Computer Vision
One Diffusion Model, Two Roles: Guided Trajectory Planning and Safety-Critical Scenario Generation in Closed-Loop Simulation
Arka Pal, Rajesh Kumar, Hannes Eriksson et al. (+4 more)
Diffusion probabilistic models can capture the multi-modal, interaction-rich distribution of joint future trajectories in driving scenes. We show that a single pretrained
diffusion traffic model can serve two complementary roles in the autonomous driving development loop: as an ego motion planner, and as a controllable generator of safety-critical scenarios for stress-testing the planners. On the planning side, we introduce a Single-Stream Dual-Stream (SSDS)
diffusion-
transformer decoder that fuses scene context via joint
attention rather than late cross-
attention, improving closed-loop performance on nuPlan. We further propose Decoupled Annealing Posterior Sampling with Energy (DAPSE), a training-free guidance scheme that injects arbitrary energy functions at the clean-sample level, avoiding the first-order approximation errors while requiring no auxiliary networks. Beyond planning, we leverage the same
diffusion model as a controllable scenario generator to create realistic long-tail driving interactions for closed-loop evaluation. Through inference-time guidance, selected
agents are steered toward safety-critical behaviors, including aggressive cut-ins, lead-vehicle braking, and combined longitudinal-lateral interactions, while preserving realistic traffic behaviors. Evaluated in closed-loop nuPlan simulations with independent black-box planners, the generated scenarios expose failure modes that remain hidden under standard
benchmarks. Although the SSDS-based planner achieves stronger nominal performance, it experiences larger degradation under these challenging scenarios, demonstrating that
benchmark superiority does not necessarily translate to robustness. These results demonstrate that a single learned traffic prior can simultaneously improve motion planning and provide a realistic framework for systematic planner robustness evaluation.
Read the full abstract ▾
All authors Arka Pal, Rajesh Kumar, Hannes Eriksson, Rémi Lacombe, Arvid Laveno Ling, Ankit Gupta, Maciej Wozniak
Networking
QUASAR: Quantum Satellite Architecture and Routing Simulator
Yaliang Shi, Zi Wang, Bangguo Yuan et al. (+2 more)
The deployment of Low Earth Orbit (LEO) satellite constellations is an important step toward global-scale
quantum networking. However, evaluating satellite
quantum network protocols under spatiotemporal orbital dynamics and
quantum physical constraints remains computationally expensive and challenging. In this paper, we propose
QUASAR, a lightweight simulator for evaluating
entanglement distribution in satellite-based
quantum networks.
QUASAR provides a decoupled architecture that integrates dynamic orbital
topologies, time-varying optical transmittance, and
quantum memory decoherence into network- layer attributes. To demonstrate its capabilities, we abstract and implement two representative hardware architectures: Simultaneous Downlink and On-Orbit Stitching. We further introduce an
Entanglement Distribution Rate (EDR)-Aware Spatiotemporal Routing (EASR) heuristic as a reference workload. Our case study examines how
QUASAR supports different satellite architectures, routing workloads, realistic orbital traces, concurrent requests, and scalable event-driven execution. With over 85% lower network-layer update latency than continuous polling,
QUASAR provides a practical and extensible framework for future satellite
quantum network protocol evaluation.
Read the full abstract ▾
All authors Yaliang Shi, Zi Wang, Bangguo Yuan, Gaojie Wu, Zhiwei Zhao
Artificial Intelligence
Artificial Intelligence in Equity and Crypto Markets: Progress, Profitability Evidence, and the Limits of Automated Investing
Linsen Zhu and Mengqing Cai
Artificial intelligence (AI) now supports investment workflows from data and prediction through research, portfolios, execution, and tool use. Technical capability, however, is not evidence of investment profitability. This critical state-of-the-art review examines public research available through 31 August 2026 on listed equities, exchange-traded funds, centralized crypto spot, perpetual futures, and on-chain
markets. We organize evidence with an alpha-translation chain: point-in-time information must yield a stable signal, feasible positions, executable orders, and risk-adjusted returns after costs. Across machine learning, time-series foundation models, financial
language models, reinforcement learning, and
agents, the examined record shows real but mainly upstream progress in prediction, text processing, portfolio design, and workflow integration. Evidence is thinner for durable net performance. Temporal contamination, repeated selection, survivorship, weak
benchmarks, implementation costs, venue mechanics, and capacity can break translation to net alpha. Strong historical results coexist with predictor decay, corrected look-ahead failures, mixed prospective evidence, and few audited live-capital records. Crypto adds informative state but requires separate treatment of spot, perpetual, and decentralized cash flows and execution. Within the public evidence examined here, no general AI architecture is shown to deliver persistent, cross-regime, capacity-aware net alpha. More credible claims require point-in-time data and models, decision-aligned objectives, joint portfolio–execution evaluation, controlled adaptation, prospective tests, and authority-matched governance. These conditions can improve evidence and implementation; they do not guarantee profit.
