

Interactive agent benchmarks and multi-turn reinforcement learning increasingly place a second language model in the role of the user. This simulated user controls what information the agent receives and when, yet current benchmarks score only the agent and do not directly measure whether the user correctly executed its assigned role. We introduce UserProxyBench, an evaluation layer over the tau-bench family, and the User Fidelity Score (UFS), which measures adherence to the benchmark's private user instructions using task-grounded rubric criteria scored independently of agent success. Holding the agent fixed at GPT-5.5 and varying only the user proxy across 375 enterprise tasks changes mean task reward by 15.2 points, while 24.4% of successful episodes contain a user-specification violation. The dominant failure is premature disclosure: users provide information before it is requested. This behavior has little effect on task reward, yet among successful episodes it causes the agent to make 1.06 fewer tool calls on average, changing the interaction being evaluated while preserving the reward. Finally, across seven proxies we identify an empirical cost-fidelity frontier, enabling practitioners to select the least expensive simulator that satisfies a required fidelity level.

Emergence, features of a whole that no part shows alone, is central to complex systems but rarely measured so as to locate its source. Viewing a system structurally, as a network of interacting parts, we evaluate emergence as what observing the coupled whole reveals beyond, or erases from, the combined observations of the parts. For observations that build what they see route by route, we show that every emergent feature is produced by its own route across the interface between parts, a route that traces it to the components and interactions responsible. We show the number of all such boundary-crossing paths bounds the total emergence discrepancy, and equals the emergence of measures that count routes. The bound is computed locally around the interface, validated through network examples, and in a nervous system the routes show that interneurons relay most of the emergence. Our work associates emergence with path structure in the network system that can be anticipated, attributed, and engineered.

Understanding gender biases in large language models (LLMs) is increasingly important as these systems become embedded in decision-support tools with real consequences. Prior research has focused only on a small set of models, leaving open the extent to which gender biases are common and heterogeneous across LLMs. We address this gap across ten models released between April 2025 and June 2026, spanning nine vendors, using two paradigms: gender attribution to stereotyped phrases (Study 1) and moral judgment of abuse or torture against a woman or a man to prevent a catastrophic outcome (Study 2). In Study 1, two of ten models attributed masculine-stereotyped phrases to female writers more often than the reverse, while three models showed the opposite pattern. In Study 2, several models converged on a male-disadvantaging asymmetry that was directionally consistent with a documented human tendency to protect female targets from harm, though the specific conditions under which this asymmetry emerged varied by model; three other models, by contrast, showed no variation across conditions. These results indicate that gender-related biases are common in LLMs. Their direction and magnitude, however, are highly heterogeneous, to the point that some models behave in diametrically opposite ways to others. Bias auditing should therefore be treated as an ongoing, multi-vendor process, rather than a one-time assessment.

We give an explicit solution to the five expert prediction with expert advice partial differential equation (PDE) in the finite-time horizon setting. The solution formula establishes that the adversary's rank strategy $(1,0,1,0,0)$ is globally optimal, and the COMB strategy $(1,0,1,0,1)$ is optimal exactly on the set where $x_1=x_2$ and $x_3=x_4$. The formula is derived from the solution of the geometric-stopping problem given in our companion paper through the transform principle of Bayraktar, Ekren and Zhang, which links the two problems by a Laplace transform. Inverting the transform term by term expresses the solution through a series of Gaussian and complementary error function kernels. The optimality of $(1,0,1,0,0)$ is reduced to the signs of $41$ one-variable Gaussian series, which are certified with computer assistance by Poisson summation, first-mode domination and interval arithmetic on $1616$ rational cells. The proofs of our main theorems, certificates included, are also formalized in the Lean proof assistant.

