Anna Zykova-Myzina, Timofei Gritsaev, Daniil Tiapkin, Nikita Morozov
原文摘要
This paper explores policy gradient algorithms for training stochastic policies to sample from structured discrete probability distributions under the Generative Flow Network (GFlowNet) framework. Building on extensive theoretical connections between GFlowNets and entropy-regularized reinforcement learning, we derive equivalents of standard policy gradient algorithms for training GFlowNets, as well as experimentally explore their various methodological aspects, including baseline training and advantage estimation. Most importantly, our work is the first to derive and successfully apply proximal policy optimization to GFlowNets, showing its improved convergence speed and data efficiency compared to standard GFlowNet training objectives on benchmarks ranging from synthetic energies to molecular graph generation.
On July 1, 2025, Vietnam reduced its 63 provinces to 34, abolished its 696 districts, and replaced its 10,035 commune-level units (wards) with 3,321. ViAdminShift, a Vietnamese language resource about this event, consists of 7,995 benchmark items, 2,281 retrieval passages from pre- and post-reform Wikipedia revisions and dated templates, and old-to-new tables that a released script rebuilds from the official conversion table (not redistributed). Changed items store the pre-reform answer (except 121 address items), a $2\times2$ design uses untouched units as controls, and 400 addresses appear without diacritics. The tables agree with an independent lineage on 10,564 of 10,565 official pairs, disagreements were settled against the resolutions, and the probing-track scorer agrees with both authors' blind labels (Cohen's $κ=0.91$ and $0.89$). Data, outputs, code, and a datasheet are released. In the probing track, at least 20 of 67 models know the reform. Fourteen of them give the new province alone for at most 11% of plain questions about old addresses (old province only: 14-100%); a bare date recovers up to 100%. Post hoc (80 province-free addresses, single runs), with reasoning off, today's date in the system prompt leaves 68-100% old-province-only answers for 10 of 12 models; with reasoning, an "as of" prefix in the question leaves none for seven of eight. On the 1,262-item benchmark subset, with same-source old and new passages, small models unaware of the reform give 14-30% outdated answers and the five benchmark-track knowers (also differing in size or provider) 0-5%.
A depth lower bound for ReLU networks asserts that every sufficiently shallow network fails on some input. What does a proof of such a bound compute? We study this question for exact computation of the maximum of $n$ real inputs, with depth measured by the number of hidden layers. For every separately fixed prime $q$ and depth bound $K$, we give a uniform $\mathsf{TC}^0$ refuter for rational networks with at most $K$ hidden layers and more than $q^K$ inputs, provided that the reduced denominators of their nonbias weights are not divisible by $q$. Biases may be arbitrary rationals. The refuter finds a rational input on which the network fails, and its correctness is provable in $\mathbf{VTC}^0$, the bounded-arithmetic theory of uniform constant-depth threshold computation. The underlying numerical depth lower bound follows from Averkov, Hojny and Merkert (ICLR 2025), extending Haase, Hertrich and Loho (ICLR 2023); our contribution is an internal proof that makes its geometric argument into a feasible refuter. On the construction side, $\mathbf{VTC}^0$ verifies locally checkable certificates for the balanced integer-weight maximum network of depth $\lceil\log_2 n\rceil$. At $q=2$, the construction and refuter meet: for each fixed $K$, the theory verifies that integer-nonbias-weight networks with arbitrary rational biases can compute the maximum with at most $K$ hidden layers exactly when $n\le 2^K$. We also give certified rational constructions of polynomial encoding length and depth $O(\log n/\log\log n)$. These results instantiate a witnessing duality: bounded proofs of certified existence and of refutation correspond, respectively, to threshold-computable constructors and refuters whose guarantees the theory proves.
We release all 3,718 episodes from the ArmnetBench v0.1 evaluation (arXiv:2607.24481), which contains 600 teleoperated demonstrations and 3,118 rollouts of seven imitation-learning and vision-language-action policies on 12 tabletop tasks, recorded on low-cost SO-101 arms; 73% of the rollouts are failures. Every policy for a task was trained or fine-tuned on the same 50 demonstrations and rolled out on the same cell, so the corpus contains many policies' failures on shared tasks. An operator labelled every rollout successful, suboptimal or failure. Each episode has three camera views, joint states and actions and is available on the Hugging Face Hub in LeRobot v3.0 and a reward-model format. This note documents how the episodes were collected and the limitations that could affect their use.
