Generating physically based rendering (PBR) texture maps for a 3D mesh from a single reference image is highly ambiguous, as the observed appearance entangles material identity, surface reflectance, and illumination. While current generative methods produce realistic albedo maps, they still struggle to infer accurate metallic and roughness attributes for relighting. We present UltraTex, a cascaded multi-view framework that moves beyond appearance matching by integrating physical knowledge into multi-view diffusion-based PBR generation. UltraTex first jointly predicts albedo and world-space shading normals to generate more detailed albedo maps, and then predicts metallic-roughness (MR) maps conditioned on the generated albedo, reference image, and material semantic priors. To inject material semantics, a lightweight adapter transforms hidden-layer features of a vision-language model (VLM) into intrinsic material representation tokens, providing part-level knowledge anchors for MR estimation. To model material-light interaction, photometric representation alignment guides intermediate diffusion features toward ground-truth material renderings under controlled lighting. This makes metallic and roughness variations observable through highlight intensity, specular sharpness, and local reflectance contrast, providing stronger guidance for fine-grained material prediction without explicit lighting supervision. Extensive experiments show that UltraTex generates high-quality textures and outperforms existing open-source and commercial methods in visual fidelity, material accuracy, and physical plausibility.
We introduce $Z$-sight, a unified robot foundation model that couples vision-language-action (VLA) and generative latent world modeling. Video-based world action models (WAMs) show the value of joint future-and-action generation for control, yet inherit the heavy cost of video generators. Recent latent-space policies avoid pixel generation, yet typically align intermediate features or rely on frozen target encoders, decoupling future prediction from the VLM's native feature space. $Z$-sight instead predicts the future directly in the visual token space in which its VLM perceives the present. Coupling a foresight expert and a continuous action expert with the VLM in a mixture-of-transformers, $Z$-sight jointly denoises future spatial tokens and actions at inference without pixel reconstruction or external encoders. Trained on human video and robot trajectories, $Z$-sight achieves 81.9% success on SIMPLE (+9.6 points over the strongest baseline) and 73.3% across six Unitree G1 tasks (+20.0 points) at 110 ms per chunk, while remaining competitive on tabletop benchmarks.
World-action models jointly predict future observations and actions, yet reconstruction-oriented visual representations do not explicitly emphasize the spatial relationships and object-state changes needed for manipulation. We introduce Phoenix, which learns representations of scene structure for world-action modeling. Specifically, Phoenix Encoder enriches DINOv3's general visual semantics through dual-teacher distillation of geometric and spatiotemporal semantic knowledge. Next, CS-VAE causally aggregates and compresses Phoenix Encoder features into generative latents while preserving scene information and spatial layout. Finally, Phoenix jointly predicts these latents, complementary appearance latents, and robot actions, linking scene evolution with action generation. To study pretraining data efficiency, we curate approximately 39,000 hours of egocentric and robot manipulation data for representation learning and use the robot demonstration subset for WAM pretraining. Following benchmark-specific post-training, Phoenix achieves state-of-the-art performance across simulation and real-robot benchmarks spanning single-arm, bimanual, and dexterous manipulation. It demonstrates strong robustness to perturbations, achieving success rates of 94.8% on LIBERO-Plus and 72.4% on RoboTwin 2.0 C2R, alongside strong cross-embodiment transfer and real-world manipulation performance on Bench2Dex and RoboDojo-RealWorld, respectively. Pretraining-scale analysis shows larger performance gains from data scaling than a video-VAE-only baseline, indicating stronger scaling benefits and more effective data utilization. Moreover, Phoenix maintains high success rates when adapted with limited downstream demonstrations, demonstrating strong few-shot adaptation. Our code, pretraining data, and models are available on the project page to help the community reproduce our results.
In biomechanical systems, observable performance is often used as a proxy for underlying organization, although similar outputs may arise from different adaptive configurations. This study considers the vertical dimension of occlusion (VDO) as a constraint applied to an adaptive neuromechanical system. A single-case design in a patient with Parkinson's disease enabled repeated intra-individual gait observations under six occlusal probes. Three complementary analytical levels were examined: (i) an aggregated scalar score of observable performance, (ii) a conceptual dynamical systems framework, and (iii) an exploratory UMAP representation of 55 standardized biomechanical variables from 270 M1 observations. The revised Level 1 analysis showed that the relative ranking of OC2.5 and OC3 depended on score construction, while their scalar distributions remained close. The Level 3 embedding showed substantial overlap among all six probes and did not identify independently separated condition-specific clusters. OC2.5 and OC3 displayed limited centroid displacement but broad observation-level overlap. The principal result is therefore representational non-identifiability: neither the aggregated score nor the selected low-dimensional embedding uniquely identifies an occlusal-condition-specific system state. VDO is interpreted as a constraint parameter rather than a causal determinant. The findings are exploratory, model dependent, and non causal. They do not establish distinct physiological states, an optimal VDO, clinical thresholds, or diagnostic, predictive, mechanistic, or prescriptive validity.
