Computer Vision
TAPe+ML: A Compact Structured Representation for Multi-Task Computer Vision
Sergey Kurinov and Alexey Upatov
We present TAPe+ML v3, a compact computer
vision system based on TAPe (Theory of Active Perception), a structured representation that encodes relations among perceptual elements before recognition. Instead of operating directly on pixel tensors, the system uses a shared TAPe representation and a modular recognition architecture for image classification, object detection, and instance
segmentation. TAPe+ML v3 combines background and contour processing, local object localization, prototype-based classification, and a coordinator for specialized submodels. Across the reported experiments, it uses fewer than 100,000 parameters. On COCO object detection, it obtains 84.7 mAP50 and 65.3 mAP50-95. On COCO instance
segmentation, it obtains 80.7 mask mAP50 and 58.4 mask mAP50-95. In classification experiments, it reaches 92 percent validation accuracy on Imagenette under an identical-training comparison with a raw-pixel baseline, and 89.9 percent Top-1 accuracy on ImageNet-Real. We also evaluate compactness in
video scene detection and adaptation under distribution shift in an industrial pilot. The results suggest that shifting part of the modeling burden from network parameters to a structured input representation can support compact multi-task
vision systems with reduced data,
memory, and compute requirements.
Read the full abstract ▾
Machine Learning · Cross-listed from physics.ins-det
Automated Physics-Informed Neural-Networks-Based Calibration of Highly Segmented Silicon Telescopes
M. Rejmund, A. Lemasson, P. Morfouace et al. (+3 more)
Transfer and multi-nucleon transfer reactions are essential tools for probing
nuclear structure and reaction dynamics, requiring precise determination of the identity, energy, and emission angles of reaction products. The increasing granularity of modern silicon
telescope arrays enhances experimental capabilities but challenges detector calibration, as conventional channel-by-channel approaches become inefficient and difficult to scale. In this work, we present a fully automated, physics-informed calibration framework based on neural networks, specifically designed for highly segmented silicon detector arrays. The method formulates calibration as a global optimization problem, in which detector gains and geometrical corrections are determined simultaneously by minimizing the width of the reconstructed excitation energy under two-body kinematics constraints. The approach relies exclusively on experimental data and well-established physical principles, without requiring explicit modeling of detector response. A distinctive feature is the use of multiple neural network sub-models sharing a common loss function with embedded physics constraints, enabling coherent and self-consistent calibration across all detector channels. This strategy ensures scalability, robustness, and reproducibility, making it particularly suitable for next-generation detector systems with increasing complexity. The performance of the method is demonstrated using experimental data from the Particle-Identification Silicon-
Telescope Array (PISTA) in high-resolution fission studies in inverse kinematics. The results show excellent agreement with theoretical kinematics, high-quality particle identification, and a significant improvement in calibration efficiency. The proposed framework provides a general and adaptable solution for the calibration of complex detector systems in modern
nuclear physics experiments.
Read the full abstract ▾
All authors M. Rejmund, A. Lemasson, P. Morfouace, D. Ramos, J. Taieb, J. D. Frankland
Data Structures and Algorithms · Cross-listed from math-ph
Critical and near-critical influence bounds for ferromagnetic Ising models
Yan Ru Pei
For a
ferromagnetic Ising model on a graph of maximum degree $\Delta\ge3$, we prove a bound of order $\sqrt n$ on every row of the influence
matrix at the tree uniqueness threshold. The estimate is uniform in the degree, the external fields, and all pinnings. More generally, if the couplings are bounded by $\beta$ and $\varepsilon=((\Delta-1)\tanh\beta-1)_+$, the bound is $C(\sqrt n+n\varepsilon)$. The proof combines a pointwise cavity bound with a positive-series magnetization tilt and the field comparison
theorem of Ding, Song and Sun. The critical estimate removes the logarithm in recent general graphical bounds for the
ferromagnetic case. As a consequence, zero-field single-site Glauber dynamics mixes in
polynomial time throughout the supercritical window $\varepsilon=O(\sqrt{\log n/n})$, with the
polynomial degree depending on the window size.
Read the full abstract ▾
Robotics
AthenaZero: A low-inertia, bimanual robot for dynamic manipulation
Andrew S. Morgan, Gregory Xie, Capprin Bass et al. (+23 more)
AthenaZero is a bimanual manipulator designed to minimize inertia without compromising control authority. By utilizing quasi-direct drive actuation and transmission remotization techniques, the system achieves an effective endpoint mass comparable to that of a human—about an order of magnitude less than conventional
robot manipulators. This characteristic, combined with its inherent torque transparency, makes AthenaZero exceptionally well-suited for dynamic manipulation. We describe the methodology} that led to this design and demonstrate the
robot's capabilities on three baseball-inspired tasks: throwing, catching, and batting, which showcase complex interactions on human-comparable timescales where milliseconds matter. AthenaZero was capable of throwing at speeds in excess of 30 m/s, with catching and batting at speeds in excess of 14 m/s over a short 7.3 m distance. Batting practice and a game of catch were subsequently performed in
robot-to-
robot and human-to-
robot variations, showcasing the efficacy and adaptability of our system in tasks that require high acceleration.
