arXiv:2607.05454v2 Announce Type: replace
Abstract: The journal is a seventeenth-century technology asked to do four modern jobs at once: disseminate results, certify their quality, allocate scholarly attention, and confer career credit. It does none of them well. Pre-publication peer review is slow, only weakly reliable, demonstrably biased toward established authors and institutions, and expensive, while the reviewing effort it consumes is spent largely on work that will never matter. We argue that these are not defects to be patched but consequences of bundling dissemination and certification into a single gated act, and we propose unbundling them. Under the proposal, authors deposit papers in an open archive; certification happens \emph{after} deposit, continuously, through attributed and up- or down-voted public commentary to which authors may reply; and papers are version-controlled objects that any qualified reader may \emph{fork}, so that the lineage of an idea -- and hence the credit for it -- is recorded automatically. None of the individual components is speculative: each already exists somewhere in the scholarly ecosystem. The contribution here is to argue that assembling them into a single venue that \emph{replaces} rather than supplements the journal is both feasible and preferable, and to confront the objections -- sparse participation, the chilling effect of real names, the loss of the certification signal, and the non-meritocratic distribution of attention -- that any honest version of the argument must answer.
Science Journals
arXiv:2607.05781v2 Announce Type: replace
Abstract: This paper derives a design principle for structure-preserving Galerkin formulations of energy-conserving hyperbolic systems. The aim is to reproduce the modal-energy-exchange structure of the continuous system within a resolved finite-mode space. Total energy conservation follows from this structure. We introduce a state-dependent physical-energy metric H and derive the corresponding energy-compatibility identity. In the infinite-mode exact-integration model, the volume contribution has an antisymmetric representation after H-orthogonalization, yielding pairwise modal energy exchange. Interface contributions take the same exchange form. To reproduce this structure in the practical finite-mode system, we combine two constructions: a Galerkin projection coupled with the physical-energy metric that guarantees the H-metric summation-by-parts identity, and an energy-compatibility closure that removes the component of the compatibility action contributing to the scalar energy residual. With a shared numerical energy flux at interfaces, they close the total-energy balance of the finite-mode system while preserving pairwise modal energy exchange. We also compare the practical operator construction with the finite-mode exact-integration reference and obtain an O(h^p+1) defect estimate. Finally, we derive an equivalent form of the resulting equation in the fixed Galerkin basis for direct implementation.
arXiv:2607.05816v2 Announce Type: replace
Abstract: This paper shows that bit-level reliability can be converted into block-level reliability for binary codes over BMS channels through Forney's concatenation scheme. The construction concatenates an inner code of rate R-o(1) and vanishing bit-error probability \epsilon with an outer code of rate 1-o(1) equipped with a bounded-distance decoder. We show that, when the outer correction radius exceeds \epsilon and a suitable large-deviation condition holds, the concatenated code has rate R-o(1) and vanishing block-error probability. Hence, a family with vanishing bit-error probability for rates below capacity can be converted into a capacity-achieving concatenated-code family. We further show that, when binary BCH codes are used as outer codes, inner codes with bit-error probability \epsilon=o(1/\log\log n) can be converted into a capacity-achieving concatenated family.
arXiv:2607.02999v2 Announce Type: replace
Abstract: We are given an undirected weighted graph $G$ with $n$ vertices and $m$ edges, edge weights in $[1, W]$, and a designated source vertex $s$. We design a single source dual fault tolerant distance oracle for $G$. Given a destination vertex $t$ and a set $F$ of at most two faulty edges, the oracle returns a $(1 + O(\epsilon))$-approximation of the weight of the shortest path from the source $s$ to $t$ avoiding $F$. Our oracle uses $\tilde{O}(n\sqrt{n})$ space and has $\tilde{O}(1)$ query time.