该论文研究自适应数据分析(ADA)中随机性是否必要,证明在分析师计算能力无界时,任何确定性机制在约 n 个查询后就会失败,从而表明随机性是严格必要的。
cs.CR · cs.DS · cs.LG
查看原文摘要
The Adaptive Data Analysis (ADA) problem formalizes the challenge of preventing false discovery and overfitting when a dataset is repeatedly reused. Formally, our input is a dataset containing $n$ i.i.d. samples from an unknown distribution $\mathcal{P}$ over a domain $\mathcal{X}$, and our goal is to answer a sequence of $k$ adaptively chosen statistical queries with respect to $\mathcal{P}$. The main question is how many queries we can support (i.e., how large $k$ can be), primarily as a function of the number of samples $n$. This question has been intensively studied and is relatively well-understood for randomized mechanisms: there are computationally efficient mechanisms that support $k \approx n^2$ queries, and no computationally efficient mechanism can answer $k \gg n^2$ queries. In this paper, we address a fundamental question: is randomness necessary for ADA? Despite a decade of work on ADA, this question remains open. A folklore observation dating back to the initial works on ADA is that randomness is {\em not} necessary when the analyst is computationally bounded. Yet, the necessity of randomness against computationally unbounded analysts has remained elusive. Our main contribution resolves this gap in the information-theoretic setting. Perhaps surprisingly, we show that randomness is {\em strictly necessary} to answer a non-trivial number of adaptive queries: when the analyst is unbounded, any deterministic mechanism can be forced to fail after just $k = \tilde{O}(n)$ queries.
KV cache reuse accelerates LLM serving, but existing systems manage host DRAM and NVMe SSD as a tiered hierarchy where DRAM acts as an intermediate cache. Production LLM workloads break both premises of this hierarchy. First, bimodal reuse defeats DRAM temporal filtering, since workloads are dominated either by one-shot blocks with long reuse distances or by multi-gigabyte agentic sessions that exceed DRAM capacity, putting SSD on the critical read path. Second, DRAM staging and eviction-ordered layout bottleneck SSD restoration. Holding host staging buffers for in-flight prefixes causes contention under load that aborts prefetches, while incremental eviction scatters a session's blocks across the SSD and prevents large sequential reads. We present FlatCake, a flat KV storage architecture that exposes host DRAM and NVMe SSD as peer media in a unified 64-bit address space. Exploiting the fact that KV blocks are write-once, immutable, and reused together within a prompt or session, FlatCake places each block directly onto its target medium in session-contiguous, layer-major order at creation time and never migrates it. FlatCake streams coalesced SSD reads directly into a bounded GPU window via GPUDirect Storage and decouples DRAM and SSD request queues to prevent head-of-line blocking. We implement FlatCake in SGLang and compare it against host-staged (Mooncake) and GDS-enabled (Mooncake-GDS) tiered storage, as well as SSD-direct storage (Tutti). Across three production agentic, coding, and conversational traces with DeepSeek-V4-Flash, FlatCake reduces mean TTFT near the baselines' latency knees by $1.40\times$--$2.42\times$ over the fastest baseline. On the agentic Claude Code trace, FlatCake sustains a 91.6% cache hit rate under load (versus 71.1% for all three baselines) and delivers up to $2.16\times$ higher token throughput.
When a capable model and a human reviewer cooperate on the same decision, which human still adds value? We argue this is a question about a single cognitive trait, metacognitive calibration: the alignment between a reviewer's behavioral confidence and the actual correctness of their judgments. We develop a trait theory of reviewer value to make it precise. The theory makes five claims: value is governed by calibration rather than by activity, speed, agreement, or tenure; calibration leaves observable traces in behavioral telemetry; those traces are non-diagnostic on their own, because the fast, low-revision signature of a calibrated expert is indistinguishable from that of a confident novice until it is gated by outcomes; human calibration and a model's self-knowledge are the same construct, so the joint system should defer to whichever agent is better calibrated on each item; and individual calibration is causally identifiable only above the level of the single judgment. We ground each claim in a six-month production window of an AI-assisted accounting system (541621 reviewer events, 483 reviewers), and find that a purely behavioral expertise profile inverts the ground truth, scoring an overconfident cohort (5.2 percent correction survival) higher than genuine experts (99.2 percent); that reviewer quality is flat across tenure, so the trait is not learned on the job; and that the same calibration logic that separates good from bad reviewers, applied to the model, concentrates 92.5 percent of human effort on genuine errors. We close with the boundary the theory must respect: under majority-vote aggregation, individual causal value is unidentifiable.