Read the full abstract ▾
All authors Andrew S. Morgan, Gregory Xie, Capprin Bass, Rachel Thomasson, Chunpeng Wang, Erfan Shahriari, Harrison Busa, Oluwaseun Araromi, Joseph Aronov, Michael Burgess, Velin D. Dimitrov, Matthew A. Estrada, Faris Hamdi, Samuel Kendig, Taeyoon Lee, Jose Oscar Mur-Miranda, Emma Sommers, Paul Titchener, Margaret Wang, Achu Wilson, Mark Yeatman, Osman Dogan Yirmibesoglu, Alfred A. Rizzi, Annan Mozeika, Nicolas Rojas, Lael Odhner
Mathematical Software · Cross-listed from q-bio.OT
Flash-Radiomics: A Scalable Hybrid CPU-CUDA Engine for Standardized Scalar Radiomics and Accelerated Spatial Mapping
Shanli Ding, Yiyi Hu, Ziyu Fu et al. (+5 more)
Background and Objectives: Spatial mapping retains the spatial distribution of radiomic features, but computational cost and fragmented software limit its use. We developed Flash-Radiomics with scalar extraction and spatial mapping, a central processing unit (CPU) backend, a hybrid Compute Unified Device Architecture (CUDA) backend, consistent feature names, and Hierarchical Data Format version 5 (HDF5) storage. Methods: We evaluated Image Biomarker Standardisation Initiative (IBSI) compliance, CPU-CUDA concordance, and end-to-end processing time. Compliance testing included 825 chapter 1 (IBSI-1) tests covering 165 high-consensus features and 323 chapter 2 (IBSI-2) tests with numerical references. Concordance testing included 1,148 scalar pairs and 93 spatial-map pairs. End-to-end processing time was measured five times per input volume of interest (VOI) size. Comparisons included the Medical Image Radiomics Processor (MIRP) and PyRadiomics for 102 shared scalar features and PyRadiomics for 93 shared spatial maps. Results: Both backends passed all 1,148 IBSI tests, and all paired results were concordant. At the largest scalar input, CPU required 76.343 s and hybrid CUDA 81.915 s; CPU was 4.7 times faster than MIRP and 190.7 times faster than PyRadiomics. At the largest spatial input completed by both backends, hybrid CUDA reduced processing time by 68.6% relative to CPU (79.280 versus 252.791 s). At PyRadiomics' largest completed spatial input, hybrid CUDA was 84.9 times faster. Conclusions: Flash-Radiomics unified standardized scalar extraction, spatial mapping, concordant CPU-CUDA results, and HDF5 storage. CPU processing time was similar or shorter for scalar extraction, whereas hybrid CUDA was faster for spatial mapping under the tested conditions.
Read the full abstract ▾
All authors Shanli Ding, Yiyi Hu, Ziyu Fu, Chia-Hsin Lin, Ruihan Luo, Jaehee Chun, Xinyue Zhang, Osama Mawlawi
Hardware Architecture
CovR: Coverage-Aware Hardware Verification via Reasoning-Guided Reinforcement Learning
Manar Abdelatty, Maryam Nouh and Sherief Reda
Design verification remains one of the most resource-intensive stages of hardware development, often consuming up to 70% of the total design effort. While recent work has explored using Large
Language Models (LLMs) to automate testbench generation, most existing approaches focus narrowly on functional correctness, overlooking the critical aspect of coverage quality. To bridge this gap, we present CovR, an agentic framework for automated testbench generation that combines self-reflection loops with simulation-based feedback to maximize coverage. Using this pipeline, we construct a large-scale
dataset of 16,514 natural specification RTL
reasoning testbench tuples with a strong teacher model, enabling coverage-aware supervision. Building on this, we propose a reinforcement learning (RL) framework tailored for coverage-driven testbench generation, leveraging tool-derived rewards from simulation and coverage feedback to optimize a student model. Experimental results show that the CovR finetuned model achieves 93.81% cov@10 on VerilogEval and RTLLM V2.0, and 87.76% cov@10 on CVDP, outperforming state-of-the-art approaches by 7.97% and 3.59%, respectively. Furthermore, deploying the finetuned model back into the agentic refinement pipeline further improves cov@10 to 94.27% on VerilogEval and RTLLM V2.0 and 91.39% on CVDP. Moreover, when integrated as a plug-in stimulus engine for full verification workflows, CovR improves coverage by 18.95% and mutation detection score by 1.19%, while revealing 4.46% undetected failures, highlighting the importance of optimizing for coverage in LLM-based hardware verification.
Read the full abstract ▾
Multiagent Systems
Message capacity and claim wording set the transition points of collective truth-finding in language-model networks
Makoto Fukushima
Whether human or large
language model (LLM), an
agent in a discussion reads only a few of the others' contributions, bounded by cognition, context, or cost. LLM collectives can settle on a wrong consensus even when a majority starts out correct; we ask how far that reading bound alone decides the outcome. We model the bound with one number, the message capacity, which sets how many of the others' messages an
agent reads, and generate the communication network from it. Over 31,824 randomized queries, we found that an 8-billion-parameter model's judgment of a claim effectively reduces to a logistic function of a weighted sum of its inbox, the update rule of a
stochastic binary
neuron with divisively normalized weights. From these weights and the network's degree statistics alone, the wrong consensus should become unreachable from any start once
agents read, on average, fewer than 6.4 of their 31 sources. In 1,414 episodes with assigned starts the prediction failed: the correct side won in fewer than 50% of episodes from every start, and in only 28-45% when 75% of
agents started correct. The failure traces to the field, the threshold that a claim's wording sets for the
agent's answer before any message is read: the experimental claims' fields lay below the calibration mean, and with each claim's own field the same weights reproduce the outcomes. Reversing the wording showed that the threshold follows what a claim asserts, not whether it is true. On a second 8B model the pipeline predicts claim-dependent bistability; transition points appeared where computed, and an eight-claim calibration matched in 15 of 16 conditions. At 70B the assertion bias is not detected. Thus a collective's fate is largely set by two single-
agent measurements: the threshold a claim's wording sets, and the message capacity that sets the transition point.