Data agents aim to fulfill natural-language instructions over heterogeneous data, including tables, files, and databases. However, data agents face a challenging agent-data gap: heterogeneous data resides outside the agent, while the agent can access it (e.g., column names and file paths) only through generic tools. Existing approaches either let agents directly explore raw data sources or inject manually constructed semantic layers into prompts. However, neither scales well to large heterogeneous data sources nor adapts to different agent behaviors. In this paper, we introduce EvoOntology, a self-evolving ontology layer for data agents. EvoOntology encapsulates the ontology as an MCP server comprising a schema layer, a content layer, and a tool layer, enabling agents to actively query and interact with the ontology at runtime. To this end, we introduce a builder agent for autonomous ontology construction and a self-evolution loop that continuously refines the ontology through attribution-guided typed edits that are accepted only after a backbone-conditional paired evaluation. Experiments on three well-adopted data-agent benchmarks with four LLM backbones demonstrate that EvoOntology consistently outperforms strong baselines and existing semantic-layer approaches, effectively bridging the agent-data gap and enabling more effective interaction with heterogeneous data. Code: https://github.com/ruc-datalab/EvoOntology
Choosing a robot action model requires evidence about task capability, adaptation cost, deployment requirements, and the reliability of reported conclusions. Success rates alone offer limited guidance to researchers with different computing budgets and goals. We propose RoboSeed, a multidimensional evaluation framework for vision-language-action (VLA) and world-action models (WAM), within a continuing program to train, evaluate, and openly release robot models with verifiable evidence for model selection. The framework defines five dimensions: capability, downstream training affordability, deployment affordability, locally verified research value, and reproduction evidence. It maps empirical measurements and audit criteria to 0-100 profiles using fixed normalization anchors, with balanced, training-constrained, and deployment-constrained preference settings. Interface and protocol checks establish when comparisons are appropriate. Missing measurements remain explicit, incomplete aggregate scores are withheld, and weighted input coverage identifies evidence gaps. We apply the framework to representative VLA and WAM models, drawing on local training, released-weight evaluation, task-level audits, and an evaluation-system study. The results reveal budget-dependent backbone choices, sensitivity to adaptation recipes, and differences between language and visual task interfaces. Reproducible scoring code and machine-readable evidence support extension to additional models. The study provides dimension-level assessments and priorities for further measurement, while incomplete serving traces and heterogeneous protocols preclude overall numerical rankings.
Harald Kugler, Leonard Henckel, Sebastian Weichwald
原文摘要
There are many causal discovery algorithms, but few guiding principles for choosing between them. Methods are usually compared through three lenses: identifiability, consistency, and benchmarks. Each introduces assumptions about the data-generating process that are often untestable. We argue that there are no shared standards for comparing, prioritizing, or even discussing those assumptions. We support this claim by showing that (i) identifiability assumptions are often preferred by convention rather than by principle, (ii) uniform consistency requires further untestable assumptions, while the weaker pointwise consistency has arguably narrowed algorithm design, and (iii) benchmarks are not neutral tests but carry an epistemic role of their own: they quietly fix which regimes matter, which guarantees come into play, and what counts as success. In doing so, they shape what algorithm comparisons are taken to show and thus perform some of the same epistemic work as theory, but less transparently. In this position paper we argue for a more transparent, integrated practice: Make the epistemic standards for discussing identifiability and consistency assumptions explicit, especially their failure modes and scope conditions; design benchmarks to operationalize those standards, by stress-testing and ablating both. Then, theoretical and empirical results jointly clarify when, why, and how algorithms may work rather than accumulate as disconnected advances.
Hendrik Möller, Hanna Schoen, Robert Graf, Matan Atad, Nathan Molinier, Anjany Sekuboyina, Bettina K. Budai, Fabian Bamberg 等 17 位作者
原文摘要
The human spine commonly consists of seven cervical, twelve thoracic, and five lumbar vertebrae. However, enumeration anomalies may result in individuals having eleven or thirteen thoracic vertebrae and four or six lumbar vertebrae. Although the identification of enumeration anomalies has potential clinical implications for chronic back pain and operation planning, the thoracolumbar junction is often poorly assessed and rarely described in clinical reports. Additionally, even though multiple deep-learning-based vertebra labeling algorithms exist, there is a lack of methods to automatically label enumeration anomalies. Our work closes that gap by introducing "Vertebra Identification with Anomaly Handling" (VERIDAH), a novel vertebra labeling algorithm based on multiple classification heads combined with a weighted vertebra sequence prediction algorithm. We show that our approach surpasses existing models on T2w TSE sagittal (98.30% vs. 94.24% of subjects with all vertebrae correctly labeled, p < 0.001) and CT imaging (99.18% vs. 77.26% of subjects with all vertebrae correctly labeled, p < 0.001) and works in arbitrary field-of-view images. VERIDAH correctly labeled the presence 2 Möller et al. of thoracic enumeration anomalies in 87.80% and 96.30% of T2w and CT images, respectively, and lumbar enumeration anomalies in 94.48% and 97.22% for T2w and CT, respectively. Our code and models are available at: https://github.com/Hendrik-code/spineps.