We study fair and strategy-proof allocation of multiple divisible resources with Leontief utilities, motivated by cloud computing. The canonical mechanism, Dominant Resource Fairness (DRF), satisfies sharing incentive (SI), envy-freeness (EF), strategy-proofness (SP), and Pareto optimality (PO), but can be highly inefficient in terms of utilitarian social welfare. Under the classical approximation benchmark, no mechanism satisfying even one of SI, EF, and SP can improve on the trivial worst-case guarantee. We therefore adopt the recently introduced fair-ratio benchmark, which compares a mechanism only with the welfare-maximizing allocation that itself satisfies SI and EF. For two resources, we first introduce Adaptive-Speed Fairness (ASF), a unified parametric framework that captures previous mechanisms as special or boundary cases. Every ASF mechanism satisfies SI, EF, and PO, and we derive a general sufficient condition that guarantees SP. Optimizing within this framework yields a strategy-proof mechanism with asymptotic fair-ratio $2/(2\sqrt{2}-1)\approx 1.094$, substantially improving the previous best guarantee $3-\sqrt{3}\approx1.268$. We complement this upper bound with a lower bound of $1.0789$ for all ASF mechanisms. To overcome this limitation, we introduce Corrected Resource Balancing (CRB), which uses the full resource-load structure together with a one-agent incentive correction. CRB satisfies SI, EF, SP, and PO and achieves fair-ratio at most $1+1/n$. Together with a general $1+\Omega(1/n)$ lower bound, this establishes the optimal asymptotic order $1+\Theta(1/n)$ for two resources. Finally, for every $m\ge3$, any deterministic mechanism satisfying SI and SP has fair-ratio exactly $m$. The same lower bound holds for randomized mechanisms satisfying ex-post SI and truthfulness in expectation.

Generative AI (GenAI) tools entered higher education classrooms faster than the field was able to study their effects on learning. One concern is that GenAI may displace the critical thinking and AI literacy that students will need after graduation. This paper reports a Design-Based Research pilot of a GenAI-assisted critical thinking framework, in which ChatGPT was used as a thinking partner in an undergraduate research methods and statistics course during Spring 2025 (N = 14). The mixed-methods design combined pre- and post-intervention measures of statistical learning (AASCDM), AI literacy (MAILS), and critical thinking (WGCTA) with instructor field notes, student artifacts, and student-AI interaction logs. Pre-post tests showed gains on every AASCDM dimension and on eight of nine MAILS dimensions, while WGCTA percentiles did not change. Qualitative analysis identified four themes: the ways students positioned the LLM (as answer generator, validator, or co-thinker); the depth of student engagement (procedural vs. conceptual); occasional humanizing of the tool; and the role of curriculum design in shaping each of the prior three. Read together, the findings indicate that one semester of GenAI-assisted instruction can move domain learning and self-reported AI literacy but does not move standardized critical thinking, and that the modal student-LLM relationship is one of validation instead of dialogue. We end with design principles for the next iteration of the framework and implications for research on adaptive and personalized learning.

Autonomous vehicles interacting with passengers through natural language must reason beyond immediate commands. Passenger intent may span multiple stages of behavior, depend on future events, refer to surrounding agents or landmarks, and remain relevant as driving conditions evolve. Existing language-enabled driving datasets largely focus on short, localized interactions, leaving these longer-horizon forms of passenger intent comparatively underexplored. We introduce doPlan, to our knowledge the first publicly available, human-annotated real-world dataset designed to study passenger language as persistent task context. Built on nuPlan, doPlan contains 5,154 human-written passenger instructions spanning 169.1 hours of cumulative instruction-aligned context over 50.9 hours of unique driving, with annotation windows ranging from 30.0 to 508.8 s. The annotations capture immediate, deferred, event-conditioned, persistent, and multi-stage passenger intent. The dataset, annotation interface, and supporting resources are publicly available at https://github.com/Mi3-Lab/doPlan. We evaluate four language-conditioned driving models and find that sensitivity to passenger language does not reliably translate into behavior consistent with the requested direction. More broadly, among 2,161 examples with a matched future maneuver, the first associated maneuver occurs a median of 24.6 s after the evaluation point, and only 9.8% occur within the models' common 5 s prediction horizon. These findings highlight the need to connect persistent passenger intent with successive planning decisions. doPlan provides a setting for studying how unresolved goals can be retained, grounded in evolving scenes, and tracked across multiple stages, including how a planner determines when a future goal becomes relevant to the current plan.