AI systems are built across many organizations that contribute models, data, compute, capital, talent, and other assets. While the layers of the AI stack are relatively well understood, the relationships between actors and components across layers remains sparsely documented. In this work, we focus on the organizations that make up the AI supply chain: the complex network of actors that contribute to the development and deployment of AI. We motivate the examination of the organizational AI supply chain---highlighting concerns about concentration, power, chokepoints, cascading failures, fragmented accountability, and risk---then describe our pipeline for mapping the AI supply chain. Our pipeline yields our main contribution: a multilayered, large-scale dataset of actors and relationships in the AI ecosystem, annotated with dates, descriptions, types, tags, and more. This work takes a step toward systematically mapping and measuring the AI supply chain and its properties, cataloging how models, data, chips, infrastructure, capital, and even talent flow between organizations. This data enable critical examinations of industry relationships and yield insights into the political economy of AI development and concentration. We provide several illustrative analyses, including studying the most central actors, analyzing network concentration, and identifying communities and chokepoints. While our dataset is fairly comprehensive, it is also necessarily incomplete: public documents omit private deals, and coverage is shaped by search design. Together, the approach, data, and findings aim to serve as a resource for studying the AI ecosystem. The dataset is publicly available at https://huggingface.co/datasets/sarahcen/ai-supply-chain and can be explored interactively at https://aisupplychains.org/.
Grana: Compositional Reward Modeling for Long-Horizon Robot Manipulation
Zi-ang Cao, Kaiyuan Zheng, Yinzhen Xu, Haonan Duan, Jiasheng Gu, Fengyuan Hu, Nan Zhu, Kun Lei 等 17 位作者
AI 速读 基于摘要
Grana 是一种视觉-语言机器人奖励模型,通过开放词汇子任务描述识别未完成子任务并估计其连续进度,再组合子任务进度与已完成子任务信用来估计长时程任务的整体进度。
cs.RO
查看原文摘要
Robot reward modeling faces two challenges: generalizing across tasks and accurately tracking progress in long-horizon manipulation. General-purpose reward models demonstrate broad task transfer, but transfer from short-task pretraining to long-horizon tasks remains underexplored. Long-horizon progress models often rely on task-specific subtask classifiers or fixed action vocabularies, limiting the reuse of learned progress knowledge when adapting to new tasks. To address these challenges, we introduce Grana, a vision-language robot reward model that learns to identify unfinished subtasks and estimate their continuous progress using open-vocabulary subtask descriptions. We train a language-conditioned model to estimate progress in short-horizon tasks and individual subtasks of long-horizon tasks. For long-horizon tasks, we combine the selected subtask's predicted progress with credit for preceding completed subtasks to estimate whole-task progress. General progress prediction requires training the model across diverse tasks with progress supervision that accounts for imperfect execution. Elapsed-time proxies can misrepresent progress when relative subtask durations vary, or failures reverse local progress. We therefore introduce an annotation protocol that converts sparse subtask boundaries and action-quality annotations into dense progress targets, without requiring annotators to assign numerical progress scores. Pretraining with this supervision on 200 diverse, mostly short-horizon manipulation tasks enables zero-shot reward prediction on unseen tasks and improves progress prediction after adaptation to multi-minute, long-horizon tasks. Experiments demonstrate (i) generalizable reward prediction, including adaptation to five long-horizon tasks, (ii) benefits from pretraining and composing subtask progress estimates, and (iii) improved downstream policy performance.
Anticipating or Speculating? Making Futures Pay in AI Ethics
John P. Nelson
AI 速读 基于摘要
作者主张区分“预见”与“臆测”,并据此批评为医生与医疗AI判断分歧制定统一决策规则的做法。
cs.CY · cs.AI
查看原文摘要
Philosophers have recently been trying to develop uniform decision rules for instances in which physicians' judgments diverge from the recommendations of medical artificial intelligence (AI) systems. I argue that this effort is fatally flawed because the performance and appropriate use of algorithmic recommendation systems by physicians are highly context-specific, empirical medical matters rather than epistemological ones. In such instances, general decision rules can be stipulated only to the extent that they abstract away from relevant, indeed determinative, contextual circumstances. The alleged problem of physician-AI disagreement, and others like it, play into longstanding disputes in technology ethics about the viability and utility of envisioning possible futures: is this useful "anticipation" or wasteful "speculation"? I suggest that the two may be distinguished. Anticipation explores plausible futures to help inform present actions under irreducible uncertainty. Speculation abstracts away uncertainty, developing formal solutions to stipulated circumstances whose relationship to the real present is often quite unclear. Efforts to develop crisp philosophical solutions to physician-AI disagreement, among other problems, fall into the latter category. Technology ethics should remember that fruitful futures thinking must both begin from and return to the messy, uncertain present.