LaBraM, an early EEG foundation model, demonstrated that large-scale EEG pretraining can yield strong representations for diverse downstream tasks. Building on this foundation, we expand pretraining data and model capacity while retaining its core encoder and objective. We train ten encoders with 5.8 million to 1.97 billion parameters on three EEG collections spanning 25,067 to 79,910 hours, adapting signal preparation and training to heterogeneous recordings. Across 23 tasks, Atom-L* reaches 59.05% mean balanced accuracy at its best evaluated full fine-tuning (Full-FT) checkpoint, 3.61 percentage points above the strongest baseline EEG foundation model. Quark-G leads linear probing (LP) at 54.33%, a 1.24-point margin. Both also lead average task rank. At the last evaluated checkpoints, larger encoders improve Full-FT more consistently than LP, and stronger LP does not guarantee stronger Full-FT. Preserving patch structure improves LP over average pooling. Data-fraction experiments suggest capacity-dependent saturation, while checkpoint trajectories show that downstream performance can decline as pretraining loss falls. Joint classification with a shared Qwen3 readout reaches 50.50% with Molecule-H and Qwen3-8B; increasing language-model size provides no consistent gain. These results demonstrate the value of scaling LaBraM and the importance of matching pretraining scale to downstream adaptation.
Fine-grained manipulation couples instruction grounding, local visual precision, and sustained execution. We present FineCraft, a framework that adapts pretrained models for fine-grained manipulation in the MetaFine Challenge. A task-routed mixture of policies selects independently adapted DINOv2-DiT or pretrained $π_{0.5}$ policies through a common observation-action interface. Local visual features help identify the instructed object parts, demonstrations and expert continuations teach the required actions, and new observations guide replanning during execution. FineCraft ranks fourth with an overall score of $0.878$, exceeding the published reference on all five tasks. Official results and local diagnostics characterize performance across visual perturbations, instruction changes, and multi-stage manipulation. Code is available at https://github.com/eqcy/FineCraft
Repository-level AI coding benchmarks measure task success but do not directly capture the conditions required for software maintenance completion. We study multi-condition fulfillment, where a maintenance task must satisfy multiple operational conditions, sometimes with directional dependencies. Using a fixed 18-criterion taxonomy, we analyze 2,925 Apache Fineract issues and 944 tasks from SWE-bench Pro, SWE-PolyBench Java, and SWE-EVO. We examine condition breadth, corpus-level co-occurrence, and source-grounded directional requirement relationships. Among 1,177 benchmark task pairs matched on issue classification, modified-file count, and code churn, the mean fulfillment-set Jaccard distance is 0.8138, and 57.4% have disjoint condition sets. Condition count alone does not explain this contrast: SWE-bench Pro averages 2.47 conditions per task, versus 1.78 for Fineract. Temporal (B4) and numerical (B2) conditions have a phi coefficient of +0.430 in Fineract versus -0.013 in the pooled benchmark corpus. Among temporally tagged tasks, numerical conditions are also specified in 48.85% (254/520) of Fineract cases versus 5.81% (5/86) of benchmark cases. We document 16 directional relationships in Fineract requirements involving time, calculation, and state preservation, and inspect four benchmark cases to illustrate how relevant conditions appear in evaluation assertions. These examples do not quantify benchmark-wide oracle omissions. The results show that similar surface task profiles can mask substantially different fulfillment-condition sets. We recommend reporting condition breadth, co-occurrence, and directional relationships separately. The corpus comparisons are descriptive and may reflect domain differences; they do not establish causality or production failure rates.