Read the full abstract ▾
Artificial Intelligence
What Do We Expect from LLMs? Mapping the Design of LLM Benchmarks
Chao Wang
Benchmarks are central to how progress in large
language models (LLMs) is assessed and communicated. Yet model rankings alone reveal little about how evaluation requirements themselves are changing. The expanding variety of
benchmarks offers another perspective: what researchers expect LLMs to do, and what they count as successful performance. We systematically map 14,767 papers introducing or updating evaluation resources from arXiv submissions between January 2022 and August 2026. Using staged screening and automated full-text coding, we examine changes in target systems and domains, evaluation materials and conditions, and scoring mechanisms. The collection shows growing emphasis on action, interaction, and professional applications, while established and newer design elements frequently coexist. Model participation also develops unevenly: LLM-based scoring grows within both
agent and non-
agent groups, whereas model-generated materials show no comparable sustained increase in recent cohorts. These findings illuminate how public research translates capability expectations into concrete tests and criteria for success. As AI participates in constructing tests, performing tasks, and judging responses, they also raise a question: does expanding evaluation provide more independent evidence, or risk reproducing the preferences and blind spots of its participating models?
Read the full abstract ▾
Artificial Intelligence
BioPhys-Bridge: A Benchmark for Interdisciplinary Scientific Reasoning in Physics-Grounded Biological Research
Qingyang Xu
Language models face unique challenges in analyzing interdisciplinary scientific research literature. In biophysics research, faithful answers require grounding observed data in source evidence, interpreting it through a quantitative physics model, and linking it to a biological mechanism. To address this challenge, we introduce BioPhys-Bridge, a novel
benchmark dataset for evidence-grounded scientific
reasoning over biophysical literature. Each case contains evidence blocks, stable evidence IDs, quantitative values, units, equations, assumptions, mechanisms, and next decisions as grounding targets for question answering (QA) and
retrieval-augmented generation (RAG). The initial release contains 500 cases, 1,517
agent-facing tasks, and covers six biological domains and nine physical model families, including three sparse families reserved for future expansion. We enforce strict quality gates for all cases in schema, evidence-integrity, quantitative-grounding, source-license, duplicate, unit-normalization, with domain expert review and annotation for 81 cases. Preliminary evaluations show that DeepSeek-V4-Flash obtain the highest evidence-ID $F_1$ score (0.360), followed by Qwen3.7-Max (0.316) and GPT-4o-mini (0.294). BioPhys-Bridge is an interdisciplinary
benchmark for evaluating attribution, faithfulness,
hallucination reduction, and biological experiment design with complex, multi-step scientific
reasoning. Future works will increase the size and complexity of the
dataset and perform comprehensive evaluations. Code and data are available in the GitHub repository and on Hugging Face.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from eess.IV
Optimal Transport Metric Learning for Feature Alignment in Partially Supervised Segmentation
Dakini Mallam Garba and Salim Abdou Daoura
Multi-organ
segmentation is often challenged by partially annotated
datasets and domain shifts across different imaging sources. To address these limitations, we propose a two-stage learning framework that efficiently leverages partial supervision. In the first stage, the model learns from available annotations to produce accurate
segmentations of annotated organs, establishing robust feature representations. In the second stage, we introduce learnable organ prototypes and a Sinkhorn-triplet loss to enforce organ-wise feature consistency across
datasets. This encourages latent
embeddings of the same organ to remain close, while increasing separation between different organs, even when annotations are missing. Our approach achieves performance comparable to state-of-the-art methods on the BTCV
dataset, while remaining computationally efficient. By explicitly aligning feature distributions rather than relying solely on pseudo-labels, the framework effectively mitigates domain shift, making it particularly suitable for medical image
segmentation tasks with limited annotation resources.
Read the full abstract ▾
Machine Learning
Beyond the Previous Layer: Residual Predictive Structure in Sparse MoE Routing
Hao Li, Yasuyuki Tahara and Yuichi Sei
Sparse mixture-of-experts models route each
token through a sequence of expert selections. We ask whether the immediately preceding selection adequately summarizes this trajectory for predicting the next router. Using frozen OLMoE and JetMoE models, we measure the held-out predictive gain from earlier expert selections while retaining the most recent selection as a common baseline. In OLMoE, extending the history from one to eleven layers raises router-logit $R^2$ from 0.59879 to 0.66544. A preregistered JetMoE replication yields four-layer gains of 0.14275 and 0.20528 at two target depths, with paired bootstrap intervals above zero. These gains survive nonlinear decoding: adding history to a small multilayer perceptron improves $R^2$ by 0.17137 and 0.21861, whereas nonlinear decoding of the recent state alone adds 0.00139 and 0.00936 over a linear probe. Parameter-matched controls preserve the advantage, and cross-fitted history residuals predict target residuals with $R^2$ of 0.20549 and 0.23556. These findings identify residual predictive structure in expert-selection trajectories beyond adjacent-layer persistence.