Predicting the consequences of an agent's actions requires modeling both the body's motion and the environment's response. Most video world models attempt to learn both from numerical actions, even though geometry and kinematics already specify the intended motion. We introduce Doppel, an embodiment-agnostic video world model conditioned on camera-aligned renderings of future body configurations and numerical action trajectories. By explicitly providing the intended motion, Doppel can focus on predicting object motion and interaction-driven changes in the environment. Rendered motion also provides a common visual interface across different embodiments and action spaces. To train on robot datasets without camera extrinsics, we recover camera poses and align rendered trajectories with recorded observations. We jointly train a single shared video backbone and conditioning adapter on eight datasets spanning eight robot and human embodiments, from single- and bimanual robot arms to a dexterous humanoid and human hands and bodies, using about 600 hours of video without embodiment labels. Across three robot-arm datasets, Doppel improves PSNR over action-only baselines by 1.00 dB on held-out trajectories and 1.20 dB under out-of-distribution evaluation, while predicting more consistent object motion. Training on all eight embodiments (Doppel-8) improves robot predictions over the three-robot model (Doppel-3) on four of six evaluation sets. Project page: https://doppelwm.github.io/.
Real-world harness engineering is driven by explicit and implicit user requirements, which guide development and validation. However, existing harness evolution methods often overlook implicit requirements and interaction needs, while using a single evolution agent to modify the harness and benchmark scores to select changes. This process often produces narrow patches and leads to benchmark-specific local optima. We introduce Agentic Harness Engineering Beta (AHE-Beta), a requirement-centered loop that brings real-world engineering practices into harness evolution. The loop consists of three steps: (1) User-driven evaluation lets simulated users raise their requirements as issues while they work with the harness over multiple turns. (2) Team-based modification implements those requirements as generalizable changes rather than task-specific patches. (3) Requirement-driven selection merges a change only when it resolves a verified requirement and passes review. Empirically, on held-out tasks, AHE-Beta raises pass@1 from 21.9% to 33.3% while using 40.8% fewer tokens than Base Harness, and achieves higher pass@1 than Claude Code, Meta-Harness, and AHE. In a blind human A/B test, real users prefer AHE-Beta in 31 of 48 valid pairs and rate it higher on all five dimensions, matching the directional preference of the user simulator. The human A/B results show that AHE-Beta achieves its goal of aligning harness evolution with real-world user requirements. Without further evolution, the gains transfer to GLM-5.3 and DeepSeek-V4.1-Flash with pass@1 improvements of 15.8% and 5.3% over Base Harness. Together, these results show that requirement-centered harness evolution is a more general method for improving agent capabilities.
In this paper, we propose a Unified Law for Visual Tasks, which could be further divided into Four Laws of Computer Vision, with the goal of enabling a single framework or method to uniformly train on and perform nearly all visual tasks. Our starting point is an inspiration and hypothesis: if the world information is encoded into image, which we call World Snapshots, then image processing is world processing, and image generation is World Snapshot generation. Furthermore, if all visual and visualization tasks can be reformulated as generation tasks, then a unified method, World Snapshot Models (WSMs), can perform the vast majority of visual tasks, much as large language models have largely unified natural language processing. This paper validates these laws across more than 1,000 trained and evaluated visual tasks (defined by distinct input-output mappings) and presents a unified training framework for future visual-task research and applications.
Agentic image restoration composes operations to handle mixed degradations, yet its toolbox often requires storing multiple restoration networks. We propose PrismRestore, which constructs a compact bank of callable restoration actions from a single frozen all-in-one model. Our approach builds on task-dependent yet overlapping channel responses in the shared model. An offline task-prior search uses activation and gradient statistics to group channels with similar response profiles, then searches task-specific modulation coefficients over non-representative channels while leaving representative channels unmodulated. The resulting prior bank enables explicit action selection and composition without additional restoration backbones or task-specific adapter modules. A state-conditioned policy learns to compose these actions using Group Relative Policy Optimization. Its objective separately normalizes fidelity and perceptual feedback within groups of trajectories sampled for the same input, and combines terminal evaluation with return-to-go derived from stepwise quality changes. Experiments on synthetic composite-degradation and real-world multi-task benchmarks show improvements over the evaluated all-in-one baselines and restoration quality competitive with multi-model agents. PrismRestore achieves these results with a single restoration backbone and faster inference than the evaluated agentic baselines.