We study deterministic search by $n \geq 2$ unit-speed robots starting at the origin of the Euclidean plane. We introduce the concept of \emph{segment visibility} whereby a stationary target $x$ at distance $d=\|x\|>0$ is discovered at the first instant $t$ when $x$ belongs to the segment joining a pair of robots at time $t$. Discovery time is compared with the omniscient optimum $d$. We distinguish two knowledge models according to whether the robots know $d$ in advance. For $n=2$ robots, in the \emph{known-$d$} model, we give a symmetric four-phase algorithm and prove that its competitive ratio is optimal: $1+\sqrt3+π/6$; the matching lower bound follows from the known-distance two-robot shoreline-search theorem through an explicit reduction to segment detection. For $n=2$ robots, in the \emph{unknown-$d$} model, a three-circle construction and a quantitative initial-path estimate yield the stronger lower bound $\sqrt3+π/6+\sqrt{17}/4 \approx3.286425990$, giving an explicit positive cost for not knowing the radius; we also analyze a feasible antipodal-diameter logarithmic-spiral algorithm with optimized ratio $8.928516954\ldots$, thus proving that the competitive ratio when $d$ is unknown is at least $ \sqrt3+π/6+\sqrt{17}/4 $ and at most $8.928516954\ldots.$ For $n \geq 3$ robots, the competitive ratio of finding the target is exactly $\sec (π/n)$ regardless of whether or not $d$ is known to the robots.
Knowledge is how understanding accumulates across people and generations: durable records that outlive their producers and that others can check, combine, teach, and correct. AI agents are now built and run by many independent users and organizations, and each agent is upgraded, replaced, or retired within months. Yet what an agent learns stays trapped in model parameters, a transient context window, or a system-specific memory store that is appended to and retrieved from rather than processed. We argue that the next generation of AI should be knowledge-centric, built around an explicit, continuously evolving knowledge layer independent of any single agent. In this loop, agents discover knowledge from their environments and trajectories; the layer internalizes it by turning fragmented, heterogeneous records and experience into composable, revisable understanding; and agents across systems and generations use the result and return new outcomes for revision. While discovery and use have advanced rapidly with large language model agents, internalization remains the bottleneck. We argue that internalization should become a persistent object of training and maintenance-centered on condition-preserving conceptualization of events and deductive, inductive, and abductive reasoning across noisy, heterogeneous representations (text, tables, graphs, rules, and programs). To power this layer, we call for knowledge foundation models that provide shared neural capacity for knowledge computation, and outline a research agenda showing how an explicit knowledge layer delivers reuse, cost efficiency, auditability, and continuity across agent generations.
Spherical transformer configurations can accelerate grokking through simultaneous modifications of several model properties. The present study investigated the contributions of these modifications in the published Sphere+Fourier configuration using 960 controlled modular-arithmetic runs. A complete factorial analysis varied residual projection, readout weight normalization, logit scale, and weight decay. Across four matched settings with weight decay, the successful-run mean crossing times without projection were 8-37 times those observed with projection. However, residual projection was insufficient to induce grokking without additional interventions. Various other unprojected configurations also generalized. Since projection altered the magnitude entering the readout, these comparisons could not separate geometry from output scale. A separate 100-run fixed-horizon study demonstrated faster first crossing for the complete configuration across five train/test partitions. Faster crossing did not necessarily correspond to greater persistence over extended training. Control experiments demonstrated rapid grokking without Fourier initialization. Task and depth sweeps also identified conditions in which the configuration failed to reach the criterion by 50K epochs. Accordingly, the observed acceleration depended on interactions among the components and did not extend uniformly across the tested tasks or depths.
Xuwei Liu, Zhimin Zhang, Xiaoguang Hu, Zhongwei Liu, Yuanhui Sun, Jun Huo, Hua Wu, Haifeng Wang
原文摘要
Machine learning supports materials design, atomistic simulation and experimental characterization, yet models and workflows remain fragmented. We present PaddleMaterials, an open-source framework integrating heterogeneous materials-learning tasks while preserving task-specific representations and scientific semantics. It unifies data processing, training, evaluation, pretrained-model management and workflows across six task classes: property prediction, structure generation, machine-learning interatomic potentials, electronic-structure fields, spectrum elucidation and microscopy image processing. Task interfaces record physical quantities, units, aggregation conventions and references, making cross-task composition verifiable. A shared compiler backend enables switching from dynamic-graph to compiled execution through one setting, using graph optimization, operator fusion and hardware-targeted code generation without changing model definitions. Across 65 model instances in 11 groups, compiled execution accelerated every instance in eight groups, with group median speedups of 11%-224% and group maximum speedups of 11%-336%. Case studies on LiMnO2 and toluene link structure generation or spectrum elucidation, interatomic-potential relaxation, property prediction and electron-density prediction, using same-geometry density functional theory (DFT) calculations as references. For LiMnO2, predicted formation energies for generated and relaxed structures differed from same-geometry PBE+U references by at most 7 meV per atom, while the computational cost of model steps was less than one-twentieth of a DFT single-point calculation. PaddleMaterials extends unified materials workflows beyond conventional property, potential and structure-generation tasks while supporting verifiable cross-task operations and compiled execution.