Read the full abstract ▾
Information Theory
Minimum Schubert Codewords and Second-Minimum Grassmann Codewords
Muskan Khaneja and Prasant Singh
In this paper, we give a classification of the minimum weight codewords of Schubert codes $C_\alpha(\ell, m)$ by settling the
conjecture proposed by Ghorpade and Singh in 2018, for all values of $q$ and all $\alpha$. We use this classification to prove that a codeword of the Grassmann code $C(\ell, m)$ has the second minimum weight if and only if it is indexed by an element of $\bigwedge^{m-\ell}V$ that can be written as the product of a decomposable $(m-\ell-2)$-vector and an alternating $2$-vector of rank $4$. Finally, we give an enumeration of the second minimum weight codewords of the Grassmann code.
Read the full abstract ▾
Artificial Intelligence
Compact-Memory LLM Agents via Online Max-Member Clustering and Atom-Aware Packing
Jiahe Geng, Jinpeng Wang and Kun Yuan
Many long-horizon LLM deployments face tight
prompt budgets: latency, cost, and context limits make full-context prompting impractical as interaction length grows. The key question is then not raw recall alone, but which
memory design gives the best quality–
token trade-off in the compact-
memory regime. We present RSM-full, an online clustered-
memory pipeline designed for a strong quality–
token Pareto point. RSM-full combines two design choices: a cosine-gated max-member merge write rule and an atom-aware grouped context packer. On AMA-Bench, our primary compact-
memory benchmark, it reaches $83%$ of Full-Context quality at $32%$ of the
token cost at a $4$k budget; under four-seed averaging it beats the closest streaming-clustered baseline (Online K-Means) by $+3.5$–$6.0$,pp ($p{<}.001$) across the whole ${\sim}2.6$k–${\sim}5$k regime. Three-seed ablations show most of this gain comes from the merge rule ($+5.7$,pp over Online K-Means and matched-$\tau$ DP-means) and the grouped packer ($+5.0$,pp over flat concatenation). The pattern reproduces on RealMem, an independent long-horizon persona-
memory benchmark: RSM-full improves on Budget-RAG ($+0.69$,pp, $p{=}.006$), is on par with BM25-RAG (paired $\Delta{=}{+}0.27$,pp, $p{=}.47$; we do not claim BM25 equivalence in the equivalence-test sense), and significantly outperforms Streaming-Proto ($+2.97$,pp) and the closest reproduced 2025 agentic-
memory baseline A-MEM ($+1.65$,pp, $p{<}.001$). Across
benchmarks the message is consistent: under tight budgets, compact-
memory performance is driven mainly by how streaming
memories are merged and how retrieved content is assembled. Overall, RSM-full is most useful when answeroughly $2k$–$5k$
prompt tokens, where itdefines a strong compact-
memory Pareto point; higher-
token baselines remain stronger outside this regime.
Read the full abstract ▾
Software Engineering
ARIA - An Agentic Framework for Autonomous Testing of Infotainment Systems
António Azevedo, Bruno Lima and João Pascoal Faria
Automotive infotainment validation still relies on manual testing, slow, costly, and incompatible with agile releases and OTA updates. Scripted automation only partly helps: it couples test logic to implementation, yielding brittle, high-maintenance suites. Existing LLM-driven frameworks mostly target web/mobile apps, using single- or dual-
agent setups that overload one or two models with perception, planning, action selection, and validation at once, prone to
hallucinations and unproductive exploration loops given infotainment complexity. We present ARIA (Autonomous Real-time Infotainment Assessment), a multi-
agent LLM framework that autonomously runs end-to-end tests on Android infotainment systems via visual interaction, using a closed-loop pipeline of four specialized
agents per step plus a report stage. From single-sentence scenarios (path, action, expected outcome), ARIA runs the interactions and produces reports, reproducible scripts, and visual evidence per step. Evaluated on a manufacturer's physical Android infotainment system across 30 scenarios, ARIA completed 28 (93.3%) with a verdict (2 errored), 20 of which (71.4%) matched ground truth. It caught all 5 known defects, no fault passed as working; its 8 false positives stem from
navigation/image limits and unsupported gestures, showing multi-
agent LLMs can run infotainment tests industrially while exposing the cost of a low false-positive tolerance. A single-
agent baseline confirms the multi-
agent design's value: on the first pass, before stronger-model revisitation narrows the gap, it shows a far higher false-positive rate (72.0% vs. 52.6%), conflating navigational difficulty with system failure. We report first-pass/post-revisitation results,
token/call/cost per scenario, and show via repeated runs that stability tracks complexity, with fault detection perfectly consistent, pointing to CI integration of visual testing.