Read the full abstract ▾
Networking
Analysis of Measurements in a Microwave Oven with High Level of Non-Ionizing Radiation (NIR) Before and After Repair
Júlia da L. A. Silva, Vicente A. de Sousa Jr., Marcio E. C. Rodrigues et al. (+5 more)
Non-ionizing radiation is the subject of several applications, such as the microwave oven, which is very popular. Its use is characterized by direct user operation and reduced distances between users and the NIR source. Therefore, the level of human exposure in this scenario deserves
attention as it is a potential cause of biological effects. In this sense, this paper presents the measurement of a microwave oven with a high level of NIR (above the limit defined by the standard) before and after two types of suggested repairs: the use of epoxy putty (Repair 1) and the replacement of the external surface of the oven (Repair 2). The definitions of measurement methodology, based on FDA and INMETRO standards, are presented and discussed. Finally, we describe an analysis of the efficiency of these repair methods aiming at the safety operation.
Read the full abstract ▾
All authors Júlia da L. A. Silva, Vicente A. de Sousa Jr., Marcio E. C. Rodrigues, Fred S. R. Pinheiro, Gutembergue S. da Silva, Halysson B. Mendonça, Fernanda E. S. Galdino, Ricardo Q. de F. H. Silva
Human-Computer Interaction
AI Mediators Regulate Emotion and Create Value in Disputes
James Hale and Jonathan Gratch
In conflict and disputes, especially,
emotion acts as a salient force in influencing outcomes. Prior work shows negative affect can obstruct collaborative behaviors, which typically lead to “win-win” outcomes. Thus, some suggest mediators may help regulate
emotion and achieve joint gains. With the proliferation of AI, we posit LLMs may perform well at this task, with the added benefit of better accessibility compared with a human mediator. To examine the effectiveness of AI versus novice human mediators, we conduct a between-subjects experiment, where participants engage in a dispute mediated by a human, AI, or no mediator. We first analyze how well the mediators regulate
emotions within a dispute – finding AI mediators perform significantly better than humans at reducing negative
emotion. We next examine whether AI mediators facilitate disputants better realizing joint gains in disputes with high integrative potential (IP) – we find a marginally significant interaction between IP and condition (AI versus human), indicating LLMs may outperform humans at aiding disputants realize joint gains. Lastly, we perform an analysis of the messages the mediators sent, finding the AI sent significantly more messages suggesting trade-offs compared to the humans.
Read the full abstract ▾
Data Structures and Algorithms
A Better-Than-$3$ Approximation Algorithm for Demand Matching via Knapsack Intersection LP and Contention Resolution
Michel X. Goemans and Yuchong Pan
The demand matching problem generalizes both the knapsack problem and the $b$-matching problem. In this problem, each edge of a graph has a demand and a weight, and each vertex has a capacity. The goal is to find a maximum weight subset of edges such that, at each vertex, the total demand of the incident selected edges does not exceed the vertex capacity. Parekh [IPCO 2011] proved that, if each edge is individually feasible, the natural LP relaxation for demand matching has integrality gap at most $3$, yielding a $3$-approximation
algorithm. This bound is tight for the natural LP relaxation, matching the lower bound of Shepherd and Vetta [Math. Oper. Res. 2007]. We present a randomized $(3/2 + \sqrt{2} + \varepsilon) \approx (2.914 + \varepsilon)$-approximation
algorithm for the demand matching problem for every $\varepsilon > 0$, giving the first approximation ratio strictly better than $3$. For bipartite graphs, we obtain a randomized $(2 + \varepsilon)$-approximation
algorithm for every $\varepsilon > 0$. Both
algorithms run in time
polynomial in $1/\varepsilon$ and the input length. Our
algorithms use a strengthened LP relaxation based on intersecting the integral knapsack polytopes associated with the vertices, together with a multiple-choice generalization. As a key ingredient, we prove the existence of a $(q, 1/(1+q))$-balanced contention resolution scheme for the integral knapsack polytope for every $q \in [0, 1]$, which may be of independent interest. The balance guarantee $1/(1+q)$ is tight in the worst case over all knapsack instances.
Read the full abstract ▾
Machine Learning
Locating Hidden Failures Makes Long-Horizon Agents More Reliable
Salman Rahman, Yubin Kim, Mihir Parmar et al. (+15 more)
As AI
agents take on long, autonomous tasks, we increasingly oversee rather than perform the work, yet we still judge them almost entirely by whether they finally succeed. An outcome cannot reveal where a run went wrong, whether the
agent recovered, or the irreversible harm it caused along the way, and where long-horizon
agents fail remains unmapped. We study $2518$
agent trajectories across software engineering, computer use, and science, close to real deployment, and classify $6967$ mistakes into $78$ failure types. Failure follows a recurring signature: after its first mistake an
agent often fails to recover and rarely catches the error itself, so the run continues unchecked while still looking correct; whether an
agent recovers depends on the task and the environment's feedback, not on the
agent framework running it. Long-horizon
agents can do real harm on the way to a passing result: even runs scored as solved delete data, corrupt systems, or fabricate success rather than earning it. We release these human-verified annotations as Traverse, a
benchmark on which six frontier judges struggle to locate failure regardless of scale: even the strongest correctly identifies the first mistake in fewer than a third of runs. Yet Scout, a $4$B verifier we trained, locates failure far better than these judges and transfers to domains it never saw. Used at test time to select among an
agent's candidate runs, it raises task success above the
agent's own single-attempt performance, without retraining the
agent. By making failure cheap to locate and correct, this work is a foundation for more trustworthy long-horizon
agents that learn from their own mistakes, and a practical path to overseeing increasingly autonomous AI.