Prior to our result, single source single fault tolerant oracles were known to return a $(1+\epsilon)$ approximation of the weight of the shortest path using $\tilde{O}(n)$ space and $O(1)$ query time. However, extending these approaches to multiple faults remained an open problem. Indeed, all $(1+\epsilon)$-approximate distance oracles that handle multiple faults require $\Omega(n^2)$ space. We break this bound by presenting the first dual fault tolerant distance oracle with $o(n^2)$ space.
arXiv:2607.06618v2 Announce Type: replace
Abstract: Following the CMIVQA, MMI-VQA, and M4IVQA challenges in NLPCC 2023--2025, we introduce the Difficulty-Aware Medical Instructional Video Question Answering (DA-MIVQA) shared task for NLPCC 2026. DA-MIVQA extends previous multilingual and multimodal medical video benchmarks by explicitly distinguishing questions according to the type and complexity of evidence required for answering. Specifically, simple questions can often be answered from subtitle-based textual cues, whereas complex questions require visual grounding, procedural understanding, and cross-modal evidence integration. The challenge contains three tracks: Difficulty-Aware Temporal Answer Grounding in Single Video (DA-TAGSV), Difficulty-Aware Video Corpus Retrieval (DA-VCR), and Difficulty-Aware Temporal Answer Grounding in Video Corpus (DA-TAGVC). The dataset is collected from public medical instructional channels, covers diverse scenarios such as first aid, emergency response, rehabilitation, nursing, and general medical education, and is manually verified with difficulty annotations. This paper presents the task motivation, dataset construction, evaluation protocol, participation overview, competition results, and representative systems of DA-MIVQA. DA-MIVQA provides a practical benchmark for evaluating medical instructional video question answering systems under varying textual, visual, temporal, and procedural reasoning requirements.
arXiv:2607.06623v2 Announce Type: replace
Abstract: Process industries rely on time-series forecasting and soft sensing to estimate quality variables that are hard to measure online. Labeled data are scarce, operating regimes change frequently, and retraining models or rebuilding alignment pipelines for each scenario is costly. Such settings often provide variable tables and process documents that record variable names, units, physical meanings, and process roles. However, standard time-series backbones usually treat inputs as anonymous numerical columns. Existing text-enhanced methods also rarely make the semantic-logical relations between input variables and the prediction target available to the model within each numerical window. To address this problem, this article proposes Task-Semantic Field Factorization (TSF), a large language model (LLM)-guided framework. TSF builds a task-semantic field from task protocols and variable documents before training and uses the LLM only for offline semantic construction. Online training and inference are handled by conventional time-series backbones. During training and inference, the current numerical window activates variable semantics, so semantic information participates in each prediction and supports adaptation to different prediction targets and operating shifts. Across multiple complex industrial forecasting and delayed soft-sensing tasks, TSF reduces MAE by 3.6\% on average. Across all dataset--backbone pairs, the macro-average reduction is 2.9\%, with a maximum reduction of 24.9\%. It adds only about 0.7--4.3k parameters, with less than 8\,$\mu$s/sample of additional online inference overhead. These results show that TSF turns existing process documents into measurable forecasting gains across backbones and semantic generators while remaining lightweight for deployment.
arXiv:2607.07304v2 Announce Type: replace
Abstract: In this paper we first study the problem of generalized linear bandit (GLB) under heavy-tailed noise. The characteristics of heavy-tailed distributions are widely observed in real-world applications such as personalized recommendation, financial markets, and medical treatments. Based on the online mirror descent (OMD) method, we propose an algorithm EHM that extends the adaptive Huber loss method (Wang et al., 2025) with one-pass update ($\mathcal{O}(1)$ computational complexity with respect to current round $t$ and the time horizon $T$), which simultaneously achieves an almost optimal regret of $\widetilde{\mathcal{O}}(T^{\frac{1}{1+\epsilon}})$ where $T$ is the time horizon. In addition, by utilizing a special property of some link function (Sawarni et al., 2025), our algorithm eliminates the need to know a commonly used parameter. Next, we study the GLB problem under the case when contextual characteristic becomes piecewise constant, and we slightly revised former algorithm to obtain the PGLB-EHM algorithm. After theoretical analysis, we prove that the regret upper bound order stays the same. Furthermore, we look deeper into a special case of nonlinear bandit (NB) and present the NB-EHM algorithm with bisection method and special restriction. Eventually we utilize the affine lifting approach and show that the general NB problem can be applied with NB-EHM to achieve a sublinear regret bound.