Computer-Aided Design (CAD) underpins modern engineering, yet converting existing shapes into editable models still demands substantial expert effort. Most AI systems emit the entire CAD program in a single pass, never inspecting the intermediate geometry. In contrast, human engineers build a part feature by feature, checking after each operation what remains to be modeled. We introduce CADENA (Spanish for "chain"), a model that reconstructs a 3D mesh as a parametric CAD program, growing its sequence of operations one at a time and comparing the target with the currently predicted geometry at every step. We also address the lack of benchmarks for evaluating reverse-engineering methods on mechanical parts, introducing CADENA-Bench, a benchmark that measures performance across categories of mechanical parts. CADENA outperforms prior methods on CADENA-Bench and on the DeepCAD, Fusion 360, and MCB datasets. Code is available at https://github.com/zhemdi/cadena, model weights at https://huggingface.co/kulibinai/cadena, and CADENA-Bench at https://huggingface.co/datasets/kulibinai/cadena-bench.
Diffusion probabilistic models have become a cornerstone of modern generative AI, yet the mechanisms underlying their generalization remain poorly understood. In fact, if these models were perfectly minimizing their training loss, they would just generate data belonging to their training set, i.e., memorize, as empirically found in the overparameterized regime. We revisit this view by showing that, in highly overparameterized diffusion models, generalization in natural data domains is progressively achieved during training before the onset of memorization. Our results, ranging from image to language diffusion models, systematically support the empirical law that memorization time is proportional to the dataset size, consistent with a kernel-regression bound on the time required to fit the empirical score at low noise. Generalization vs. memorization is then best understood as a competition between time scales. We show that this phenomenology is recovered in diffusion models learning a simple probabilistic context-free grammar with random rules, where generalization corresponds to the hierarchical acquisition of deeper grammar rules as training time grows, and the generalization cost of early stopping can be characterized. We summarize these results in a phase diagram. Overall, our results support that a principled early-stopping criterion - scaling with dataset size - can effectively optimize generalization while avoiding memorization, with direct implications for hyperparameter transfer and privacy-sensitive applications.
Nowadays refinery optimization utilizes sheer amounts of data, which can be handled with modern Linear Programming (LP) software, but the interpreting and applying the results remains challenging. Large petrochemical companies use massive models, with hundreds of thousands of input matrix elements. The LP solution is mathematically correct, but simplifications are made in the model, and data supply errors may occur. Therefore, further insight is needed to trust the results. The novelty of the paper consists of enhancing the LP solver with a memory, which makes possible to find potential errors in the input data, finding unusual relations in the data on different decision levels, by introducing an anomaly detection tool. Analyzing the potential "anomalies" in the input and output of LP, the subsequent LP solutions can be improved. The Anomaly Detection tool proposed comprises three methodologies. A transformed version of the popular ECOD methodology is applied. New methods are proposed to handle high-dimensional data: choosing the most informative pairs. Then, this is used alongside two 2D Anomaly Detection algorithms, revealing several business opportunities and data supply errors in the MOL refinery scheduling and planning architecture. The proposed tool learns connections in past data and runs very efficiently on present data revealing unusual connections sorted by their importance. Up to this step, optimization and Anomaly Detection happened automatically, and thereafter, decision makers may take a look at the most relevant anomalies found, and decide how to further improve the subsequent LPs. The tool also has periodic retraining built in.
We study two-layer neural networks trained by stochastic gradient descent (SGD) on a multi-index teacher model, focusing on how first-layer parameters orthogonal to the teacher subspace affect generalization. We decompose the first-layer weights into a signal component aligned with the teacher and an orthogonal complement, and analyze their joint SGD dynamics together with the second layer in the proportional limit. We show that although SGD updates to the complement are initially isotropic, the training dynamics induce teacher-dependent spectral spikes in the complement Gram matrix. This emergent anisotropy provides a concrete mechanism by which representation variance is amplified, and generalization degrades. To isolate this effect, we introduce Decomposed Dynamics, an analytically tractable surrogate whose asymptotic test-error dynamics coincide with those of SGD. In simulations with a ReLU student, freezing the directions of the orthogonal complement improves generalization relative to training them. Finally, we show that the signal component rapidly aligns with the second-layer weights, revealing strong cross-layer coupling even in this shallow setting. Together, our results provide a dynamical explanation for how feature learning outside the teacher subspace harms generalization.