Petr Kuznetsov, Pierre Sutra, Guillermo Toyos-Marfurt
原文摘要
An obstruction-free implementation guarantees progress to every operation that is given enough time to take steps in isolation. But, as we show in this paper, the mere presence of concurrent operations alone does not have to prevent progress; only incomplete conflicting (non-commuting) operations may do so. This progress condition, that we call conflict-freedom, is a natural generalization of obstruction-freedom that promises efficient implementations for objects exhibiting semantic commutativity. We show that, as with obstruction-freedom, every sequential object has a read-write conflict-free linearizable implementation. Our conflict-free universal construction is based on a novel generalization of the instrumental commit-adopt object, interesting in its own right.
Transient pressure diffusion in heterogeneous porous media becomes difficult to resolve efficiently when permeability is discontinuous and spans several orders of magnitude. In this paper, we develop a domain-decomposed random-feature evolutional deep neural network (DD RF-EDNN) that separates spatial approximation from temporal evolution. Permeability-informed random features are compressed into an orthonormal pressure space, and a conservative finite-volume operator is projected onto this space so that only the reduced coordinates are advanced in time. This formulation retains the dissipative structure of the discrete flow problem without repeated neural-network optimization. The analysis quantifies dictionary and singular-value truncation errors, proves contractivity under an energy-consistent dissipativity correction, and derives a conditional grid-level estimate for homogeneous dynamics and time-independent data admitting a steady lifting. Numerical experiments on low-permeability inclusions, high-conductivity blocks and channels, and three-dimensional grids assess the accuracy of the method across distinct coefficient configurations. The method reaches a final-time relative $L^2$ error of $9.54\times10^{-4}$ in the principal benchmark and maintains errors on the order of $10^{-2}$ at a permeability contrast of $10^3$. Numerical diagnostics further show that the observed error balance and temporal convergence are consistent with the analysis. These results support DD RF-EDNN as a structure-preserving and interpretable reduced formulation for transient simulations on prescribed heterogeneous media.
This paper reviews control architectures for building heating, ventilation and air-conditioning (HVAC) systems, addressing the tension among energy demand, occupant comfort and equipment operability, and situating proportional-integral-derivative (PID) regulation within a layered control stack that also includes rule-based and model-based supervisory control and data-driven reinforcement learning (RL). It examines the thermostat-PID layer from two-position and scheduled setpoint control through proportional, PI and PID regulation and their tuning, then formulates RL for building control as a Markov decision process. Its central contribution is a taxonomy of where RL meets PID: RL as a gain tuner for an existing loop, as a direct replacement for the regulatory controller, as a hybrid with model predictive control, and as supervisory setpoint and schedule generation. Across these classes, the review establishes that the physical loop remains invariant while the intervention point changes, and that the classes differ in their requirements for actuation authority, state information, modelling, commissioning and constraint handling. It further establishes that PID is inexpensive and well codified but limited by nonlinearity, time variation, discrete mode switching, loop interaction and slow plant dynamics, while RL's admissible position must be argued rather than assumed. The paper concludes that these architectures are not competing philosophies of control but nested points of attachment on one HVAC plant, with hybrid implementations possible, and that deployment trade-offs must be assessed according to the specific control interface and scaffolding required.
Hamidreza Eivazi, Eslam Sharaawy, Stefan Wittek, André Hebenbrock, Raphael Ginster, Steffen Blömeke
原文摘要
Battery degradation remains a central challenge in the development and broad application of sustainable energy technologies. Accurate degradation prediction is challenging, as battery aging emerges from complex and heterogeneous interactions among cycling behavior, operating conditions, and cell chemistry. Existing machine-learning approaches typically focus on deterministic end-of-life predictions or modeling degradation curves within fixed chemistry settings, leaving probabilistic state of health (SOH) trajectory modeling across heterogeneous chemistries underexplored. We introduce FlowBatt, a use-inspired conditional generative framework that adapts flow matching with a diffusion transformer (DiT) backbone to battery degradation trajectory modeling. FlowBatt models full SOH trajectories as probabilistic degradation processes conditioned on early-cycle capacity data. To strengthen this conditioning pathway, we use explainable AI analysis to diagnose and improve the encoder that maps early-cycle capacity matrices into conditioning vectors. We evaluate FlowBatt on five public benchmark settings spanning diverse chemistries and aging conditions, comparing it with supervised and diffusion-based trajectory models as well as established baselines for remaining useful life (RUL) prediction under shared data splits. FlowBatt achieves the best trajectory-wise SOH prediction performance on all five datasets, with errors below 2.5%, and delivers competitive RUL prediction performance. By generating multiple plausible trajectories, FlowBatt provides empirical prediction intervals for battery degradation modeling. These results indicate that flow-matching-based trajectory modeling is a promising framework for probabilistic battery health prediction.