On-policy distillation (OPD) trains a student on its own generated responses using dense, token-level supervision from a stronger teacher. Vanilla OPD treats all teacher signals equally, assuming that the teacher's supervision is equally important for every token. However, teacher signals at different tokens may have very different effects on the student's performance: some correct important reasoning errors, while others have little effect on the final answer. Motivated by this observation, we introduce Dr. OPD (OPD Done Right), which defines the optimal weighted OPD to maximize the student's performance. We formulate Dr. OPD as a bilevel optimization problem in which the student learns from weighted teacher supervision, while the weights are selected to maximize the expected reward of the resulting student. To solve Dr. OPD, we develop an efficient iterative solver that updates the token weights and student policy alternatively. At each round, it updates weights in closed form and then takes one gradient step on the resulting weighted OPD objective. Under regularity conditions, we show that this weighted update achieves a higher expected reward than a vanilla OPD update. Empirically, across strong-to-weak and same-size distillation on math and code, Dr. OPD consistently outperforms all evaluated baselines. In particular, in the strong-to-weak distillation setting, Dr. OPD improves average math performance by $9.7$ points over vanilla OPD, and enables the smaller student to surpass its larger teacher.

An agent skill is a reusable, actionable natural-language artifact that guides an agent to perform a task effectively under a given harness. Recent studies have explored the optimization of agent skills, contributing to a growing collection of publicly available skills spanning diverse tasks, domains, and harnesses. Despite millions of publicly shared skills, existing skill optimization methods largely overlook this accumulated knowledge, instead relying solely on expensive agent rollouts to iteratively refine skills for a target task. To address this, we propose Retrieval-Augmented Skill Optimization (RASO), a framework that leverages an external skill corpus as prior knowledge throughout skill optimization. RASO retrieves relevant knowledge from existing skills and adapts it to the target task and harness via Cross-Harness Adaptation, accounting for mismatches in both domain and harness. RASO comprises two complementary stages: Retrieval-Augmented Skill Initialization (RASI) constructs a knowledge-grounded initial skill without requiring agent rollouts, while Retrieval-Augmented Skill Update (RASU) iteratively refines the skill by retrieving external knowledge guided by execution feedback. Across four agent benchmarks and two models, extensive experiments show that RASO consistently outperforms baselines without retrieval-augmented skill initialization and updating.

Physics-Informed Neural Networks (PINNs) embed governing equations into deep learning, but enforce them only through loss residuals, leaving highly oscillatory wave behavior to be discovered by optimization. As a result, methods that achieve relative $L_2$ errors below $10^{-3}$ on standard manufactured Helmholtz benchmarks can fail on practical radiation problems involving singular excitations, absorbing boundaries, and wave fields spanning tens of wavelengths. Architectural physics embedding addresses this limitation by factorizing the field into analytically derived oscillatory kernels and learnable envelopes. However, the kernel dictionary must be manually constructed and scales with the number of elementary units, growing exponentially with the depth of hierarchically structured systems such as antenna arrays and metasurfaces. We propose PE-EK-PINN (Physics Embedded with Evolving Kernels), which treats physics kernels as reusable learned representations rather than fixed analytical inputs. A converged subsystem field is frozen and promoted to an evolved kernel, whose transformed copies are reused to represent higher-level configurations without deriving new governing equations. The resulting hierarchy makes the peak number of active kernels independent of system size and reduces cumulative training cost from $O(N)$ to $O(\log N)$. Experiments on dipole arrays, composite line-source geometries, and cross arrays demonstrate the dramatic training cost reduction, while achieving a reduced or comparable relative $L_2$ error. One notable example is PE-EK-PINN solves a $256$-dipole array more than 30 times faster than direct PE-PINN.