Reconstructing an editable CAD model from a 3D shape remains a challenging engineering task. Existing methods can propose CAD operations, but no single source of proposals works equally well across different part geometries and stages of reconstruction. We introduce CADFather, an autonomous agentic system that coordinates complementary tools to recover parametric CAD programs from 3D meshes. A vision-language assistant inspects renders of the target and intermediate reconstructions, then decides which candidate CAD programs to extend, which tools to invoke, how many proposals to generate, and when to finish. Learned and algorithmic tools propose CAD operations, while numerical optimization refines the parameters of existing programs. Proposed or refined programs are executed and evaluated to provide feedback for subsequent decisions. The agent maintains alternative candidate programs for each target part and preserves the best valid result throughout reconstruction. CADFather uses pretrained generation and assistant models without additional training. We evaluate reconstruction quality and execution validity on the full DeepCAD, Fusion360, and MCB test sets, as well as on CADENA-Bench, CADBench, and BenchCAD. We additionally analyze computational cost and the trade-off between cost and reconstruction quality.
We give a parallel algorithm that maintains a one-sided $(2+\varepsilon)$-approximation of all core numbers under batches of edge insertions and deletions. For fixed $\varepsilon>0$, the expected amortized work is $O(\log n)$ per edge update, and each batch has deterministic polylogarithmic depth. This improves on the previously known $(4+\varepsilon)$-approximation and $O(\log^2 n)$ amortized work bounds. This approximation factor approaches the factor-two limit for polylogarithmic update and query time under the OMv conjecture. Our algorithm is based on assigning each vertex a "pressure" and splits each edge's unit load smoothly between its endpoints according to their pressures. After each batch of updates, our dynamic algorithm adjusts the pressures of vertices in parallel to restore balance between the pressures and loads. We show that by maintaining balanced pressures we can obtain good approximations of core numbers. Our construction also gives a $(1+\varepsilon)$-approximation of the densest-subgraph value with logarithmic expected amortized work and supports batch-dynamic maximal matching, clique counting, forest decomposition, and coloring in low-arboricity graphs with improved or matching work bounds. We give a practical implementation of our batch-dynamic coreness algorithm and find that it is able to achieve smaller empirical error than the state-of-the-art parallel batch-dynamic implementation, while running faster. On six real-world graphs, it runs $32\times$ faster than that implementation in geometric mean and $4.1\times$ faster than a state-of-the-art heuristic baseline. On three graphs with up to 1.8 billion edges, it is $9.2\times$ faster than the heuristic baseline.
Overhead aerial imagery feeds dense-prediction pipelines that are routinely trained in one geographic region and deployed in another, yet no benchmark measures how shadow detection models transfer across regions. Existing overhead shadow datasets either lack a geographic transfer protocol or rely on automatically generated labels. We introduce ShadowTransfer, a benchmark of 4,500 human-verified shadow masks across three U.S. cities at two native resolutions, paired with a leave-one-city-out evaluation protocol. Under vanilla training, holding out the test city costs 3.5 to 7.8 mIoU across four architectures, the gap remains significant when the training set grows from 225 to 900 images, and six of eight adaptation methods degrade more than vanilla training on average. We propose a three-probe framework that diagnoses where and how the transfer breaks: the failure operates in feature space, beyond the reach of pixel-histogram matching, shows up as a confidence collapse on shadow pixels while background confidence stays intact, and is recoverable by retraining different architectural components depending on the model: encoder for some, decoder for others. These findings replicate on INRIA building footprints, suggesting they describe geographic transfer beyond shadows. Guided by the diagnosis, we build Spectral Information Bottleneck (SIB), a plug-in module with one component per finding. Across the 9 (architecture, held-out city) cells, SIB has the smallest mean gap to the within-city upper bound of the ten methods we evaluate (6.42 mIoU), level with FDA. The benchmark, LOCO protocol, code, and trained models are released to support reproducible evaluation of geographic transfer in overhead dense prediction.
Specification mining extracts temporal rules that describe relationships between events in recorded system traces. Applying these rules in distributed monitoring systems is more difficult because observations come from different nodes, whose clocks may not be synchronized, and events may be lost or reordered during transmission. In this paper, we study how to aggregate the coverage of mined rules across observers, determine whether their temporal windows are needed during monitoring, and characterize their detection under event loss. Our evaluation uses IPv6 packet traces collected from 21 receivers across four continents. We show that local mining followed by global confirmation recovers centralized rule coverage under suitable conditions while substantially reducing communication overhead. In contrast, merging traces with unaligned clocks can distort coverage by introducing incorrect temporal matches. A window-gain criterion identifies rules that require temporal state, while our loss analysis and broker experiments demonstrate why contiguous losses can be more damaging than independent losses. The results identify separate requirements for aggregating observations, selecting monitor state, and handling event loss.