Simon Matrenok, Nicolas Menet, Nicolas Flammarion, Andreas Krause
原文摘要
LLM post-training via reinforcement learning has shown consistent improvements in accuracy at the cost of reducing behavioral coverage - a phenomenon known as sharpening that can stall progress on difficult tasks. Repeated sampling objectives such as $\mathrm{pass@k}$ and $\mathrm{max@k}$ are known to mitigate sharpening, but the mechanism behind it remains poorly understood. Moreover, existing approaches for $\mathrm{@k}$ optimization use dedicated algorithms, and no general framework exists for adapting established policy optimization methods from $\mathrm{@}1$ to $\mathrm{@k}$. We derive the optimal policies for KL-regularized $\mathrm{pass@k}$ and $\mathrm{max@k}$ objectives in closed form. Our explicit solutions reveal that $\mathrm{@k}$ optimization naturally redistributes learning pressure based on prompt difficulty, directing it toward harder problems. Crucially, we show that standard $\mathrm{@}1$ objectives achieve these exact same optimal policies through prompt-adaptive regularization or reward transformation. Ultimately, this yields an optimizer-independent explanation for why $\mathrm{@k}$ objectives reduce sharpening on easy tasks while prioritizing success on harder ones. Building on these insights, we introduce $\mathrm{@k}$ Adaptive Pressure for Policy Optimization (KAPPO). Following a one-time reference calibration, KAPPO seamlessly adapts existing $\mathrm{@}1$ methods to $\mathrm{@k}$ via prompt-adaptive regularization or reward transformation, without otherwise altering their training procedures. Experiments with several RL methods on mathematical reasoning and code generation demonstrate that KAPPO provides direct control over sharpening. Crucially, we demonstrate that $\mathrm{@k}$ pressure can be fully realized with single completion training, matching the performance of multi-completion $\mathrm{@k}$ optimizers at a fraction of the cost.
We prove that any linear $3$-query locally correctable code $C \colon \mathbb{F}_2^k \to \mathbb{F}_2^n$ with constant correction radius must have length $n \geq 2^{Ω(\sqrt{k})}$. This improves on the prior best lower bound of $2^{Ω(\sqrt{k/\log k})}$, due to [AG24], and shows that the folklore construction of linear $3$-LCCs from Reed-Muller codes, which achieve length $n = 2^{O(\sqrt{k})}$, is optimal up to constant factors in the exponent. Our proof builds on the approach of [KM24a], further refined in [Yan24,KM24b], that uses "chain derived equations" in combination with Kikuchi graphs. The crux of our argument is to show that a certain edge-coloured Kikuchi graph has a unique colour cycle: a cycle in which some colour is used exactly once. We do this by controlling the distribution of colours at each vertex in the graph using a random walk mixing argument and a novel hypergraph decomposition based on edge expansion of the hypergraph's "flattening" into a graph.
Zichong Zhu, Zhihao Liu, Andrei Sharf, Zhanglin Cheng
原文摘要
Procedural modeling offers compact, editable, and reusable representations for architectural content, but building such models still depends heavily on manually authored grammars. We study image-based procedural building reconstruction from in-the-wild images, with the goal of recovering an executable procedural description that can be instantiated as an editable 3D building asset. Because a single image provides only partial and view-dependent evidence, we focus on recovering the dominant visible facade(s) and attach the recovered facade program to a simple mass model for building-level instantiation. Our method decomposes this task into two structured prediction stages. First, a Global Structure Layout Tree recovers facade-scale split-and-repeat organization through a fine-tuned layout model with render-guided visual-symbolic reflection. Second, a Component Parameterization schema maps representative facade tiles to structured procedural attributes for local architectural component synthesis. These two representations are then deterministically compiled into executable Computer Generated Architecture (CGA) programs, avoiding the brittleness of direct low-level program generation. Experiments show that the proposed framework handles both regular facades and challenging in-the-wild images, while maintaining high structural consistency, procedural executability, and fine-grained architectural detail. The recovered programs further enable controllable and reusable 3D building asset generation.