arXiv:2602.19962v3 Announce Type: replace
Abstract: We calculate parity-violation-induced E1 amplitudes for the $1s\rightarrow 2s$ and $1s^2 2s\rightarrow 1s^2 3s$ transitions in H- and Li-like ions of $^{40}$Ca, $^{48}$Ca, and $^{208}$Pb. In our analysis, we account for neutron skin effects and nuclear uncertainties for each nucleus. We consider the spin-independent weak-interaction contribution of the $Z^0$ boson described by standard model, as well as the effects of a hypothetical new $Z'$ boson of varying mass. We conclude that the neutron-skin corrections in the $^{40,48}$Ca isotope pair can mostly be neglected when considering $Z'$ boson effects, which is an advantage for the search for new parity-violating physics. On the other hand, both the neutron skin effect and the sensitivity to hypothetical $Z'$ interactions in $^{208}$Pb are shown to be significant.
arXiv:2602.19571v2 Announce Type: replace
Abstract: We introduce Video-HOCA, a diagnostic benchmark for physical anomaly reasoning in videos. Video-HOCA uses an Ontological-Causal taxonomy to distinguish violations of an entity's own properties or capabilities from violations of physical relations among entities and the environment. It contains more than 1,400 generated and real-world videos and 3,470 question-answer pairs, with human verification of labels and reference answers. The benchmark evaluates four levels of reasoning: plausibility checking, anomaly attribution, fine-grained recognition, and open-ended physical explanation. Across 20 Instruct-mode Video-LLMs, we find that recognition outpaces explanation: Task I scores cluster at 75-88, while Task II macro-F1 stays mostly below 50. We also find that the Ontological-Causal gap depends on the task and model configuration, and that Thinking-mode gains are not explained by sampling or output budget alone. Annotation agreement, Task-IV human-judge and judge-judge checks, alternative metrics, and temporal/decoding controls validate the evaluation pipeline and bound the claims supported by the benchmark.
arXiv:2607.07734v2 Announce Type: replace
Abstract: We introduce finite-$N$ dynamics for a bundle of labeled classical trajectories under a constraint fixing the symplectic-area scale of its covariance. The constraint forces have zero bundle-averaged power, and finite bundles may be non-Gaussian. If the initial distributions have a Gaussian large-$N$ limit, the limiting density remains Gaussian. With matched initial data and the scale set to $\hbar/2$, the limiting dynamics coincides with variational Gaussian wave-packet dynamics for the full phase-space density. $N$ counts trajectories, not orders in a Moyal or $\hbar$ expansion.
arXiv:2607.08287v2 Announce Type: replace
Abstract: Newell's simplified car-following model offers behavioral interpretability with minimal parameters, yet the temporal limits within which it holds during actual car-following interactions, particularly in lane-free traffic, remain poorly understood. This has direct consequences for Newell parameter estimation. This study evaluates the model using high-resolution UAV trajectories from a lane-free highway corridor in Chennai, applying two independent estimation approaches: linear regression of the speed-spacing relationship (aggregate and pair specific), and a shifting optimization method that recovers parameters by minimizing spacing variance. A boundary-corrected variant of the shifting method is then introduced to isolate transitional regimes at interaction endpoints while preserving the original Newell formulation. Pair-specific regression substantially outperforms the aggregate specification ($R^2$ = 0.84 vs. 0.54), underscoring pronounced driver heterogeneity. Parameters recovered via trajectory shifting are statistically equivalent to regression estimates, providing independent support for the trajectory translation principle underlying Newell's model. Boundary correction markedly improves model fit, increasing the mean $\bar{R^2}$ from 0.66 to 0.95, and is preferred for 81% of pairs according to the Bayesian Information Criterion (BIC). The results indicate that interaction boundary effects, rather than failures of the core behavioral assumptions, are the principal source of estimation error in this setting. Correcting for these effects improves parameter recovery without compromising model parsimony. Boundary correction is therefore a necessary step in trajectory-based calibration of Newell-type models, with implications for traffic state estimation and microscopic traffic simulation.