The final processing step in compression systems is generally an entropy coding stage, which converts symbols into a bit stream based on their probability distribution. A distinct class of entropy coding methods operates not by mapping input symbols to discrete codewords but by operating on intervals or ranges. This approach enables a more accurate approximation of the source entropy, particularly for sources with highly skewed or varying symbol distributions. Representative techniques in this category include traditional arithmetic coding, range coding, and methods based on asymmetric numeral systems (ANS). Due to their near-optimal compression efficiency, these interval-based entropy coders are integral to modern data compression systems. The complexity of these methods depends mainly on three processing steps: the core routines of encoding and decoding, the interval-based determination of the correct symbol at decoder, and the efforts of keeping updated with respect to the varying symbol distribution. The interval-based symbol determination at decoder typically demands for a searching procedure. Previous literature showed that the search can be replaced by a table-based approach with only O(1)-complexity but having the side-effect that the adaptation of the multi-symbol statistic becomes infeasible because of the high time-consumption of adapting the table. We compare different approaches to fast adaptive coding of multi-symbol alphabets and propose a new adaptation process using a special ring-buffer technique enabling the adaptive table-based decoding procedure in combination with the replacement of a division by a bit-shift operation at encoder and decoder core routines. The proposed technique is faster than alternative approaches for alphabets from about 12 to 64 different symbols when comparing the overall encoder+decoder time in a practical implementation.
Equivariant Modeling with Polynomial Features via Minimal Homogeneous Generators
Michael Liaofan Liu, Joe Kileel
AI 速读 基于摘要
提出一个基于极小齐次生成元构造等变多项式特征的通用框架,并配套自动计算与拟合的Python包。
cs.LG · stat.ML
查看原文摘要
We present a general framework for modeling and learning equivariant functions between finite dimensional semisimple real group representations, unifying constructions developed across many application areas. Our model is a linear combination of equivariant polynomial features whose coefficients are arbitrary learnable functions of invariant polynomial features. We take as features minimal homogeneous generators for the invariant algebra and the covariant module over the invariant algebra, up to prescribed degrees. We provide a Python package that automatically computes these features and fits the resulting models to training data. The algorithm for feature computation requires explicit isotypic decompositions of the input and output representations, along with explicit Clebsch-Gordan decompositions of the tensor product of irreducible representations, and achieves speedups over alternative methods in computational algebra. Numerical experiments in scientific modeling demonstrate the full learning pipeline and yield accurate predictors with few trainable parameters, using the model both for end-to-end prediction and within layers of larger architectures.
CAD practitioners often work with proprietary parts that they are unwilling to share with external services. For individuals and small companies, bringing this work in-house also requires computational resources they can afford. We introduce CADMAGEDDON, an open-source agent that reconstructs editable CAD from meshes and can run locally on single-GPU systems and current-generation laptops. The method decomposes this complex task into a chain of simple visual decisions about what to construct and what to correct. Geometric tools support precise construction, allowing the agent to focus on visual reasoning. Using the mid-sized Qwen3.8-27B model, CADMAGEDDON outperforms prior approaches, including systems based on frontier models, in reconstruction quality on all the main benchmarks. CADMAGEDDON integrates with FreeCAD, bringing its reconstructions into practitioners' existing CAD workflows. Code: https://github.com/kulibinai/cadmageddon
User simulators offer a scalable way to evaluate interactive agents, but how to assess their quality remains underexplored. In particular, it is unclear whether existing user simulator evaluation frameworks reliably detect meaningful failures in simulated user behavior. This uncertainty makes it difficult to interpret simulator evaluation. To understand what a high benchmark score tells us about simulator quality, we (meta-)evaluate the metrics themselves. To instantiate this meta-evaluation, we deliberately perturb simulated replays to induce failures such as altered goals, omitted constraints, premature stopping, and acceptance of broken work, then test whether the metrics penalize these failures. Evaluation methods from three prior studies detect only 5-6 of the 10 perturbations; their combined scores even favor populations instructed to accept broken work over unmodified controls. Motivated by these gaps, we propose a new evaluation suite that combines utility of user simulators for evaluating agents and two measures of fidelity to real users: population-level fidelity assesses collective interaction patterns, while trace-level fidelity assesses the preservation of individual users' goals, constraints, and persona. Under the same detection criterion, our evaluation suite detects all 10 perturbations. Evaluating six user simulators on 351 real multi-turn sessions in which 348 distinct users across 23 languages and agents work together to create and refine artifacts such as full-stack web apps, games, design assets, and automated workflows, we find that population-level measures can miss when a simulated user pursues the wrong goal, ignores a constraint, or behaves out of character, highlighting the need to assess fidelity at comprehensive granularities.