Read the full abstract ▾
All authors Salman Rahman, Yubin Kim, Mihir Parmar, A. Ali Heydari, Genglin Liu, Simon A. Lee, Weizhi Zhang, Arian Hosseini, Ahmed A. Metwally, Yuzhe Yang, Baharan Mirzasoleiman, Xin Liu, Pavel Izmailov, Saadia Gabriel, Mark Malhotra, Shwetak Patel, Daniel McDuff, Hamid Palangi
Robotics
Multi-Session Multimodal Underwater Mapping with Acoustic and Optical Imaging
Precious Philip-Ifabiyi, Valerio Franchi, Fausto Ferreira et al. (+1 more)
Accurate seafloor mapping is essential for marine science, archaeology, and environmental monitoring. However, integrating data from different sensors, such as side-scan sonar and optical cameras, collected across separate survey sessions, remains challenging due to positioning drift and sensor offsets. This paper presents a multi-session,
multimodal underwater mapping framework based on factor graph optimization. The method jointly optimizes vehicle trajectories, 3D landmark positions, sensor extrinsics, and per-session global
alignment transformations. By combining rigid inter-session corrections with local trajectory deformations, it compensates for both inter-session offsets and intra-session distortions from accumulated
navigation errors. The proposed methodology was validated on real-world
datasets collected along the Catalan coast. Results show measurable improvements in map consistency over both unoptimized and rigid-
alignment baselines across all metrics, including Pixel Accuracy and mean Intersection over Union. The method achieves a 3.4% improvement in pixel accuracy over the unoptimized baseline, corresponding to improved semantic labelling across approximately 14700 $\text{m}^2$ of mapped area. Qualitative results further show consistent co-registration between sonar and optical maps, even in the presence of significant trajectory distortions and inter-session misalignments. These findings demonstrate the potential of the proposed framework to generate coherent
multimodal seafloor maps from heterogeneous
underwater surveys.
Read the full abstract ▾
All authors Precious Philip-Ifabiyi, Valerio Franchi, Fausto Ferreira, Nuno Gracias
Computational Engineering · Cross-listed from math.OC
Day-ahead Coordination of Virtual Power Plants within Active Distribution Networks using Deterministic Bi-Level Optimization
Laura M. Barajas-Arguello, Rafael A. Núñez-Rodríguez, Daniel Gebbran et al. (+1 more)
This paper proposes a deterministic bilevel optimization framework for the coordinated operation of Virtual Power Plants (VPPs) embedded in an active distribution network. The Distribution System Operator (DSO) acts as the upper-level leader, minimizing a weighted combination of expenditure, active losses and voltage deviation subject to nonlinear AC power flow constraints, while each VPP operates as a lower-level follower that maximizes its profit under the uniform price signal issued by the DSO. Unlike most existing formulations, which linearize the lower-level subproblem to obtain a Mixed-Integer Linear Program (MILP), the proposed model retains the full AC Optimal Power Flow (AC-OPF) equations, producing a bilevel Mixed-Integer Nonlinear Program (MINLP). The lower-level problem is replaced by its Karush-Kuhn-Tucker (KKT) optimality conditions and the Strong Duality
Theorem, yielding a single-level Mathematical Program with
Equilibrium Constraints (MPEC). Complementarity conditions are then linearized via the Fortuny-Amat big-M transformation. The framework is validated on the IEEE 33-bus feeder over a 24-hour horizon, with four distributed resources aggregated into a single VPP. Compared with individual dispatch against a regulated time-of-use tariff, aggregation reduces active losses by 10.5 %, the accumulated voltage deviation by 18.3 %, and the bus-hours below 0.95 p.u. from 132 to 29. These gains cost 0.31 % in social cost and 0.26 % in DSO expenditure, while the rent of the aggregator is preserved.
Read the full abstract ▾
All authors Laura M. Barajas-Arguello, Rafael A. Núñez-Rodríguez, Daniel Gebbran, Clodomiro Unsihuay-Vila
Machine Learning
Symmetry without a manifold: intrinsic dimension on orbits
Chon-Fai Kam, Miloud Bessafi and Frédéric Cadet
The standard
geometric derivation of neural scaling exponents takes the intrinsic dimension of a data
manifold as its input. On modular addition in $\mathbb{Z}_p$ that derivation has no input. The exact algebraic solution is an orbit of $\mathbb{Z}_p$ acting by isometries. Transitivity alone makes the ratio statistic underlying the standard dimension
estimator a point mass, so the
estimator is undefined, and here the two nearest neighbour distances coincide exactly. Breaking the
symmetry at scale $\epsilon$ returns a number, but one that tracks $1/\epsilon$ with no scale free plateau. We show that the failure is general, since on any finite orbit of a group acting by isometries the
estimator reports the resolution at which the set is probed rather than a dimension. What replaces the power law is exponential in hidden width, $L(h)=L_\infty+A\exp(-c\,h^{\alpha})$, with $R^2$ between 0.982 and 0.995 against 0.857 to 0.906 for a power law admitting the same floor and fitted under the same protocol. Where the data supply is sufficient the rate belongs to the regulariser rather than to the group, since weight decay moves $c$ by a factor of 47 while group order moves it by 1.10, a residual below seed to seed resolution, for every fixed $\alpha$ between 0.75 and 2. The critical width falls with group order rather than rising, against capacity counting that assigns a fixed number of
neurons to each irreducible representation.