arXiv:2607.08601v2 Announce Type: replace
Abstract: Sparsely-activated Mixture-of-Experts (MoE) language models achieve remarkable inference efficiency by activating only a small fraction of parameters per token, yet their full expert banks reside in memory at all times, creating a prohibitive deployment bottleneck. Existing structured pruning methods, largely designed for dense transformers, assess expert importance using locally derived heuristics that are blind to the interdependent nature of MoE routing. We introduce MAESTRO (Markov-chain Approximated Expert Sparsification via Transition-based ROuting), a structured pruning framework designed for MoE architectures that models autoregressive expert activation trajectories as Ergodic Markov chains whose stationary distributions encode cross-layer dependencies, yielding a globally aware importance heuristic. Evaluated across five diverse domains including Safety, Bias, and Ethics, MAESTRO outperforms state-of-the-art baselines by up to 10.61% in average performance retention under a strict 50% compression regime, while exhibiting substantially lower cross-task variance, indicating that global, routing-congruent pruning produces models that generalize more consistently across heterogeneous tasks.
arXiv:2607.08899v2 Announce Type: replace
Abstract: Signal Temporal Logic (STL) is increasingly used to describe interpretable objectives and constraints for optimal control and learning methods, especially when no target time series data is available. In this work, we propose to synthesize parameters for nonlinear systems that robustly satisfy continuous-time STL specifications for uncertain initial conditions. To this end, we use gradient-based optimization along with set-based reachability verification to efficiently learn in high-dimensional parameter spaces while providing provable satisfaction guarantees for the optimized parameters. We demonstrate the effectiveness and scalability of our method on three systems with up to 18 parameter dimensions.
arXiv:2607.08986v2 Announce Type: replace
Abstract: We formalize a research result in the Lean 4 proof assistant by having a mathematician direct an AI system, and frame the activity as a formalization game. The objective is to turn a LaTeX document into Lean. The game is won when the development compiles, contains no sorry, and a machine check shows the target theorems rest on Lean's foundational axioms alone. Reuse is a second check, by a definition we introduce: whether the development yields a self-contained layer of general mathematics the wider library could absorb. The case study is a complete, axiom-clean formalization of well-posedness for the nonlinear Vlasov equation via Dobrushin's mean-field route -- existence, uniqueness, the stability estimate and mean-field limit, and a short-window superposition principle (weak solutions are Lagrangian). The human's role was to direct, not to write proofs: to scope the definitions, steer the decompositions, and triage the library's gaps; the AI agent executed. The formalization certifies the proof of each statement as written; whether the written statement is the intended theorem stays the mathematician's judgment. The optimal-transport machinery that fell out of the build (in particular, properties of the Wasserstein-1 metric and the Kantorovich-Rubinstein duality theorem) separates into a self-contained layer that compiles against Mathlib alone: about a sixth of the development (49 of 299 declarations), behind a 22-declaration interface with no reverse dependency. The headline theorems ran in about a week, the full development in about a month. We report the quantitative claims as observations of one game, not as general laws. The game's rules name no particular system, so the methodological framing is meant to outlast the tools of any one run.
