Science & Mathematics

Neuro-spectral architectures (NeuSA) are a class of physics-informed neural networks (PINNs) that have proven effective for solving time-dependent PDEs in the frequency domain. However, training and inference often rely on costly numerical integration of the learned vector field, typically using sequential Runge-Kutta time stepping, which limits temporal parallelism. We introduce PaPi, an integration scheme inspired by weakly nonlinear flows that enables time-Parallel inference through exponential integration and a Duhamel-Picard scheme. We show that the Duhamel-Picard iteration converges exponentially for all flows, and provide explicit error bounds that depend on how close the flow is to linear. Experiments show that PaPi accelerates training and inference by up to two orders of magnitude, achieving training times of under one second for low-dimensional PDE learning problems, while maintaining competitive predictive accuracy.

Three-dimensional time-of-flight secondary ion mass spectrometry (3D ToF-SIMS) builds chemical volumes by stacking images acquired between sputter cycles. Standard reconstruction assumes a flat starting surface and a single sputter rate, so samples with topography, or with layers that sputter at different rates, yield distorted interface positions and depth scales. This article is a practical guide to correcting these artifacts. We group the available corrections by the information they need: the SIMS data alone, atomic force microscopy (AFM) topography before and after the profile, or AFM topography also at intermediate stages. All three are applied to a model Pt/TaOx/Ta/Pt/Ti crossbar memristor device measured with in situ SIMS/AFM at ten stages, together with the data preparation they share. Each correction method is evaluated using the known device geometry and the measured surface evolution. The data-only correction requires no additional instrumentation and restores relative topography, but its depth scale depends on nominal layer thicknesses and remains unconstrained below the last calibrated interface. AFM before and after the profile fixes the start and end surfaces but assumes uniform sputtering in between, which fails next to the device, where the surroundings change material during the profile. Multi-stage AFM correction follows these changes and reproduces the structure of the device, at the cost of instrument time and repeated imaging of a partially sputtered surface. A decision flowchart and a checklist help readers choose, validate and report a correction.

Reliable multi-view forecasting requires videos from different cameras to describe the same evolving 3D scene. To this end, we introduce Video-Point Model (VPM), a world model that conditions on multi-view video histories and a language instruction to jointly forecast future videos from arbitrary viewpoints and the 3D trajectories of persistent scene points. We extend a pretrained world model with camera-conditioned multi-view processing that integrates complementary observations across cameras, and add a trajectory head that predicts how physical points move through time. Each trajectory tracks a single point across all views, and supervising these trajectories alongside video generation therefore anchors appearance changes in every camera to a shared description of 3D motion. To support training and evaluation, we build VPM-Data, a simulation corpus pairing synchronized multi-view videos with persistent point trajectories, and VP-WorldBench, a forecasting benchmark spanning simulated and real environments. We further extend VPM to robot control by adding robot actions as a token modality in the generator. Action tokens share the point-supervised backbone and are denoised jointly with future videos, thereby grounding the policy in the same 3D-aware representation. On VP-WorldBench and existing 4D generation benchmarks, point supervision improves both video fidelity and 3D motion accuracy over a video-only counterpart, and in simulated and real-robot experiments, it also raises the manipulation success of the action-enabled model.

RTL repair benchmarks should measure whether a model can fix broken RTL given a specification. To this end, we construct and hash-bind RTLRepairDataset, 5,784 (broken, fixed) training pairs; RTLRepairBench, 459 held-out cases with 374 syntactic and 85 behavioral mutations; and an evaluation harness supporting simulation and formal equivalence. We then use this infrastructure to identify two measurement confounds. First, simulation and formal scoring give opposite signs for the apparent effect of providing a specification in one evaluated cohort; reset-aware rescoring attributes almost all of the simulation-formal disagreement to reset and first-cycle comparison semantics. Second, on 274 mined model failures, the specification-only minus specification-conditioned recovery point estimate ranges from -8.5 to +18.3 percentage points across six models. Thus, a conventional repair score need not isolate whether the model used the faulty RTL. We translate these findings into evaluation controls: report initialization, reset, first-cycle, and unknown-value semantics; disclose oracle coverage and undecided cases; and include a specification-only baseline. Together, the resources and audit provide both a benchmark for behavioral repair and a protocol for diagnosing what contributes to its measured score. We release the benchmark, the training set, construction code, sanitized aggregate evidence, and reproduction materials at https://github.com/Evaaaz/RTLRepairBench.