Read the full abstract ▾
Computer Vision
Rapid Loss of the Sierra Nevada's Largest Trees Driven by Fire
Fabien H. Wagner, Dan J. Dixon, Christopher W. Woodall et al. (+15 more)
Large trees disproportionately contribute to biomass storage, habitat structure, and ecosystem functioning. However, their distribution and health dynamics remain poorly quantified at a regional scale. Here, a deep learning model (U-Net-ID) and canopy height models derived from sub-meter aerial imagery from 2020 were used to delineate all individual trees with crown area $\geq$ 100 m$^2$ across the Sierra Nevada Floristic Province. The model was trained using more than 3.3 million synthetic tree crowns and achieved a median Intersection over Union (IoU) of 0.602 when validated against an independent
dataset of 20,273 crowns. A total of 6,515,705 large trees were mapped, occurring across approximately 78.7% of the Sierra Nevada Floristic Province. The spatial distribution of large trees showed associations with elevation, temperature, and precipitation. Using Sentinel-2 time series from 2020 to 2025, tree health dynamics were characterized by extracting spectral trajectories for each crown and applying BFAST breakpoint detection
algorithm combined with a disturbance classification framework to identify mortality, disturbance, and recovery trajectories of individual trees. Wildfires, estimated from CAL FIRE fire perimeters, were identified as the dominant driver of large-tree mortality, killing 10% of all large trees in the Sierra Nevada, with mortality strongly concentrated during the extreme 2020-2021 fire seasons.
Read the full abstract ▾
All authors Fabien H. Wagner, Dan J. Dixon, Christopher W. Woodall, Mayumi C. M. Hirye, Felipe Saad, Griffin Carter, Ricardo Dalagnol, Lorena Alves, Cynthia Creze, Stephen C. Hagen, Zhihua Liu, Christopher Mihiar, Adugna Mullissa, Le Bienfaiteur Sagang Takougoum, Bryan Shaddy, Yan Yang, Dafeng Zhang, Sassan Saatchi
Computer Vision
Audio for Sports Highlight Detection: A Comparative Empirical Study
Hao Xu, Meenakshi Sarkar, Vishnu Raj et al. (+1 more)
Sports highlight detection aims to identify the most exciting and meaningful moments from long sports
videos. While existing methods often emphasize visual or visual-
language representations, sports
videos contain rich audio cues, including commentator
speech, crowd reactions, whistles, ball impacts, and referee calls. In this work, we revisit the role of audio in sports highlight detection and ask a simple question: how far can audio alone go? We construct lightweight audio-only baselines using pretrained audio representations and compare them with visual-only and audio-visual methods on the SV-Highlights
benchmark. Surprisingly, our audio-only GRU baseline achieves strong performance and outperforms several existing audio-visual methods under our supervised evaluation setting. Furthermore, a simple audio-visual fusion baseline achieves the best performance across all metrics, indicating that audio and visual cues provide complementary information. To better understand the contribution of audio, we conduct source-separated analysis and show that vocal/commentary audio is more informative than background-only audio, while their combination performs best. We also analyze interpretable audio cues and find that highlight clips exhibit higher RMS loudness, peak loudness, and mid-frequency energy than non-highlight clips, although substantial distribution overlap indicates that loudness alone is insufficient. Our findings suggest that audio is an underexplored but highly informative modality for sports highlight detection and should be treated as a primary signal rather than merely an auxiliary cue.
Read the full abstract ▾
All authors Hao Xu, Meenakshi Sarkar, Vishnu Raj, David Gunawan
Hardware Architecture
Demystifying Gate-Level Localization of RTL Trojans
Navid Nader Tehrani, Azadeh Davoodi and Rasit Onur Topaloglu
Hardware Trojans are malicious modifications that compromise functionality or leak sensitive data. They pose a severe threat, particularly when inserted at the Register Transfer Level (RTL). After synthesis, these Trojans are often concealed by optimizations in gate-level netlists. Recent efforts, including the ICCAD 2025 contest, emphasize golden-chip-free detection using machine learning (ML) on labeled netlists. In this work, we show that RTL Trojans exhibit stable structural and signal-flow patterns post-synthesis, enabling effective detection through targeted heuristics rather than generic ML feature learning. We propose LoRD, a lightweight heuristic-based approach that exploits these distinctive subgraph signatures, achieving near-perfect detection and localization on the contest testcases. Com- pared to a
transformer-based ML baseline and top five teams, LoRD achieves on-average a score of 2.957 (out of 3) for Trojan- implanted designs without the data and tuning overhead.
Read the full abstract ▾
Artificial Intelligence
Collaborative Memory for Multi-Agent VLM Systems
Huixin Zhang, Shao-Jun Xia, Di Wang et al. (+3 more)
Vision-
language model (VLM)
agents combine specialized perception, tools, and
reasoning to address complex visual tasks. In multi-
agent settings, different
agents inspect different image regions,
video frames, or visual representations, so collaboration extends beyond distributed
reasoning to distributed perception. This makes shared visual context a central problem in VLM
agent collaboration. In this paper, we frame
memory hierarchy, cross-
agent sharing, and consistency mechanisms around the need to reconcile interpretations and update dependent
reasoning. Effective collaboration requires
agents to build on contributions from other
agents, recover missing visual context, and reconcile differing interpretations as new evidence emerges. Shared visual
memory preserves not only images or textual summaries but also the dependencies among observations,
agent interpretations, and subsequent
reasoning. Together, these design considerations shape how information flows and evolves across VLM
agents. The proposed framework provides a foundation for building reliable and resource-efficient
agent teams.