This report audits evaluation of a long-term-memory retrieval chain on the 500 LongMemEval-S development questions. Its strongest historical reader lane scores 479 and 475 under an adapted GPT-4o rubric; re-judging the same pass-1 answers changes three labels and yields 478. Fixed-answer knowledge-update re-scoring gives 70/72 under the upstream template and 69/72 under the modified template. Reader lanes span 93 to 479 on fixed packets; paired tests between the two strongest historical lanes establish neither superiority nor equivalence. A different-family reader, configured without client tools or operator files, scores 474, 1.0 percentage point below the headline pass (paired 95% interval [-3.0,+1.0]). Live reader request bodies were not retained. With the same requested reader label, route and judge snapshot, the full package scores 474 versus 454 for baseline sessions, a difference of +4.0 percentage points [95% interval +2.2,+6.0]. Eighteen of the 23 gains, and no losses, occur where baseline packets lacked listed evidence; this post-hoc split does not identify a component effect. In recovered LoCoMo data, token-F1 gains do not survive answer-line extraction. A negative control rejects a verifier that repairs three wrong drafts but breaks eleven correct ones. All questions were used to develop the components; no untouched holdout was evaluated. These findings do not establish a new leaderboard leader or transferable memory advantage. The A/D comparison has one pass per arm, including six reused identical-prompt outcomes, with no pinned reader snapshot; B/C and repeats remain unrun. Original headline requests cannot be reconstructed and stages 1–4 remain closed. Released artifacts support packet inspection and saved-verdict recounting and re-scoring; they do not reconstruct the method.

The rapid growth of artificial intelligence (AI) data centers has introduced new challenges to power system operation. As their power demand becomes larger and more variable, quantitatively characterizing their demand flexibility is increasingly important for effective power system coordination. However, heterogeneous workload characteristics and resource requirements make this flexibility difficult to characterize directly. This paper proposes a framework for assessing the grid-compatible demand flexibility of AI data centers via batch workload temporal shifting. An averaging-based resource usage processing method is developed to map fine-resolution CPU, GPU and memory usage into unified time intervals compatible with power system operation. A workload temporal scheduling model is then formulated to shift batch workloads while preserving execution continuity, delay constraints, and server resource capacities, and is coupled with a utilization-dependent server power model to translate workload scheduling decisions into server power demand. Two complementary flexibility metrics are evaluated: short-term peak demand shaving and the maximum duration of sustained power reduction. Numerical results based on real GPU cluster traces demonstrate that workload temporal shifting can provide quantifiable and grid-compatible demand flexibility for AI data centers with limited disruption to computing workloads.

We present RGOR (Reference-Grounded Oral Refinement), an audio-driven lip-sync framework that renders the mouth of the specific person being dubbed rather than a generic one. Existing lip-sync systems follow the audio closely and keep the face recognizable, yet the mouth they render is an average mouth: the shape and texture of the lips, the arrangement of the teeth, and how much of them shows as the mouth opens are not that person's. The problem persists because nothing in current training or evaluation asks for the person's own mouth: perceptual losses accept any plausible mouth, face identity is carried mostly by the skin around it, and the released inference code of inpainting systems uses the unmasked target frame as the reference, which hides the gap. To address this, RGOR conditions every generated frame on frames from separate enrollment recordings of the same person and on HD patches of the mouth that bypass the VAE, and trains the generator against a paired judge that compares each rendered mouth with the person's reference and learns to reject a realistic mouth of someone else. We further build an evaluation protocol and use it to compare open-source and commercial lip-sync systems on held-out identities. Experiments show that RGOR achieves the best or second-best result on most metrics, and preserves the person's own lip and dental detail while keeping synchronization and the rest of the face intact.