We study the minimum persistent memory required to distribute a graph state through a quantum switch. Specifically, consider a central switch that assists $n$ remote users in distributing a graph state, using independently generated user-switch Bell pairs. Since Bell-pair generation is probabilistic, different links generally succeed at different times. When a Bell-pair is generated, the qubits may need to be stored in memory while waiting for other links to succeed. Recently, new protocols have been proposed to reduce this storage overhead by processing and releasing some switch qubits before all Bell links are available. However, these protocols do not characterize the minimum persistent switch memory required at each stage. Moreover, the local-cover method in the state-of-the-art Piecemaker protocol is computationally expensive for general target graphs. In this work, we characterize the minimum persistent switch memory required after any subset of users' Bell links are generated and processed. We then develop an online stabilizer-transition protocol that attains this limit, which can be implemented in polynomial time in the number of qubits $n$. Remarkably, this optimal protocol can provide substantial memory reductions over existing heuristic approaches. In particular, for the path graph $P_n$ over $n$ users, Piecemaker protocol requires at least $\lfloor n/2\rfloor$ qubits of peak persistent memory for \textit{every Bell-links generation order}, whereas under the order $(1,2,\ldots,n)$, our protocol requires only one persistent switch qubit. Finally, the same stabilizer-transition framework also provides a direct forward construction for sequential graph-state generation using the minimum number of quantum emitters, equal to the maximum of the height function.

Neural implicit representations provide memory-efficient signed distance field (SDF) parameterizations for large-scale LiDAR scenes, but the geometry recovered from sparse and often noisy inputs carries jitter and artifacts. Classical geometric optimization techniques resolve such artifacts with surface energy functionals, whose minimizers correspond to smooth surfaces. Motivated by this principle, we formulate neural LiDAR mapping as the optimization of energy functionals over the parameter space of neural implicit representations. Specifically, we regularize the signed distance field with an energy functional defined on the shape operator of its level set, which represents the curvature of the surface, thus encouraging smoothness along the surface. This adapts classical geometric priors to the scale of million-point LiDAR maps. Experiments on large-scale datasets show reconstructions with improved smoothness at preserved geometric fidelity, and no additional inference-time cost. The code is publicly available at \href{https://github.com/HrishikeshVish/nieo}{https://github.com/HrishikeshVish/nieo}

Computer-use agents are limited by the interfaces humans give them. An agent may type a spreadsheet cell by cell when a single line of code would do, or print all two thousand lines of a file through a shell to find one function. Both the interfaces and the knowledge of when to use each are built by hand, which cannot keep pace with new applications, workflows, and models. We argue that interface design should itself be an agent capability. We introduce Agent S4, an autonomous computer-use scientist that builds Evolving Agent Interfaces from task experience. A meta-agent examines task agents' trajectories on synthetic training tasks, without success labels, and constructs and revises interfaces and usage knowledge in a repository. Population-based evolution then selects which repositories to develop further by their performance on validation tasks the meta-agent never sees, favoring changes that generalize beyond the seen tasks. On OSWorld V2, Agent S4, with Qwen3.8-27B and the evolved repository of interfaces, sets a new state of the art in open-source computer-use agents, reaching 21.3% binary score, 3.8 times the same model with fixed interfaces. This is above both GPT-5.5 (13.0%) and Claude Opus 4.8 (20.6%), at about 4.3% and 1.5% of their respective costs. On ScienceBoard, the evolved interfaces improve the success rate from 41.4% to 58.6% with the same model. These results suggest that useful interfaces can be evolved from experience, making autonomous interface design a source of improvement complementary to advances in models and harnesses. We open-source our code at https://github.com/simular-ai/agent-s.

TS2Vec is one of the strongest self-supervised methods for time series classification, introduced in 2022. It is designed to be trained anew on every dataset: self-supervised pre-training followed by linear probing. Since then, the field has shifted towards foundation models that reuse one encoder across datasets. Yet TS2Vec offers a compact convolutional architecture that operates directly on temporal samples and accepts variable-length inputs without resampling. We ask whether these properties can be retained when the encoder is pre-trained only once, and show that they can: TS2Vec can essentially be turned into a foundation model. Three changes suffice: pre-train on a synthetic corpus, feed the encoder a raw amplitude view and an instance-normalized shape view of the input through two small stems, and distill one specialist encoder per view into a single encoder. The result, TS2Vec-FM, is a 1.6M-parameter dilated convolutional network that, kept frozen, reaches 83.3% on UCR-128 and 75.3% on UEA-27 under linear probing, within 0.3 pp of MantisV2, the current state of the art, and above TS2Vec trained separately on each target dataset. What we find most interesting is how different TS2Vec-FM is from other foundation models in design: near-parity is reached without attention, without a patch tokenizer, and without intermediate-layer extraction, using an "old-school" convolutional encoder and a two-layer MLP stem.