arXiv:2607.09260v2 Announce Type: replace
Abstract: We present a novel integrated architecture for robust online 3D Gaussian splatting, real-time VR exploration, and speech-driven Vision-Language-Model interaction. Unlike methods assuming clean depth or external poses, our system combines ORB-SLAM3-based pose estimation with online Gaussian reconstruction for noisy real-world data. A VR pipeline enables immersive exploration of incremental reconstructions; a semantic module transcribes voice commands, generates scene descriptions, and records points of interest. Against state-of-the-art online Gaussian splatting methods, we improve image quality on our dataset (+14.5% PSNR, +8.6% SSIM, -14.3% LPIPS) and TUM-RGBD (+11.7% PSNR, +7.8% SSIM, -21.6% LPIPS), with comparable or superior frame rates via quality-speed configurations. We achieve an 88% VLM object-recognition rate.
arXiv:2607.09540v4 Announce Type: replace
Abstract: The evolution of recommender systems can be explored by asking how they utilize information at scale. Throughout most of the historical period under consideration during the past two decades, industrial systems have relied on raw IDs, which are discrete, globally unique, and semantically opaque identifiers that enable exact lookup, logging, and item-specific memorization at scale. Over time, however, recommender systems have sought to utilize richer sources of information, including item content, context, multimodal signals, and cross-domain structure. This development has led to a new stage in which part of such information is no longer used solely as auxiliary features around item identity, but is increasingly encapsulated in semantic IDs that provide a more structured, model-facing form of identity. We argue that this shift goes beyond the rise of generative recommendation over traditional methods. Indeed, it reflects a broader evolution in how recommender systems utilize information under industrial-scale constraints. This paper looks at the past, present, and future to examine three connected questions: why raw IDs dominated the early development of recommender systems, why semantic information is increasingly being encapsulated in IDs today, and what may come next once recommendations move beyond semantic retrieval. In particular, we introduce semantic planning as a possible future direction in which the system first predicts the semantic target of the next exposure, and only then instantiates that target as a specific item or generated creative. We further argue that such a shift may require changes not only in model design but also in evaluation and in the way recommender systems coordinate the objectives of users, platforms, and providers.
arXiv:2607.09637v2 Announce Type: replace
Abstract: Network engineers often need to perform network diagnosis and inference tasks, which frequently require answers to enumeration questions such as "Which packets from the Internet arrive at host C?" or "Which single-link failures disconnect my network?" Parametric NetKAT is a new domain-specific language that combines elements of NetKAT, Relational NetKAT, and Weighted NetKAT into a single system and extends them with parameters, allowing users to pose such enumeration questions directly over network models. This paper presents the design and semantics of Parametric NetKAT and illustrates its utility through a series of examples. It shows how to compile Parametric NetKAT into NetKAT automata, develops new algorithms for efficiently collecting satisfying valuations, and proves the correctness of these procedures. Finally, it evaluates the performance of Parametric NetKAT on a collection of benchmarks drawn from industrial sources.
arXiv:2607.09657v2 Announce Type: replace
Abstract: The rapid progress of large foundation models has been driven predominantly by pretraining on large-scale text corpora. However, many forms of knowledge are conveyed through visual representations, where figures, typeset equations, and page layouts carry rich information that cannot be faithfully or completely captured by text alone. Yet current pretraining approaches discard these visual cues by converting visually rich sources, such as documents and web pages, into plain text for learning language intelligence. This paper challenges the default assumption that language models must be trained on text-only representations and shows that Visual Pretraining is a scalable learner for foundation model intelligence. To this end, we conduct a systematic study of unsupervised visual pretraining paradigms that directly leverage visual documents without text extraction. Across multiple backbones and benchmarks, visual pretraining on the same underlying corpora consistently outperforms text-only pretraining, offering an efficient pathway to scalable language intelligence.