Read the full abstract ▾
Hardware Architecture
TreeFI: Value-Aware Statistical Fault Injection for Deep Neural Networks
Noam Bires, Marcello Traiola, Angeliki Kritikakou et al. (+1 more)
Reliability evaluation of deep neural networks under hardware faults commonly relies on fault injection, but exhaustive campaigns are intractable for modern models and
datasets. Statistical fault injection reduces this cost, yet existing approaches still require large injection budgets because they do not explicitly exploit a key property of floating-point faults: the effect of a bit flip depends strongly on the value being corrupted. We propose TreeFI, a value-aware statistical fault-injection methodology for FP32 single-bit faults in DNN activations and weights. TreeFI partitions each layer's value distribution into intervals with similar expected bit-flip behavior, learned using regression trees, and allocates injections across these intervals according to their relevance for failure-rate estimation. This stratified allocation preserves the target confidence and error margin while avoiding unnecessary injections in low-impact regions of the fault space. We validate TreeFI on CNN and
Transformer models using CIFAR-10 and ImageNet. On ResNet8, where exhaustive activation fault injection is feasible, TreeFI provides more accurate estimates than state-of-the-art statistical FI baselines under the same campaign setting. Across the evaluated models, TreeFI reduces the required injection budget by up to 72.1x, with average reductions of 44.9x for activation faults and 11.2x for the executed weight campaigns.
Read the full abstract ▾
All authors Noam Bires, Marcello Traiola, Angeliki Kritikakou, Elisa Fromont
Computer Vision
TourPhysics: Bringing Physics to World Models for Exploration and Manipulation from a Single Image
Xin Zhang, Yabo Chen, Zixuan Duan et al. (+4 more)
Interactive visual world models must distinguish observation from physical intervention. Camera motion reveals new surfaces, whereas intervention changes object motion, contact, and deformation. Current
video world models are largely driven by appearance priors and often lose physical or spatial consistency over long horizons. We present TourPhysics, an online framework initialized from a single image and a declarative physical configuration. TourPhysics extends PhysOmni, our ACM Multimedia 2026 work, from finite physics-grounded
video synthesis to persistent exploration and manipulation. TourPhysics combines deterministic simulation with
video generation while assigning separate roles to simulator state,
geometric evidence, generator controls, and appearance
memory. For each action, the simulator computes a finite physical and camera trajectory before the corresponding observation is generated. Accepted observations publish the terminal state and update the appearance
memory and subsequent generator controls, while the committed state and simulator geometry remain fixed throughout synthesis and retry. We further separate the simulator geometry used for projection and visibility from the relative depth used to condition the generator. A reference-anchored
memory retrieves accepted static appearance through
geometric cross-view correspondence and incorporates it through a bounded residual that reverts to the native path when no valid correspondence exists. On simulator-defined camera tours and object manipulations, TourPhysics follows prescribed camera and object trajectories more closely than the evaluated baselines, preserves the input scene, and reduces appearance drift during long-horizon revisits.
Read the full abstract ▾
All authors Xin Zhang, Yabo Chen, Zixuan Duan, Haibin Huang, Chi Zhang, Feng Xu, Xuelong Li
Machine Learning
Fast Gauss Sums via Flash Attention
Nicolaj Rux and Sebastian Neumayer
Gaussian kernel sums are the computational core of maximum mean discrepancies (MMDs), kernel gradient flows, Stein variational gradient descent (SVGD), and many other kernel methods. At the same time, softmax
attention has received an extraordinary amount of hardware-aware code engineering, culminating in flash
attention. We show that Gauss kernel sums with arbitrary, signed weights can be evaluated via flash
attention: two small input augmentations turn the normalized softmax reduction into the unnormalized Gauss sum, without writing a single line of custom GPU code. For feature dimension D>8 in fp16, this approach beats compiled PyTorch code as well as PyKeOps kernels (often significantly) in speed,
memory-overhead and accuracy. Indeed, its
memory scaling remains linear.
Read the full abstract ▾
Software Engineering
Better Understanding, Better Fixes? A Study of Hallucination in LLM-based Automated Program Repair
Xuemeng Cai, Jiakun Liu, Linhan Yang et al. (+2 more)
Large
language models (LLMs) have significantly advanced automated program repair (APR), yet existing evaluations remain largely result-centric and provide limited insight into
hallucination during repair. In APR,
hallucination may arise not only in final patches but also in the intermediate artifacts that guide patch generation. To address this gap, we perform a multi-layered analysis of
hallucination throughout the APR process. Specifically, we characterize
hallucination as the production of patches or intermediate artifacts that are not faithfully grounded in the available repair evidence. We examine repair
hallucination in final patches and understanding
hallucination in intermediate artifacts through three tasks, namely triggering testcase identification, line coverage prediction, and additional testcase generation. We then evaluate three representative LLMs on 832 Defects4J bugs through automatic evaluation and manual analysis. Our results show that both repair and understanding
hallucinations remain prevalent. Across models and settings, only 21.0%-55.9% of generated patches pass the developer-written test suite. Moreover, although more accurate intermediate artifacts are generally associated with successful repairs, this relationship does not always hold. Manual analysis of 812 sampled repairs identifies repair
hallucinations in 72.7% of cases, including patches that pass all available tests; incorrect causal localization and incorrect repair strategies account for 45.9% and 18.5% of these
hallucinations, respectively. Meanwhile, models frequently misidentify triggering testcases, mispredict line coverage involving branching control flow, and generate additional testcases with missing bug-triggering conditions or incorrect expected behavior.