Current evaluation of LLM-generated debate largely focuses on argumentative quality, leaving the value structure of generated arguments underexamined. We introduce a diagnostic metric framework for comparing how values are expressed and organized in human-reference and AI-generated debates. The framework separates value expression, the range of recognizable values a debate invokes, from value organization, how those values are distributed and sustained across turns. Grounded in Schwartz's Basic-10 value theory, it represents debates as multi-level value distributions over argumentative units, utterances, stances, and full debates, enabling metrics for value coverage, alignment, diffusion, and cross-turn dynamics. Across multiple generation systems, AI-generated debates are broadly value-expressive, often activating as many or more value dimensions than matched human references, but they organize these values differently. Their utterances are more locally value-diffuse, and source-blind judges distinguish value clarity from value focus. These findings show that debate evaluation should consider not only value presence, but also value organization. Code and supplementary materials are available at https://github.com/Maoyuan-li/Value-Expression-and-Organization.

LLM-based methods for node classification on text-attributed graphs (TAGs) routinely report state-of-the-art accuracy. We ask how these results compare with inexpensive baselines run on the same splits. On the splits of three recent studies, we tune four baseline representatives with validation-only selection and compare them with the highest mean that each paper reports for an LLM-based method in the matched setting. Correct & Smooth on TF-IDF features trails this reference by a median of 2.3 percentage points, and an MLP on frozen e5-large-v2 embeddings by 2.8 points. Correct & Smooth is within 3 points of the reference or better on 15 of 24 tasks, but the reference is significantly higher on 12 of 22 testable tasks, so several reported gains remain substantial. Reading released code next to the papers, we also document evaluation risks: selection of hyperparameters or checkpoints on the test set, labels of test-pool nodes used in training, baseline numbers reused without disclosure, and a text-label misalignment in a released Cora loader. Reports of the same method on the same split also differ by up to 12.5 points across papers whose configurations are not documented to match.

Mathematical reasoning remains underexplored for low-resource languages, including Dialectal Arabic. To bridge this gap, we introduce the DAR-MGSM (Dialectal Arabic Multilingual Grade School Math) Benchmark, the first manually curated benchmark for evaluating mathematical reasoning in Dialectal Arabic. DAR-MGSM consists of math word problems (MWPs) requiring multi-step reasoning and covers five major Arabic dialects alongside Modern Standard Arabic (MSA). The benchmark comprises two parallel versions: translated-only and translated & culturally localized. The latter version adapts entities and references to regional varieties. We evaluate eleven Arabic-centric and multilingual large language models on DAR-MGSM. The results show that the performance generally declines on the localized version, highlighting the challenges introduced by culturally grounded adaptations. However, the magnitude of this effect varies across models, with Arabic-centric models exhibiting different levels of sensitivity to localization. A mechanistic interpretability analysis shows that translated and localized variants of the same MWPs often induce distinct internal reasoning trajectories, even when aggregate accuracy remains similar. Overall, our results show that mathematical reasoning performance depends on both linguistic variety and cultural context.

Interactive webpages underpin modern applications, but reproducing both their visual appearance and multi-step behavior remains challenging for multimodal large language models (MLLMs). Scalable training data must capture user actions, state transitions, and expected outcomes, while real-world webpage code is often entangled with irrelevant scripts and external dependencies. Moreover, existing GUI-agent-based verification provides slow and potentially unstable functional feedback. To address these challenges, we propose InterWebCoder, an efficient framework for interactive webpage code generation. InterWebCoder introduces a Test-Specification-Guided and Execution-Verified Data Synthesis pipeline that records real-world interaction trajectories and synthesizes self-contained webpage code with executable test scripts. A failureattributed repair mechanism uses runtime evidence to locate and iteratively fix errors in either the webpage code or test script. We then combine test-ID-weighted supervised fine-tuning with group-relative policy optimization, directly optimizing functional correctness and visual fidelity using test pass rates and state-level visual similarity. This replaces costly GUI-agent verification with fast, reproducible executable feedback. Our pipeline produces 1,736 verified samples from 2,058 tasks, achieving an 84.35% synthesis success rate. On IWR-Bench and Vision2Web, InterWebCoder achieves final scores of 48.54 and 48.42 with a 9B backbone, outperforming 30B-level models and Gemini-3.7-Flash while remaining competitive with Claude-Sonnet-5. It reduces reward-computation latency by over 97%, eliminates external MLLM verification costs, and accelerates end-to-end training by up to 1.4x.