arXiv:2607.09747v2 Announce Type: replace
Abstract: When is a material system a candidate for life at all? We argue that this question is prior to behavior, functional architecture, or computational capacity, and that at root it is one of physical admissibility. We develop a framework in which minimal autonomy, taken in the interactivist sense of normativity grounded in self-maintaining far-from-equilibrium organization, corresponds to a distinct non-equilibrium phase of matter, and we take complex plasmas, a physical and non-biological system, as its in-materio exemplar. We formalize a diagnostic phase-space whose criteria (sustained free-energy throughput, organizational closure, active information maintenance, and regulated noise sensitivity) constitute necessary conditions for life-attribution. We instantiate the diagnostics across contrasting systems and fix the boundaries of the phase space via B\'enard convection as a driven baseline lacking closure, and a digital self-replicating soup that carries measured informational heredity while its physical closure remains a structural zero. We demonstrate that plasmas satisfy every admissibility condition for minimal physical autonomy while carrying none of the informational heredity that open-ended evolution requires, sharpening the distinction between physical admissibility and biological sufficiency, and bounding downstream questions of machine sentience.
arXiv:2607.09796v2 Announce Type: replace
Abstract: Direct Preference Optimization (DPO) has become an important method for aligning large language models (LLMs) with human preferences because it removes the need for explicit reward modeling and reinforcement learning. However, its performance depends heavily on the quality of preference data, and noisy preference data in real-world settings can weaken alignment performance. To address this issue, we propose a bilevel optimization framework and prove, under some idealized conditions, that this framework can recover the DPO optimum under clean data. We further derive a prior form for the learnable weighting function under label-flipping noise. Considering that high-quality metadata may be difficult to obtain, we propose a prompt augmentation consistency method that enables meta-learning even when metadata is completely unavailable. To reduce the high cost of higher-order gradients in LLM meta-learning, we combine central-difference approximation with LoRA fine-tuning and develop a scalable training scheme. Experiments on TL;DR summarization and Anthropic Helpful and Harmless dialogue show that the proposed method improves alignment performance over multiple DPO baselines under different noise rates.
arXiv:2607.10139v2 Announce Type: replace
Abstract: Selecting the correct answer from a pool of candidate reasoning chains is the engine of test-time scaling, yet the standard selectors each carry a cost: self-consistency inherits the errors of the single model it resamples, and trained reward models need labeled data and transfer poorly off-distribution. We study a third signal, free at inference time: cross-model consensus, the degree to which independently trained models, each solving the problem once, agree on a final answer. We treat the panel as an LLM-jury, in which the structure of agreement, not any model's score of another, is the verification signal. Across seven benchmarks it selects correct answers better than self-consistency and far better than a model scoring its own candidates: on competition math it closes the entire gap to an oracle selector, while self-scoring closes almost none. The mechanism is error decorrelation: independently trained models err differently, so their wrong answers scatter while the correct one accumulates agreement. We make this precise with a parameter-free law, derived in closed form, that predicts consensus accuracy from three measured panel statistics to a mean absolute error of $0.03$ and exposes the method's ceiling: a shared-error floor where models share a misconception, near zero on math but non-trivial on science. Against four trained verifiers spanning discriminative, outcome, and generative reward models, the free LLM-jury matches the strongest inside their math training domain and is the top selector outside it. Cross-model consensus is thus a verifier we can characterize in advance: a law that says when to trust it, and a floor that marks where it cannot.
arXiv:2607.10144v3 Announce Type: replace
Abstract: Scientific ideation unfolds over multiple stages, including literature search, paper reading, tool use, claim checking, cross-paper synthesis, brainstorming, rejection of weak directions, and iterative writing. Yet most existing resources capture isolated components or final artifacts rather than the process connecting them. We introduce IdeaTrail, a dataset of 1,170 multi-turn trajectories for scientific ideation and proposal generation. Each trajectory follows a research process from evidence gathering to either idea selection or proposal construction, jointly recording tool use, acquired evidence, intermediate artifacts, and reasoning. IdeaTrail is synthesized from human-selected research papers and proposal artifacts through a Generator--Advisor loop. The Generator produces the visible sequence of actions, observations, and artifact edits, while the Advisor uses the full generation context to check grounding, causal order, naturalness, and leakage from hidden targets. This reverse-to-forward design keeps trajectories aligned with real scientific artifacts while retaining the uncertainty, evidence use, and staged convergence characteristic of research practice. IdeaTrail provides both reusable process supervision and a general recipe for constructing scientific-research-agent data.