Group relative policy optimization (GRPO) learns only from prompts whose sampled responses disagree: a group that is entirely correct or entirely incorrect has zero reward variance, contributes no gradient, and still consumes its rollouts. Prompt-selection methods reduce this waste by steering sampling toward intermediate pass rates, but they choose the target, its width, and the uncertainty model heuristically, in raw pass-rate or logit coordinates. We show that GRPO comes with a natural coordinate for pass rates: the arc length $\psi=\arcsin\sqrt{p}$ on the Bernoulli Fisher–Rao manifold. In arc length, the expected GRPO update is uniform up to two boundary ramps; the probability of a zero-variance group is bounded by two Gaussian boundary layers of width $1/\sqrt{2G}$; pass-rate evidence has constant noise; and the gradients of the pass@$k$ and pass$^k$ objectives are Gaussians whose center and width follow from $k$ in closed form. A prompt curriculum for GRPO is therefore a Gaussian in arc length, and choosing its center amounts to choosing the objective. We turn this observation into ARCUS, a drop-in sampler that tracks every prompt with a Kalman filter in arc length, scores prompts by an objective-matched Gaussian kernel times the predicted probability of an informative group, keeps only informative groups for the unchanged GRPO update, and paces the target toward the hardest objective whose predicted yield stays within a small slack of the best. Across six mathematical reasoning benchmarks and three backbones, ARCUS improves the average accuracy of GRPO by 2.8–2.9 points and that of dynamic sampling by 1.1–1.2 points, while generating 48–57% fewer rollouts than dynamic sampling.

Constrained Reinforcement Learning has recently gained increasing attention in the field of Safe Autonomous Driving, where the general mechanism is to maximize the expected reward while keeping the overall action risk bounded. In this way, the safety issues arising in AD can be mitigated through constrained actions. However, existing Constrained RL methods still lack dynamics on the imposed constraints. For instance, the action cost adopted by the existing Primal-Dual/soft-constrained methods is often defined as static state-to-cost mapping, and the safe-action projection in hard-constrained methods relies on the static projection with the fixed feasible region boundary estimated from offline demonstrations. The above drawback tightly couples the imposed constraints to the training scenarios, leaving the AD policy hard to handle different interaction scenarios, due to the improper state-level action-cost and the static projection boundary. Consequently, in this paper, we propose Brain-SAD, a brain-inspired safe autonomous driving control framework with dynamic fear-oriented constraints. By perceiving the current vehicle-interaction scene, Brain-SAD generates dynamic fear signal as fear reaction to online decide long-term policy for regular interaction or short-term policy for urgent-collision defense. In such two policy, the above fear-reaction will be constructed as the dynamic fear constraints, respectively reflecting the overall fear cost directly coupled with action-impact, and the dynamic fear boundary of the feasible region derived from different risky neighbors, both of which will in turn serve for the online policy optimization. Experimental results show that Brain-SAD outperforms existing methods, achieving higher success rate in shorter task-completion and collision-recovery time, and exhibits stronger reliability across continuous intersections of fluctuating complexity.

JavaScript and TypeScript are widely used in modern web development, making their security critical; however, automated vulnerability detection is often constrained by the availability of high-quality training data. Here we present JsVul, a dataset curated from seven major sources. Unlike generic multi-language datasets that may retain noise – such as minified code and cosmetic edits – JsVul utilizes a language-specific pipeline. We collected pre-fix and post-fix versions of files around security fixes and, by filtering irrelevant artifacts and applying automated syntax normalization, isolated security-related changes. We ensured data integrity through multi-stage deduplication and heuristic-based labeling. Provided in a time-ordered JSONL format, JsVul supports robust model training in the JavaScript and TypeScript ecosystem and demonstrates the importance of language-aware preprocessing in building vulnerability datasets.