Read the full abstract ▾
All authors Xuemeng Cai, Jiakun Liu, Linhan Yang, Wei Ma, Lingxiao Jiang
Computer Vision
Methane Detection On Board Satellites from Unorthorectified Imagery
Luca Marini, Maggie Chen, Hala Lamdouar et al. (+3 more)
As a potent greenhouse gas, methane is a major driver of
climate change. Its effective mitigation relies on timely detection. Conventional detection methods rely on orthorectification to correct
geometric distortions and matched filters to enhance plume signals, which are steps designed for ground processing and poorly suited to onboard execution. We introduce UnorthoDOS, a
dataset and approach for training machine learning models directly on unorthorectified hyperspectral imagery, bypassing both orthorectification and matched-filter products. Our U-Net models trained on unorthorectified data approach the performance of models trained on orthorectified data (IoU 16.91% vs. 18.47% on all plumes), while both substantially outperform the mag1c matched-filter baseline (IoU 4.76%). We further demonstrate the feasibility of onboard deployment: FP16 compression halves model size with under 0.3% output deviation. The trained ML models and two ML-ready
datasets – orthorectified and unorthorectified hyperspectral imagery from the EMIT sensor – are publicly available at https://huggingface.co/
datasets/SpaceML/UnorthoDOS, with code at https://github.com/spaceml-org/plume-hunter.
Read the full abstract ▾
All authors Luca Marini, Maggie Chen, Hala Lamdouar, Laura Martínez-Ferrer, Dr C. P. Bridges, Giacomo Acciarini
Programming Languages
CPL: A Compact C-like Systems Language with Explicit Low-Level Control
Nikolay Fot and Alexander Vinarsky
This paper presents Cordell Programming
Language (CPL), a compact C-like systems
language that retains C's direct access to
memory, layout, and machine interfaces while experimenting with a smaller grammar and selected conveniences from newer
languages. Also this paper studies whether C-like are more convenient to use for compiler experiments than modern approaches and paradigms. While the
language and compiler provide primitive values, pointers, arrays, containers with methods, unions, generic functions, overloads, entry-point and section control, system calls, and inline assembly, they do not provide high-level constructs such as classes, built-in methods, a standard library, or
memory protection. The article describes the
language design, compiler pipeline, target backends, static-analysis architecture, and OS-facing use cases, then evaluates the prototype backend with reproducible x86_64 and i386 microbenchmarks against C compiler baselines. The obtained results suggest that the compiler can produce code comparable to that produced by production compilers such as GCC and Clang, as well as by small compilers such as TinyC and SmallerC.
Read the full abstract ▾
Computer Vision
InterSing: Explicit Interaction Dynamics for 3D Duet Singing Animation and Beyond
Yihan Zhou, Zikai Huang, Yuyang Yu et al. (+3 more)
We present InterSing, a framework for generating realistic 3D head animations for duet singing performances. Unlike solo singing, duet performance requires each singer to balance individual expressiveness with intermittent interaction at musically salient moments, such as phrase boundaries, synchronized rhythms, and call-and-response passages. Because these interactions are sparse and rhythm-dependent, existing audio-driven animation methods and conversational interaction models do not adequately capture their structure. Our key insight is that duet coordination can be represented as a time-varying signal that reflects how strongly performers engage with one another throughout a song. Based on this observation, we introduce interaction logits, an interpretable latent representation that models the degree of cross-performer engagement at each time step. We learn these logits using weak supervision and use them to condition an interaction-aware
diffusion model jointly driven by audio features and interaction dynamics. This formulation enables unified multi-mode generation, spanning independent motion, coordinated behavior, and smooth transitions between them. Experiments show that InterSing generates realistic and expressive singing head animations with stronger coordination and musical
alignment than existing methods, while preserving each performer's characteristic motion style. We further demonstrate that the same formulation generalizes to multi-singer performances and provides intuitive control over when and how performers engage.