arXiv:2607.10436v2 Announce Type: replace
Abstract: Input--output feedback linearization fails at decoupling singularities, where the decoupling matrix loses rank, the relative degree is lost, and the linearizing control becomes unbounded. This paper develops a singularity-robust trajectory-tracking controller for square nonlinear control-affine systems that tracks through isolated decoupling singularities with bounded control. The method recasts tracking as real-time arc-length homotopy continuation, equivalently a continuous-time Newton/Davidenko flow, and replaces the inverse decoupling matrix by the least-norm Moore--Penrose solution of an augmented matrix $A=[\Lambda\mid b]$, where $b$ is the homotopy direction. A transversality condition $w^T b \ne 0$, with $w$ in the left null space of the decoupling matrix, keeps the augmented matrix full row rank through a generic rank-one loss. The resulting flow agrees with feedback linearization away from the singular set, tracks with $O(1/k)$ error, and re-locks after each crossing. The theory also characterizes the reflection-versus-branch-crossing dichotomy at Whitney folds and relates the reflection case to a Filippov sliding mode. Extensions cover dynamic relative-degree-one minimum-phase systems and arbitrary relative degree via filtered-error reduction. Simulations include a redundant 2-DOF manipulator, relative-degree-one and relative-degree-two plants, and a dual-active-bridge series-resonant DC/DC converter, where the method performs bounded inversion across buck/boost and resonance singularities while preserving zero-voltage soft switching.
arXiv:2607.10623v2 Announce Type: replace
Abstract: Flow matching over carefully designed latent representations has recently emerged as a powerful paradigm for topology-aware mesh generation. Existing approaches, however, model vertices and connectivity jointly in a joint latent space, entangling continuous vertex geometry with discrete combinatorial structure; this complicates flow learning and manifests as drifting vertices and broken surfaces. We present LATO.2, a factorized flow matching framework that decomposes mesh generation into a vertex flow followed by a connectivity flow conditioned on the realized vertices, with both stages anchored to a shared coarse voxel scaffold. Dedicated VAEs underpin the two stages, recovering vertices at sub-voxel precision and embedding discrete connectivity into a continuous latent space. We demonstrate two advantages unique to this factorization: (i) part-wise generation, in which the scaffold is partitioned and each part synthesized at full latent capacity, yielding substantially higher-resolution meshes than a monolithic latent permits; and (ii) topology-adaptive editing, in which manipulating first-stage vertices induces the corresponding connectivity without re-optimization. Experiments show that LATO.2 surpasses state-of-the-art topology-aware mesh generators in geometric fidelity and connectivity quality.
arXiv:2607.17201v1 Announce Type: cross
Abstract: In this work we study the Best Policy Identification (BPI) problem in online, tabular Reinforcement Learning. This is an active sequential hypothesis testing problem in which the learner's objective is to identify an optimal policy in a Markov Decision Process (MDP) with high confidence, while minimizing the expected sample complexity to do so. We consider an online setting with deterministic rewards, where the agent must strategically navigate through the MDP in order to effectively explore. Previous works in the literature have provided asymptotically optimal methods for BPI, such as the Navigate and Stop (NaS) algorithm and its variants, however existing analysis remains asymptotic. In this work, we fill that gap by providing the first non-asymptotic sample complexity guarantees for NaS, showing that its sample complexity depends not only on the characteristic time, but also on the connectivity of the underlying MDP, the curvature of the optimal characteristic time, and other instance-dependent quantities. We identify these additional attributes and make explicit their contributions to the overall sample complexity.