Modern machine learning systems are trained on mixtures of data from different domains, and choosing the right mixture can substantially improve downstream performance. Despite an extensive literature on data mixing and reweighting, existing work is largely empirical and it remains unclear when auxiliary data genuinely improves scaling laws rather than merely providing more samples. To gain insight into this question, we study a high-dimensional mixed-data regression model with a shared regression function, heterogeneous covariances and noise levels, and dataset sizes that may grow at different rates. We establish the minimax risk under an ellipsoidal parameter constraint for the general covariance structure and derive deterministic equivalents for the test error of ridge regression under commutative covariances. We then specialize to a target domain and an auxiliary domain with aligned power-law covariance spectra, where the theory yields explicit scaling laws in terms of spectral decay, target regularity, and the relative growth of the two datasets. These laws identify regimes in which combining data mixtures provably yields a faster scaling rate than using either dataset alone. In particular, improving the scaling law requires a specific interplay between spectra and relative sample sizes of the domains. Our numerical experiments on language models exhibit the same qualitative phenomenon: appropriate data mixtures yield a faster decrease in target-domain test loss than training on either domain alone.

Modern computer vision models achieve high accuracy when trained on large-scale annotated datasets. In critical domains such as construction safety monitoring, data collection is costly, hazardous, and ethically constrained. This paper presents a systematic study comparing two complementary data generation paradigms, (1) Unity Simulation-based rendering and (2) Controllable Diffusion-based generation (CIA), for object detection under real data-scarce conditions. A unified experimental framework enables controlled dataset mixing across real, simulated, and generative sources, while maintaining identical model and training settings. Quantitative evaluation using Precision, Recall, mAP, and custom $\Delta$-metrics, reveals that neither simulation nor generative augmentation alone achieves optimal transferability. Unity-only training yields an mAP@0.5 drop of $-50\%$ relative to real data, while CIA-only training shows a milder $-16.5\%$ degradation. Hybrid compositions significantly improve performance, with the 90% real + 10% Unity configuration achieving the best overall mAP@0.5 of $62.68\%$ ($+7.64\%$ over baseline), and the 90% real + 10% CIA configuration maximizing precision at $74.45\%$. Results demonstrate that limited synthetic inclusion enhances generalization, while excessive substitution induces domain drift.

Computer use agents (CUAs) have demonstrated strong capabilities in completing digital tasks. However, existing CUAs either rely solely on graphical user interface (GUI) interactions, which are often inefficient and error prone, or augment GUI interactions with application specific APIs or tools, which require substantial engineering effort and are difficult to scale across applications. We argue that the next generation of CUAs should combine GUI interactions with the command line interface (CLI), leveraging the generality of the GUI and the efficiency of shell commands. A critical challenge, however, is that current models do not know when or how to use the CLI during task execution. To address this challenge, we develop a data construction pipeline that produces three types of trajectories: GUI only, CLI only, and interleaved GUI and CLI trajectories. This pipeline results in HybridCUA-8K, containing 5K hybrid trajectories and 3K verified RLVR tasks. Building on these data, we propose a training framework with two stages: supervised fine tuning on the constructed trajectories, followed by reinforcement learning with our CLI aware rewards that encourages agents to use the CLI selectively and reliably. Experiments show that HybridCUA-9B achieves 53.6% accuracy on OSWorld, improving over the base model by 14.8 percentage points, and improves performance on WindowsAgentArena by 4.0 percentage points. These results demonstrate the effectiveness and cross platform generalizability of the hybrid GUI and CLI paradigm for computer use agents.

Running a language model locally offers advantages in privacy, latency, and cost, but local hardware fits only small models, which are less capable than frontier models. The usual remedy for a hard query, escalating it to a cloud model, gives up the privacy and cost advantages of running locally. A deployment that stays local faces two decisions for hard queries instead. First, it can spend more computation on a query, e.g., reasoning before answering, which raises accuracy at a cost in latency, so it must decide which queries are worth the extra computation (efficiency). Second, some queries are beyond the local model, and delivering a wrong answer is worse than deferring the query to a human in the loop, so it must decide which answers are safe to deliver (safety). We show that both decisions can be made from the model's own hidden states. The prefill state, computed before any token is generated, predicts whether the model will answer correctly, and the answer state, at the end of the generated answer, predicts whether that answer is correct. HARISSA fine-tunes the model so that both states predict correctness, then makes both decisions with one policy that cascades through the ways of answering from cheapest to most expensive, skipping a way the prefill state predicts will fail and deferring the query when the answer it stops with is predicted wrong. On a device running a single model, HARISSA is within one accuracy point of chain-of-thought at 2.7 times lower latency. On a server holding four sizes of one model, HARISSA is more accurate than the FrugalGPT and Self-REF cascades at the same latency, and at the same deferral rate the answer state leaves fewer wrong answers than the standard confidence signals in five of six task and setting pairs.