Read the full abstract ▾
All authors Yihan Zhou, Zikai Huang, Yuyang Yu, Xuemiao Xu, Cheng Xu, Shengfeng He
Computer Vision
Sound-based Multi-Person 3D Pose Estimation
Yusuke Oumi, Yuto Shibata, Go Irie et al. (+3 more)
Can we recover the 3D poses of multiple people using only sound? This paper presents the first attempt to estimate multi-person 3D poses solely from acoustic signals. Estimating the poses of multiple individuals using acoustic signals is inherently challenging due to the superposition of motion-dependent signal variations. Unlike single-person scenarios, the presence of multiple subjects leads to overlapping acoustic signatures, making it difficult to attribute specific signal changes to an individual's pose. Furthermore, the complexity is compounded by inter-person reflections, which introduce intricate propagation delays that obscure the temporal motion-acoustic relationship. To address these issues, we propose SoundMHPE (Sound-based Multi-person Human Pose
Estimator), a novel encoder-decoder framework consisting of two key components. First, the Acoustic Multi-scale Encoder captures diverse temporal and fine-grained frequency features to isolate subtle acoustic signatures from complex, overlapping signals. Second, the Temporal Pose Decoder employs an
attention mechanism to disentangle multi-person information across successive frames. By jointly accounting for temporal dynamics and inter-person dependencies, this component precisely reconstructs frame-wise individual poses. To validate our approach, we constructed the 6-hour Acoustic Multi-person Pose (AMP)
dataset consisting of 432K synchronized frames of multi-person pose and acoustic data, and demonstrated that our SoundMHPE outperforms baseline models. Project page: https://oumi03.github.io/sound-mhpe/
Read the full abstract ▾
All authors Yusuke Oumi, Yuto Shibata, Go Irie, Akisato Kimura, Yoshimitsu Aoki, Mariko Isogawa
Machine Learning
Adaptation Interfaces for In-Context Tabular Foundation Models in Time-to-Event Prediction
Minh-Khoi Pham, Luca Cotugno, Dan Cernei et al. (+8 more)
Tabular foundation models (TabFMs) achieve strong performance on structured data, particularly for standard classification and regression problems. Yet, extending them to censored time-to-event prediction is challenging because it requires properly handling censoring and event-time dynamics. Building on our prior work, we further link TabFMs with CoxPH and DeepHit and revise the context-resampled training procedure. We evaluate temporal zero-shot reformulation, classification-based fine-tuning, and survival-head adaptation using frozen TabFM backbones on 74 single-risk data sets, and we additionally study 4 competing-risk data sets. Zero-shot inference is effective on smaller single-risk data sets, whereas supervised adaptation becomes increasingly advantageous as data sets scale. Cox provides the most reliably strong interface, especially for Integrated Brier Score (IBS) on larger data sets. DeepHit is relatively stronger for the time-dependent Concordance Index than for IBS, while cause-specific MTLR ranks highest among the TabFM survival heads in the four-data-set competing-risk analysis. Classification fine-tuning becomes more competitive with zero-shot inference as data sets grow but remains weaker for probabilistic prediction. Overall, our results indicate that effective TabFM transfer depends on the data regime and on the statistical structure represented by the chosen adaptation interface. The implementation scripts used for this work are available at https://github.com/kaylode/survival-fm.
Read the full abstract ▾
All authors Minh-Khoi Pham, Luca Cotugno, Dan Cernei, Alina Sirbu, Stefano Masi, Giuseppe Prencipe, Alessandro Pingitore, Patrizia Landi, Tai Tan Mai, Martin Crane, Marija Bezbradica
Cryptography and Security
The Security Feature Location Problem
Kevin Hermann, Sven Peldszus, Thorsten Berger et al. (+1 more)
Software security must be realized through security features such as authentication and encryption, but which features does a system implement, and where? We present security feature location: the task of relating code locations to security features, enabling developers to understand security implementations and assess whether intended security properties are enforced.
Read the full abstract ▾
All authors Kevin Hermann, Sven Peldszus, Thorsten Berger, Adam Shostack
Computation and Language
RefactorPlatform: An Open-Source Harness for Controlled Evaluation of Repository-Scale Refactoring Agents
Aziz Ben Amor, Drish Mali, Mann Acharya et al. (+2 more)
Repository-scale refactoring requires coding
agents to propagate a single change across many interdependent files without altering program behavior, yet to our knowledge no existing harness isolates the design choices that determine
agent success on this task. We present RefactorPlatform, an open-source evaluation harness that holds the environment fixed and varies each design axis explicitly: model backbone (via OpenRouter and GitHub Copilot CLI), execution regime (baseline,
retrieval-augmented, and multi-
agent), and
prompt specificity. Each run executes in an isolated workspace with live terminal streaming, per-task logging of
tokens, diffs, and transcripts, AST-based verification, and exportable telemetry for audit and reproduction. Demonstrating the platform on 100 multi-file RefactorBench tasks across four model families, we illustrate the analyses it supports: AST-aware chunking outperforms naive
token-window chunking by 25-30 percentage points (pp) across
prompt modes, whereas naive
retrieval falls below the
retrieval-free baseline; a lean
retrieval-augmented single
agent (86%) beats the sub-
agent configuration we evaluated (66%) on matched tasks with no task passing under delegation that fails under
retrieval; and
retrieval's accuracy gains absorb its
token overhead, leaving cost per successful refactoring unchanged. RefactorPlatform is open-sourced to make refactoring-
agent evaluation reproducible and auditable.