General-purpose object detectors lose accuracy on UAV footage, where targets span only a handful of pixels and onboard compute is limited. Prior work composes independently-validated architectural techniques into one detector, assuming gains reported in isolation transfer once combined. We stress-test that assumption directly. LAF-YOLOv10 integrates four techniques into YOLOv10n: a Partial Convolution C2f (PC-C2f) backbone block, an Attention-Guided Feature Pyramid Network (AG-FPN), a P2 detection head replacing the large-object P5 head, and Wise-IoU v3 regression, asking whether their combined effect matches what each contributes alone. We train LAF-YOLOv10 three times (seeds 42, 123, 256) on VisDrone-DET2019, benchmark against unmodified YOLOv10n, and use TIDE error decomposition, per-category breakdown, per-component ablation, attention/loss comparisons, zero-shot transfer to UAVDT, and held-out/test-dev evaluation to localize where the combination succeeds or fails. Composability does not hold here. LAF-YOLOv10 reaches 24.0+/-0.4% mAP@0.5 at 2.14M parameters, 7.8 points below YOLOv10n (31.8%), a deficit that transfers to UAVDT (-10.0 points) and is confirmed by held-out and test-dev evaluation (23.5%, 22.5%). Background false positives, localization error, and duplicate detections move in the direction AG-FPN and Wise-IoU were designed to push. Ablation traces the deficit to a specific source: the P2/-P5 head swap costs 2.5 points independently plus a 2.5-point interaction penalty when layered onto a backbone weakened by PC-C2f, whose own 2.0-point loss is consistent with a partial pretrained-weight transplant (73/150 backbone tensors transfer). The failure is attributable to a specific interaction, not the four components individually. Composability must be verified directly, not assumed. Code/checkpoints: https://github.com/Mudassiruddin7/Small-Object-Detection-in-UAV-Imagery.
Turning a CAD model into a machining program requires both process knowledge and geometric precision. Language models can generate such workflows by writing geometry and toolpath code for each part, but this makes numerical execution part of the model's task. We introduce MillBill, an agent that separates machining decisions from their geometric execution and verification. Given a boundary-representation part, stock, and material, a locally hosted, mid-sized open model proposes operations and cutting parameters for indexed three-axis machining. A geometric harness generates toolpaths, simulates material removal, and returns feedback for plan refinement. Setup planning accounts for what the available cutters can physically reach, rather than visibility alone. We evaluate the resulting programs by the fraction of the target surface area within the specified tolerance after simulated machining. With Qwen3.8-27B, MillBill outperforms frontier-model approaches that construct their own geometry and toolpath workflows on the evaluated machining tasks. Code: https://github.com/kulibinai/millbill
In recent years, Rectified flow (RF) has gained considerable popularity largely due to its generation efficiency and state-of-the-art performance. In this paper, we investigate how well RF automatically adapts to the intrinsic low dimensionality of the support of the target distribution to accelerate sampling. We show that, using a carefully designed end-point adapted time-discretization scheme and with sufficiently accurate drift estimates, the RF sampler enjoys an iteration complexity of order $O(k/\varepsilon)$ (up to log factors), where $\varepsilon$ is the precision in total variation distance and $k$ is the intrinsic dimension of the target distribution. In addition, we show that the denoising diffusion probabilistic model (DDPM) procedure is equivalent to a stochastic version of RF by establishing a novel connection between these processes and stochastic localization. Building on this connection, we further design a stochastic RF sampler that also adapts to the low-dimensionality of the target distribution under mild requirements on the accuracy of the drift estimates, and also with a specific time schedule. We illustrate the efficacy of newly designed time-discretization schedules with simulations on the synthetic data and text-to-image (T2I) data experiments.
Large Language Models (LLMs) now perform strongly on Alan Turing's 1950 Imitation Game, so we propose a harder challenge, the Personal Imitation Game, which asks whether a machine can replicate a specific individual in conversation rather than a generic human. Testing with 30 participants, we establish a human baseline and report the identity signals interrogators rely on: stylometric features, personality traits, and biographical facts. To demonstrate the framework, we design an LLM-based system that plays the game for one of the paper's authors, drawing on an autoethnographic dataset spanning ten years of personal digital communication. In a 20-person deception study, interrogators choose our system over the author's own friends in 7 of 10 rounds, and the author over our system in 13 of 20; neither rate is distinguishable from chance at this sample size. We consider design improvements, applications of personal AI agents, and implications for identity and privacy.