Read the full abstract ▾
All authors Huixin Zhang, Shao-Jun Xia, Di Wang, Liangxi Liu, Hainan Xiong, Zihao Wang
Information Theory · Cross-listed from quant-ph
Classical, quantum, and general probabilistic state-discrimination profiles
Mihály Weiner
For a collection of states indexed by $[n]={1,\ldots,n}$, let $F(H)$ denote the optimal unnormalized minimum discrimination error for each nonempty subfamily $H\subseteq[n]$. We call $F$ the discrimination profile and characterize exactly which profiles can arise in a general probabilistic theory (GPT). For fixed $n$, the GPT region $\mathfrak{G}_n$ is a bounded rational polytope, giving a hierarchy $\mathfrak{C}_n\subseteq\mathfrak{Q}_n\subseteq\mathfrak{G}_n$ of classical,
quantum, and GPT profiles, analogous to the local–
quantum–no-signalling hierarchy in Bell theory. For $n=3$ we determine the classical and GPT polytopes explicitly. Within $\mathfrak{G}_3$, the classical region is characterized by $\beta:=F(123)-F(12)-F(13)-F(23)\le0$, while the GPT maximum is $1$. For
qubit triples, the maximal value is attained by the trine, $\beta_{\mathrm{tr}}=3\sqrt3/2-2$, and we prove the dimension-independent
quantum bound $\beta_{\mathrm Q}\le2/3$, with a slight further improvement. Hence $0=\beta_{\mathrm C}<3\sqrt3/2-2\le\beta_{\mathrm Q}\le2/3<1=\beta_{\mathrm{GPT}}$. We
conjecture $\beta_{\mathrm Q}=\beta_{\mathrm{tr}}$ and prove this for arbitrary pure-state triples, together with further supporting evidence. For arbitrary $n$, every classical facet admitting a GPT violation already admits a
quantum violation in dimension at most $3$. A four-state
qubit example shows that
quantum success profiles need not be submodular, unlike for $n=3$. Finally, we exhibit a four-dimensional
quantum triple whose discrete profile is classical for all subfamilies but becomes nonclassical when unequal priors are allowed.
Read the full abstract ▾
Game Theory
Estimate then Predict: Convex Formulation for Travel Demand Forecasting
Youngseo Kim, Gioele Zardini, Samitha Samaranayake et al. (+1 more)
Travel demand
forecasting is essential for evaluating large-scale infrastructure projects, yet the traditional sequential four-step process can produce inconsistencies across trip distribution, mode choice, and traffic assignment. Although combined models address these inconsistencies, their practical use has been limited by simplified behavioral assumptions, computational burden, and the lack of a unified parameter-estimation framework. We propose a convex programming approach that integrates destination, mode, and route choices within a hierarchical extended logit model. The optimal primal solution characterizes the joint travel-demand
equilibrium, while the optimal dual variables recover taste coefficients, alternative-specific constants, and destination- and mode-level scale parameters. The model captures mode correlations through nested logit and route overlap through path-size logit. Its convex structure provides global optimality guarantees and enables efficient solution using off-the-shelf conic solvers. Observed travel patterns are incorporated through moment and conditional-
entropy constraints, while the route-level dispersion parameter is calibrated separately using partial link counts. Numerical experiments on eight
benchmark networks demonstrate the scalability and computational efficiency of the formulation.
Read the full abstract ▾
All authors Youngseo Kim, Gioele Zardini, Samitha Samaranayake, Soroosh Shafiee
Machine Learning
EdgeReMIND: A Scalable, Top-Ranked Memorization Baseline for Temporal Multi-Relational Link Prediction
Bryant Pollard
Temporal link prediction on the Temporal Graph
Benchmark 2.0 (TGB 2.0) faces a scalability ceiling: on the
benchmark's three largest
datasets, every existing
embedding method runs out of
memory or exceeds the time budget. These large-scale graphs are the ones nearest real deployment scale, so failing on them is a real production limitation. EdgeReMIND sets the highest reported test mean reciprocal rank (MRR) on six of eight TGB 2.0
datasets and is the only relation-aware method that runs on all of them. This linear memorization model, with learned per-relation weights over data-calibrated features, is therefore not merely a fallback where
embeddings fail but a practical state-of-the-art baseline across the
benchmark.
Read the full abstract ▾
Computational Complexity
Separating Non-redundancy and Chain Length
Joshua Brakensiek, Venkatesan Guruswami and Aaron Putterman
For a constraint satisfaction problem defined by a relation $R$, its non-redundancy $\text{NRD}(R,n)$ is the size of largest instance (as a function of the number $n$ of variables) for which no constraint is implied by the rest. Its chain length $\text{CL}(R,n)$ is the largest such instance where the constraints can be ordered so that no constraint is implied by the preceding ones. Clearly $\text{CL}(R,n) \ge \text{NRD}(R,n)$ but so far no asymptotic separation was known between these quantities. We exhibit an explicit arity $4$ relation for which $\text{CL}(R,n) \ge \omega(\text{NRD}(R,n))$.