This paper studies how empirical dialogue-flow statistics can be incorporated into Next Dialogue Act Prediction (NDAP). A KL regularization term is proposed that aligns predicted act distributions with corpus-derived transition patterns. Evaluated on a 60-class German counselling taxonomy using 5-fold cross-validation, this improves macro-F1 by 9--42% relative depending on encoder and substantially improves dialogue-flow alignment. Cross-dataset validation on HOPE suggests that improvements transfer across languages and counselling domains. In systematic ablations across pretrained encoders and architectures, the findings indicate that transition regularization provides consistent gains and disproportionately benefits weaker baseline models. The results suggest that lightweight discourse-flow priors complement pretrained encoders, especially in fine-grained, data-sparse dialogue tasks.
Scaling depth is a key route to stronger large language models, but Pre-LN architectures often suffer from depth-wise information dilution, limiting the effective use of deep layers. While cross-layer connectivity has proven effective at mitigating this problem, existing designs usually introduce additional computation or memory access. We revisit cross-layer KV sharing as an efficient form of such connectivity. Beyond cache compression, each KV reuse edge creates a direct gradient path from deep layers to source KV projections, turning KV sharing into an implicit structural regularizer for KV representation geometry. We provide a systematic analysis of this mechanism, showing that proper sharing topologies sharpen key retrieval directions while enriching the value content subspace, whereas existing adjacent or global-anchor designs do not fully exploit this effect. Motivated by this, we propose DepthGraft, a hierarchical KV-sharing architecture that reconstructs deep-layer KV from stride-aligned shallow-middle source pairs. DepthGraft distributes non-local gradient flow across source layers, while preserving prefill early completion and reducing KV cache by roughly one third. Experiments on dense and MoE models from 0.5B to 16B parameters show consistent downstream improvements under both GQA and MLA. Notably, on the 16B MoE model, DepthGraft achieves a 7.0% absolute improvement on MMLU-Pro, along with gains of 6.6% on CMMLU and 6.4% on C3. These results validate the potential of DepthGraft as a scalable design for large-scale pretraining. The project page is available at https://github.com/jianke0604/DepthGraft.
Diffusion models are evolving from general tools into personalized systems that adapt to user needs through fine-tuning and guidance. However, diffusion optimization mainly targets measurable objectives such as aesthetics, rarely examining whether they align with human intent. Human needs are multidimensional, contextual, and dynamic, whereas existing methods approximate them with fixed metrics. Because these metrics capture human needs only partially and may amplify deviations from true intent during optimization, higher scores need not yield results better matching human expectations. To understand how human needs enter and influence optimization, we systematically analyze how human intent is expressed, translated into optimizable objectives, applied to model behavior or generation, and evaluated through outputs. We further discuss trade-offs affecting human-intent alignment in personalized image generation and gaps between automated metrics and real user judgments. This human-centered synthesis reorganizes diffusion optimization and provides a systematic reference for personalized generation systems designed for human preferences.
We study dependence measures between two discrete random variables. The measures vanish at independence and satisfy the data processing inequality (DPI): stochastic post-processing of one variable cannot increase dependence. We ask which such measures admit an exactly unbiased estimator from a fixed finite number of i.i.d. samples, a property equivalent to being polynomial in the joint probabilities. We distinguish fixed-alphabet DPI, which preserves the alphabet size of the variable being processed, from dimension-changing DPI. Under one-sided fixed-alphabet DPI, every measure with a continuously differentiable extension to a neighborhood of the closed probability simplex vanishes whenever one processed state has zero probability. If the processed alphabet is larger than the other alphabet, the measure is identically zero. If it has $n$ states and the other alphabet has at least $n$, every nonzero polynomial measure has degree at least $2n$ and requires at least $2n$ samples for unbiased estimation. Both bounds are attained by the Gram determinant. Under one-sided dimension-changing DPI, every family with the same boundary regularity is identically zero. These results yield sharp task bounds for multi-task peer prediction. For two agents with $n$ signals each, $2n$ tasks are necessary and sufficient for nontrivial expected payments that vanish for independent reports and cannot increase under memoryless stochastic garbling. Under these requirements, unequal signal-space sizes force the larger-alphabet agent's expected payment to vanish identically for any finite number of tasks.