Large language models (LLMs) are increasingly proposed as decision assistants who must reason probabilistically from available evidence under explicit decision costs. We propose a decision-theoretic framework that decomposes LLMs' decision loss into two components: forming accurate beliefs from provided evidence and translating those beliefs into actions that optimize a provided utility function. Using a synthetic benchmark with known ground truth, we apply the decomposition to characterize probabilistic reasoning in frontier and open-sourced models. We further evaluate whether RL interventions targeting beliefs, decisions, or both improve these components across three domains, whether improvements transfer across components and elicitation formats, and whether decision performance can improve without improvement in belief formation. We find that targeting one component of probabilistic reasoning redistributes decision loss, improving the target without necessarily transferring to others, and that jointly targeting belief formation and decision-making improves both but hinges on matched formats between training and evaluation.

Time series anomaly detection (TSAD) plays a crucial role in healthcare, finance, industrial monitoring, and other sectors. Within and between these settings, anomalies span vastly different temporal scales, from sub-second point spikes to multi-hour drift patterns. However, most existing TSAD methods commit to a single temporal granularity, and multi-scale designs either analyze different scales in isolation or are constrained to a predefined coarse-to-fine hierarchy, both failing to sufficiently capture multi-scale interactions. To resolve this limitation, we propose Multi-Scale Autoencoder with Cross-Scale Attention for TSAD (MSCAD), a simple yet powerful semi-supervised TSAD framework founded on parallel autoencoder branches corresponding to different patch sizes. A stack of symmetric bidirectional cross-scale attention blocks enables every pair of scales to exchange information before reconstruction without allowing any single scale to be privileged. On the comprehensive TSB-AD benchmark (40 datasets, 530 series), MSCAD achieves large performance gains against 50 baselines across multiple metrics, with VUS-PR of 0.57(+9.6%) on the univariate split and 0.47(+9.3%) on the multivariate split compared to the state-of-the-art.

We study how well simple mechanisms approximate gains from trade (GFT) in two-sided matching markets with independent buyer values and seller costs, where feasible outcomes form an arbitrary downward-closed family of matchings. This model includes bilateral trade and double auctions as special cases. We focus on the Generalized Random-Offerer (GRO) mechanism, which is an equal mixture of the Generalized Sellers-Offering Mechanism (GSOM) and the Generalized Buyers-Offering Mechanism (GBOM). We determine GRO's exact worst-case approximation ratio with respect to first-best GFT, showing that it is $e$. This improves the previous $3.15$ approximation guarantee (Babaioff et al. STOC 2026) for the same mechanism. In bilateral trade, the result implies a $1/e$ guarantee for the random-offerer mechanism, improving the previous $1/\pi$ bound (Jo 2026). Our analysis uses a two-dimensional inequality in quantile space to compare first-best GFT with optimal one-sided auction profits. For each potential trade, we identify the region of buyer values and seller costs for which the first-best allocation selects that trade, and then randomly shrink this region in quantile space, yielding posted-price rules whose expected profits can be evaluated exactly. We establish tightness of the $e$ approximation by constructing markets with regular type distributions, pairwise disjoint trading edges, and a single knapsack constraint. On these instances, GRO's expected GFT approaches a $1/e$ fraction of first-best GFT, therefore ruling out any better guarantee even under these restrictions.