Modern distributed microservice deployments generate massive volumes of heterogeneous, multi-format observability logs across physically and logically partitioned infrastructure. Isolating root causes during cascading failures remains an arduous task for Site Reliability Engineers (SREs), bottlenecked by manual keyword filtering and disjointed temporal correlation across monitoring dashboards. We introduce Semantic Observability, an autonomous multi-agent framework that unifies disparate log streams into a dynamic, queryable Property Graph Database. By modeling physical infrastructure topologies, logical software dependencies, and chronological event sequences as attributed nodes and directed relations, it bridges the semantic gap inherent in flat, unstructured log repositories. Building on this unified substrate, we deploy a stateful multi-agent system orchestrated through Directed Cyclic Graphs (DCGs). The system autonomously navigates the topology, applying causal graph algorithms and temporal back-traversal to isolate the precise blast radius of anomalies. The pruned subgraph is then interpreted by a Large Language Model (LLM) reasoning engine to generate contextualized, code-level Root Cause Analyses (RCA) and actionable remediations. This paper establishes the foundational system architecture, formal relational ontology, and multi-agent control loops of Semantic Observability. We show conceptually how the framework eliminates ambient telemetry noise, reduces Mean Time to Resolution (MTTR), compresses LLM token consumption, and prevents structural hallucinations. We also specify an open-source validation framework on standard benchmarks (LogHub and Train-Ticket), laying the groundwork for a forthcoming companion paper on exhaustive quantitative evaluation and production benchmarking.

Software engineering agents solve repository-level tasks through iterative tool use, but their continuously growing interaction traces incur substantial token consumption and inference overhead. Existing compression approaches often require model training, auxiliary LLM calls, or assumptions about specific content such as source code. We present TraceLite, a simple yet effective, training-free framework for compressing heterogeneous SWE-agent observations. Our preliminary study shows that optical rendering preserves substantial textual and semantic information under moderate compression, while agent trajectories exhibit both cross-step and phase-level redundancy. Motivated by these findings, TraceLite preserves recent observations as high-fidelity text, renders historical observations as compact images, removes overlapping content, and discards phase-expired observation payloads after code editing. It requires neither offline training nor auxiliary LLM inference. We evaluate TraceLite on SWE-bench Verified and DeepSWE using GPT-5.6-Luna and Gemini-3.8Flash as backbone models. On SWE-bench Verified, TraceLite achieves 78.25% resolve rate, outperforming the evaluated compression baselines while reducing average input-token consumption and API cost by 33.05% and 21.30%, respectively. Moreover, it introduces only 0.438 seconds of compression latency per observation. These results demonstrate that TraceLite improves SWE-agent efficiency with low processing overhead and without compromising issue-resolution capability. Our code is available at https://github.com/xjywhu/TraceLite.

Evaluating and post-training robot policies through physical interaction is costly and difficult to scale, motivating action-conditioned world models as interactive simulators. However, visually plausible predictions can misrepresent contact dynamics, providing unreliable feedback for policy evaluation and learning. We introduce PhysCanvas, a state-consistent visuo-tactile world model for contact-rich robot manipulation. PhysCanvas unifies multi-view RGB, tactile images, marker displacement, task state, and actions within a unified vision space. All modalities share a video VAE and pretrained generative backbone, allowing optical touch to reuse visual and temporal priors. Motion images derived through forward kinematics and camera projection provide spatially aligned action conditioning, while an initial RGB anchor and updated reference observations guide sequential generation. Our consistency-aware post-training uses the frozen Scene Reward Model, Modality Reward Model, and State Reward Model to supervise scene preservation, visual-tactile compatibility, and agreement with real task outcomes. The resulting world model supports closed-loop policy rollouts for evaluation and provides imagined interactions and rewards for GRPO policy updates. Across four real-world manipulation tasks, when policies are rolled out in the world model, State Reward Model supervision increases the arithmetic mean of task-wise Pearson correlations between world-model and real-world success rates from 0.602 to 0.885. Policy post-training in the reward-guided world model further increases $π_{0.5}$'s mean real-world success rate from a 59.1% SFT baseline to 70.8%. These results demonstrate the value of unified multimodal modeling and consistency-aware post-training for robot policy evaluation and learning. Project page: https://zebinx.github.io/PhysCanvas_HomePage/.