Aswin Ramachandran, Jona Schulz, Maurus Derungs, Carlo Angelini, Tobias Meier, Raffaello D'Andrea
原文摘要
Progress in robot learning, particularly physical multi-agent reinforcement learning (MARL), is currently challenged by a limited availability of accessible, standardized benchmarks. While simulation-based MARL has produced remarkable emergent behaviors, from coordinated team play to complex tool use, translating these advances to physical systems remains difficult. A key barrier is infrastructural: physical platforms are often prohibitively expensive, require specialized facilities, or lack open mechanisms and designs for reproducibility. We introduce KlaskTron, an open-source, low-cost (<\$2500 USD), desktop-scale robotic testbed for adversarial MARL based on the dynamic tabletop game KLASK. The platform features dual CoreXY gantries with high-torque brushless motors, enabling the high-speed, precise manipulation required for competitive play. Crucially, we provide a complete ecosystem: hardware designs (CAD, BOM), a GPU-native digital twin in NVIDIA Isaac Lab for high-fidelity modeling and simulation, and a validated sim-to-real baseline using a learned neural actuator model. We demonstrate that this baseline enables zero-shot policy transfer, with trained agents exhibiting emergent adversarial behaviors. To ensure full reproducibility of our results, all project assets are released publicly: hardware CAD/BOM and assembly documentation at [https://idscethzurich.github.io/klask_hardware], inference software stack at [https://github.com/IDSCETHZurich/klask_software] and the simulation and MARL setup at [https://github.com/IDSCETHZurich/klask_rl].
Edge computing service deployment requires the Network Operator (NO) and Service Providers (SPs) to coordinate investment and resource allocation under uncertain demand. This paper develops an audit-based service management framework for daily coalition formation with fixed capacity and service-level agreement (SLA) reservations. A two-stage coalitional game separates infrastructure investment from operational participation. Demand follows a Geometric Random Walk (GRW), and SPs apply an ordered admission and suspension rule using reported marginal contributions. With the NO held fixed, the operational coalition value is submodular; mandatory reservations can produce negative Shapley payoffs. We derive a separable benchmark, compute operational Shapley allocations, and give a sufficient condition for convergence to an individual equilibrium. A probabilistic audit mechanism provides a conditional daily truthfulness guarantee under verified settlement and enforceable penalties. For five service classes over three years, the reference run gains 12.4% over static grand-coalition membership and recovers 94.6% of the full-information optimum. Across 30 runs, the median gain is 13.3%. Sensitivity tests examine policy parameters, capacity and alternative demand models. The gain depends on capacity and update order; abundant capacity can make the admission rule underperform static membership.
Petr Surovtsev, Pavel Korotaev, Alexander Nagaev, Alexander Kapitanov
原文摘要
Fingerspelling - the sequential signing of letters that compose proper names, technical terms, and out-of-vocabulary words - is an integral component of any sign language. Despite Russian Sign Language (RSL) being used by more than 150,000 native signers, no public fingerspelling dataset for RSL exists, in contrast to American, British, German, and several other sign languages. We introduce Znaki the first open RSL fingerspelling dataset: 37,252 video recordings (35.4 hours) of 1,593 unique sequences - proper names sampled from 16 domain categories - recorded by 68 native RSL signers (37 Deaf, 29 hard-of-hearing, and 2 certified interpreters) through a crowdsourcing pipeline with a two-stage signer qualification procedure and expert validation. The dataset covers all 33 Cyrillic handshapes, and its letter-frequency distribution matches the Russian National Corpus. As reference points for future research, we establish two baselines adapted from American Sign Language fingerspelling recognition: a Squeezeformer-based encoder-decoder, reaching a character-level accuracy ($1-\mathrm{CER}$) of 0.932 and a sequence accuracy of 0.596, and Fingerspelling-PoseNet, reaching 0.886 and 0.545, respectively.
Shangqing Wang, Syed Irtaza Haider, Shiwei Shen, Faezeh Motazedian, Rico Radeke, Laura del Rio Carazo, Frank H. P. Fitzek
原文摘要
The integration of electric vehicles (EVs) into urban mobility systems offers both opportunities and challenges for cities pursuing sustainability and energy efficiency. While simulation-based approaches have demonstrated the technical feasibility of advanced charging strategies, such as smart charging, bidirectional charging, and Vehicle-to-Grid (V2G), their real-world impact is often constrained by policy, business models, and stakeholder engagement. This paper introduces a purpose-driven technology transfer (PDTT) framework that systematically connects simulation insights with the needs of urban stakeholders. Leveraging a modular simulation architecture based on Simulation of Urban Mobility (SUMO), we evaluate the operational and sustainability impacts of advanced EV charging in complex urban scenarios, assessing indicators such as energy efficiency and CO2 emissions. The PDTT framework guides the translation of simulation results into actionable pathways for deployment, policy design, and market innovation, supporting co-creation with city planners, utilities, and communities. Our findings highlight the value of simulation as a diagnostic and engagement tool, and demonstrate how a multidimensional, stakeholder-centered approach can accelerate the sustainable adoption of advanced EV charging technologies in smart city contexts.