Read the full abstract ▾
Computer Vision
Face-voice Association across LAnguages and Gender (FLAG) 2027 Challenge Evaluation Plan
Marta Moscati, Swapnil Khandoker, Muhammad Saad Saeed et al. (+8 more)
Face–voice association models may rely on
language or gender cues in the voice rather than on speaker-specific voice characteristics, which can lead to a performance deterioration when the model has to identify a
multilingual speaker or distinguis same-gender speakers. To investigate these issues, we introduce the Face-voice Association across
LAnguages and Gender (FLAG) 2027 Challenge. The challenge formulates face–voice association as a cross-modal verification task: given a voice, identify the speaker's face from a “gallery” of faces consisting of the speaker's face and a set of negative samples. Models are evaluated on identities not present in the training data (“unseen”) and both for
languages present or absent from the training data (“heard” and “unheard”). Two evaluation settings are used to test models' reliance on gender: a standard, unconstrained and a gender-constrained one, where the latter uses a same-gender gallery. The performance of existing, baseline models in these settings reveals that models performance degrades under
language shifts and in gender-constrained settings, highlighting the need to foster the development of models that capture identity-specific aspects beyond
language and gender. The challenge provides a
benchmark dataset, pretrained baseline models, and an evaluation framework to advance face–voice association.
Read the full abstract ▾
All authors Marta Moscati, Swapnil Khandoker, Muhammad Saad Saeed, Shah Nawaz, Fatima Noor, Rohan Kumar Das, Mubashir Noman, Junaid Mir, Muhammad Haroon Yousaf, Khalid Malik, Markus Schedl
Robotics
Map2Route: Benchmarking Compositional Language-Grounded Route Planning over Semantic Maps
Muyi Bao, Hang Xu, Jingfan Tang et al. (+5 more)
We introduce Map2Route, a human-curated
benchmark for compositional
language-grounded route planning over pre-built semantic maps. Map2Route contains 1,000 episodes across 40 scenes, where instructions use relational, comparative, and nested descriptions to identify route-relevant objects and regions, while specifying ordered must-pass regions, must-avoid requirements, five categories of soft preferences, and spatial and route-stage scopes, which is partially tested by existing works. Alongside Map2Route, we propose Grounding2Route, which combines executable code-as-grounding with verification-guided repair and scope-aware planning.Across seven representative adapted baselines, Grounding2Route substantially outperforms existing methods in all metrics. Despite these gains, a substantial gap to human demonstrations remains, highlighting the difficulty of Map2Route and the considerable headroom for future progress. Additional qualitative results and resources are available on https://anonymous.4open.science/w/Map2Route-F05F/.
Read the full abstract ▾
All authors Muyi Bao, Hang Xu, Jingfan Tang, Zihan Liu, Yuxin Cai, Chen Lv, Wenshan Wang, Ji Zhang
Computer Vision
Zing-0.5: Toward Playable Worlds with Real-Time Joint Action and Text Control
Mingyang Chen, Shengdong Chen, Xiaoxiao Fu et al. (+13 more)
We introduce Zing-0.5, a 5B autoregressive world model designed for playability: users can explore generated worlds, influence unfolding events, and respond to the resulting feedback through joint keyboard and online text control. Our approach brings together three technical contributions: (1) Unified action and text conditioning, combining magnitude-aware keyboard inputs with temporally aligned text instructions and jointly annotated
videos to learn
navigation and event control within the same sequence; (2) Event-scale supervision for incremental generation, using a segment-level teacher trained on connected multi-
prompt videos to supervise a block-level causal student through distribution-matching
distillation; and (3) Low-cost real-time interaction, combining four-step generation with context-preserving streaming to support 832 x 480 inference at 24 FPS at an estimated server rental cost of approximately USD 0.009 per stream-minute. Zing-0.5 achieves an overall score of 81.0 and a consistency score of 88.5 across 158 WBench
Navigation cases. A joint-control demonstration shows a text-directed event change during continued
navigation without restarting generation. We release the model weights, inference code, and Zing-SGLang serving implementation to support further work on playable generated worlds.
Read the full abstract ▾
All authors Mingyang Chen, Shengdong Chen, Xiaoxiao Fu, Bosheng Gong, Haoyuan Guo, Bowen Li, Jiawen Li, Kejun Li, Tianpeng Li, Yin Liu, Haoze Sun, Zeyang Tian, Meng Wang, Xinmiao Wu, Jiangqiao Yan, Zining Zhao
Emerging Technologies
Co-occurrence-Aware Quadratic Assignment for Local Feature Matching in Simultaneous Localization and Mapping
Yutaka Yamada, Yohei Hamakawa, Yutaro Ishigaki et al. (+2 more)
Local feature matching, which associates keypoints in two images as keypoint pairs, is fundamental to Visual Simultaneous Localization and Mapping (Visual SLAM). Nearest Neighbor (NN) search is commonly used for keypoint matching, but it has difficulty selecting correct keypoint pairs when multiple candidates have similar costs. To improve matching accuracy, this paper proposes a keypoint matching method that considers the pairwise co-occurrence of two keypoint pairs. The keypoint matching is formulated as a quadratic assignment problem, which is an NP-hard combinatorial optimization problem, making it difficult to solve quickly on conventional computers. Recently, Ising machines have been developed as computing devices capable of solving hard combinatorial optimization problems. Using a simulated bifurcation based Ising machine, the proposed method improved matching accuracy by approximately 8 percentage points over a conventional method on the HPatches
dataset. Furthermore, we integrated the proposed method into ORB-SLAM3, a representative academic Visual SLAM system, and achieved a 3.78-fold improvement in absolute pose error (APE) and a 2.85-fold improvement in relative pose error (RPE) on the KITTI
dataset scenes where multiple same shape objects are repeatedly arranged, which are challenging for accurate self-pose estimation by the original ORB-SLAM3.
Read the full abstract ▾
All authors Yutaka Yamada, Yohei Hamakawa, Yutaro Ishigaki, Masaya Yamasaki, Kosuke Tatsumura