Read the full abstract ▾
All authors Aziz Ben Amor, Drish Mali, Mann Acharya, Vijayasri Iyer, Sébastien Bratières
Computation and Language
Cache-Aware Joint Router Adaptation for Memory-Efficient MoE Inference
Zhenhe Wu, Yaping Jin, Qinghua Xing et al. (+6 more)
Mixture-of-Experts (MoE) models activate few experts per
token, yet their full expert sets can exceed GPU
memory and require repeated weight transfers during decoding. We formulate expert-cache management as a model-side algorithmic problem and propose cache-aware post-training that jointly adapts the MoE backbone and lightweight auxiliary routers while preserving the native inference-time Top-K rule. The update-only Temporal Router learns same-layer retention across
tokens without proactive loading. The full Spatio-Temporal Router adds a Spatio Router that uses the causal predecessor's hidden state to refine the temporal cache before target-layer access. We evaluate both modes on Qwen3 and GPT-OSS across GSM8K, MATH, and CommonsenseQA. Temporal Router consistently improves hit rate and reduces expert-weight traffic over matched LM-only baselines. On Qwen3, the full mode improves adjusted hit rate by 1.15–18.03 points and reduces traffic by 4.6–53.3% relative to the strongest evaluated prefetching baseline; GPT-OSS results are competitive but task-dependent. Auxiliary-only training preserves baseline accuracy but yields modest coverage gains; joint post-training achieves substantially higher coverage. Sensitivity analyses distinguish the effects of cache capacity, refinement budget, and cache-loss weight on coverage, traffic, and quality.
Read the full abstract ▾
All authors Zhenhe Wu, Yaping Jin, Qinghua Xing, Hang Zhou, Wei He, Xianjie Wu, Xianfu Cheng, Jian Yang, Hanting Chen
Artificial Intelligence
From Language Models to World-Acting Systems: Progress and Limits of Agentic AI across Digital, Social, Virtual, and Physical Environments
Linsen Zhu and Mengqing Cai
Large
language models become consequential
agents when surrounding systems let outputs change external state. Models now call tools, operate interfaces, delegate work, retain state, inhabit generated worlds, and control
robots or laboratory equipment. Such advances are often narrated as one march toward autonomy, conflating model competence, system integration, persistence, and safe authority. This critical review synthesizes primary research and official technical specifications available by 31 August 2026. We organize the evidence along delegated authority, temporal persistence, and environmental coupling, while separating model, harness, and environment. Within the evidence examined, action-interface expansion is documented more convincingly than robust completion, recovery, authorization, or independent verification. Model Context Protocol and
Agent2
Agent improve interoperability but do not establish trustworthy delegation; multi-
agent organization adds specialization alongside cost and correlated failure. Persistent simulations and world models support training and planning but do not themselves demonstrate agency; robotics and self-driving laboratories establish bounded feasibility rather than unattended open-world reliability. We propose justified delegation as an analytical and normative heuristic, not an observed law or certified score: expand action scope only where evidence supports provenance, bounded authority, failure detection, safe recovery, and calibrated human control. This framing yields a research agenda for coupled model-harness evaluation, capability-based permissions, durable state, cross-
agent accountability, and staged physical validation.
Read the full abstract ▾
Robotics
Reasoning Without Inference Cost: Latent Semantic Scaffolding for Robot VLA Policies
Andrew Ting Yan Li, Zhuo Li, Zhelin Yang et al. (+3 more)
Vision-
language-action (VLA) models are trained by imitation and capture what action to take but not why; adding causal
reasoning improves manipulation, but current methods pay for it at inference time - generating
reasoning tokens or rolling out predicted future states at every step, a cost that compounds over long horizons. We ask whether this benefit can instead be captured during training and discarded before deployment. We introduce Latent Semantic Scaffolding (LSS), an auxiliary loss applied during human-demonstration pretraining that aligns a VLA's action-
token representations to text
embeddings of physical-
reasoning rationales through a small projection head. The head is dropped at inference, leaving the unmodified base policy with zero added cost. Our central finding concerns
alignment granularity: aligning each action
token to the rationale of its own manipulation phase (Dense LSS) rather than to a single pooled episode-level
embedding (Pooled LSS) yields representations that transfer markedly better to held-out tasks. Dense LSS attains both the best in-distribution success and the best transfer to tasks unseen during
alignment, whereas pooled
alignment over-specializes to the training task. A representational probe shows Dense LSS induces roughly twice the per-phase separability in the backbone, supporting that phase-local
alignment is the operative mechanism.
Read the full abstract ▾
All authors Andrew Ting Yan Li, Zhuo Li, Zhelin Yang, Zhipeng Dong, Quentin Rouxel, Fei Chen
Artificial Intelligence · Cross-listed from physics.med-ph
Attention-guided super-resolution of 4D flow MRI in carotid arteries
Ali Mokhtari and Dominik Obrist
Four-dimensional (4D) flow
magnetic resonance imaging (MRI) is a powerful non-invasive technique for visualizing and quantifying complex blood flow patterns in vivo. Despite its clinical promise, broader adoption is limited by low spatial resolution and sensitivity to noise, which restrict accurate assessment of critical hemodynamic biomarkers such as wall shear stress, pressure gradients, and turbulent kinetic energy. To overcome these challenges, we propose a deep learning-based super-resolution framework that integrates multi-scale feature extraction and
attention mechanisms to enhance the quality of 4D flow MRI data. The model was trained on a
dataset of 120 patients with 240 stenosed carotid arteries. High-resolution ground truth data were generated using patient-specific computational fluid dynamics (CFD) simulations based on segmented vascular geometries and physiologically realistic boundary conditions, and the resulting velocity fields served as targets for supervised learning. The proposed architecture uses convolutional block
attention modules (CBAM) to guide the network toward clinically relevant spatial features and to suppress noise in low-resolution inputs. Quantitative results show that the
attention-guided model substantially reduces the root mean square error (RMSE) compared with a baseline model without
attention, and qualitative velocity contour analysis confirms improved reconstruction of intricate flow patterns. These findings highlight the capacity of the model to restore high-fidelity flow fields under noisy conditions and support the use of deep learning to extend the clinical utility of 4D flow MRI for non-invasive hemodynamic assessment.
Read the full abstract ▾
Social and Information Networks
Finding Many Overlapping Dense Subgraphs Using Triadic Cohorts
Sabyasachi Basu and C. Seshadhri
Graphs are a standard representation for data in the social sciences, cybersecurity, computer infrastructure, bioinformatics, and more. Typical real-world graphs are sparse, meaning the average degree is small (in the tens, while the number of vertices is more than millions). When graph data is collected from a source, a major task is to perform data exploration. Thus, any region of “density" is of interest, since it indicates special structure. An important goal is to cover a significant portion of the graph using dense subgraphs. Existing
algorithms that find many dense subgraphs do not have overlapping output, and hence provide limited coverage. Other methods that produce overlapping clusters do not generate dense subgraphs. The main goal of this paper is to develop provable and practical methods that can provide overlapping dense subgraphs that can cover large portions of real-world networks. Our contribution is an
algorithm CohortRecovery that achieves this goal. We first develop a mathematical framework of triadic cohorts that captures the notion of “detectable dense subgraphs” that potentially overlap. We prove that CohortRecovery can output a set of dense subgraphs, such that each triadic cohort is almost completely contained in some dense subgraph. We give a practical implementation of CohortRecovery and demonstrate it on a variety of
datasets. It typically runs in under ten minutes on a commodity machine even on graphs with tens of millions of edges. For numerous
datasets, CohortRecovery is able to cover more than 25% of the vertices in non-trivial subgraphs of density more than 0.8, and is significantly better than a wide variety of scalable graph clustering/community detection
algorithms. Moreover, we demonstrate that output of CohortRecovery captures ground truth clusters obtained by manual curation.
Read the full abstract ▾
Computer Vision
SimFuse3D: Source-Guided Target Simulation and Confidence-Guided Multi-Stage Localization Reweighting for Cross-Platform 3D Object Detection
Yongchun Lin, Xinliang Zhang, Yun Zou et al. (+7 more)
Changes in sensor height and viewpoint alter object-level point distributions, making cross-platform
LiDAR unsupervised domain adaptation (UDA) difficult. Self-training uses labeled source scans and unlabeled target scans, yet a retained prediction may provide a useful target location while enclosing sparse foreground returns, background clutter, or points inconsistent with the predicted box. We refer to this mismatch as box-point inconsistency. We introduce SimFuse3D, which preserves the target placement and repairs the associated pseudo object using measured geometry from labeled source scans. Object
Memory retrieves a similar labeled source instance. Target Simulation places the retrieved source geometry at the target location, aligns its points with the target viewing geometry, and filters the aligned crop to approximate the target observation. Confidence-Guided Multi-Stage Localization Reweighting (CMLR) maps each target pseudo-object confidence score to a bounded weight shared by RPN localization and R-CNN box regression. All components operate only during adaptation, leaving the detector architecture and inference graph unchanged. Across six cross-platform transfers, SimFuse3D consistently outperforms Pi3DET-Net and achieves the best performance among the compared adaptation methods on nearly all metrics. On nuScenes-to-KITTI, it ranks first among the compared adaptation methods with both evaluated detectors.
Read the full abstract ▾
All authors Yongchun Lin, Xinliang Zhang, Yun Zou, Zhixuan Xiao, Liang Lei, Jianya Guo, Yuqiang Zhai, Xiaofeng Wang, HaiKuo Xu, Haoang Li
Numerical Analysis
Energy-Consistent Splitting and Decomposition Approaches for Coupled port-Hamiltonian ODEs
Marius Mönch, Nicole Marheineke, Andreas Bartel et al. (+2 more)
Operator splitting provides an attractive approach for the numerical integration of (coupled) port-
Hamiltonian systems, as it allows the underlying system structure to be exploited at the level of the individual subproblems. However, the choice of the decomposition is not unique and may strongly affect both the computational efficiency and the preservation of the energy behavior of the original system. In this work, we investigate this interplay systematically and introduce energy consistency as a criterion for assessing splitting methods for port-
Hamiltonian ordinary differential equations. We derive sufficient conditions under which a splitting based on a given decomposition inherits the energy behavior of the continuous system and use these conditions to analyze several decomposition strategies for coupled port-
Hamiltonian systems. In particular, we compare decompositions that preserve the structure with approaches that exploit lower-dimensional subsystem dynamics or separated time scales. The analysis is complemented by numerical experiments using Strang splitting and its multiple-time-stepping extension. A scalable electro-thermal
benchmark with fast electrical and slow thermal dynamics is employed to assess accuracy, energy behavior, and computational efficiency. The results demonstrate that preserving the port-
Hamiltonian structure of the subflows is essential for energy-consistent splitting, whereas decompositions that exploit subsystem structure or time-scale separation can provide substantial computational advantages. In particular, the time-scale decomposition yields significant efficiency gains for systems with pronounced multirate characteristics, while structure-destroying decompositions may lead to undesirable energy behavior.
Read the full abstract ▾
All authors Marius Mönch, Nicole Marheineke, Andreas Bartel, Kevin Schäfers, Michael Günther
Information Retrieval
AtomRec: Evolving Atomic Memory for Agentic Recommendation
Peiyu Hu, Weihai Lu, Siying Gu et al. (+5 more)
Agentic recommender systems use large
language models to maintain semantic
memory and support evidence-aware recommendation. However, existing
memory mechanisms often compress user and item information into coarse summaries and connect them with scalar collaborative links, making it difficult to preserve fine-grained preference stages or retrieve interpretable evidence as user interests evolve. We propose AtomRec, an agentic recommender with evolving atomic collaborative
memory. AtomRec represents user and item
memories as structured atomic units, builds semantic links across related
memories, and evolves related historical fields when new interactions arrive. During recommendation, it retrieves linked
memories as multi-hop evidence paths rather than isolated neighbor summaries, allowing collaborative signals to support grounded ranking. Experiments on four public
benchmarks show that AtomRec consistently outperforms state-of-the-art agentic and
memory-augmented baselines, with around 8.5% average relative improvement across metrics.
Read the full abstract ▾
All authors Peiyu Hu, Weihai Lu, Siying Gu, Zhuodong Liu, Zhaokai Luo, Yuean Niu, Zhiyong Wang, Jia Wang
Machine Learning
From Deep to Shallow: Unconstrained and Efficient Layer Merging Strategy
Petro Shulzhenko, Gabriele Spadaro and Enzo Tartaglione
Although Deep Neural Networks have become foundational in many areas of Machine Learning, high computational demands limit their application in resource-constrained environments. To address this issue, depth compression methods have been proposed to identify and linearize redundant activation functions, thereby allowing for the merging of layers without intermediate non-linearities. However, these methods face two key challenges: they cannot be directly applied to convolutions with padding due to the absence of an analytical solution for merging these layers, and they typically increase the kernel size of merged layers, thus limiting speed-up gains. To overcome these limitations, we propose an efficient strategy that enables merging of layers without an existing analytical solution, and also without increasing kernel size. We validate our approach across multiple architectures and
datasets, and measure inference speed-up gains on real embedded platforms. We publicly released the code at https://github.com/ShulzhenkoPetr/deep-to-shallow.
Read the full abstract ▾
Artificial Intelligence
Reinforcement Learning for Sequential Solar PV Policy Design under Uncertainty: An Agent-Based Approach
Iias Faiud, Jonaid Shianifar, Michael Schukat et al. (+1 more)
Designing effective and fiscally sustainable policies for
solar photovoltaic (PV) adoption requires balancing adoption gains against public expenditure under uncertainty and heterogeneous decision-making. This study formulates PV policy design as a sequential decision problem and integrates reinforcement learning (RL) with a
stochastic agent-based model (ABM) that simulates yearly
solar PV adoption under uncertainty. A policymaker
agent selects annual incentives, including capital grants, subsidised loan rates, and feed-in tariffs, over a 16-year horizon. Adoption–cost trade-offs are explored by varying policy preferences within a scalarised reward framework. Policies are learned using PPO, SAC, and TD3 and evaluated under
stochastic simulation. The results show that this approach produces a clear trade-off structure: the highest-adoption policy (TD3, $w_{\text{cost}}=0.5$) achieves approximately 4,145 adopters at a cost of EUR 41.73 million, while the lowest-cost policy (PPO, $w_{\text{cost}}=2.0$) reduces expenditure to EUR 7.27 million with 2,682 adopters. The balanced policy (PPO, $w_{\text{cost}}=1.6$) achieves 3,495 adopters at a cost of EUR 22.47 million. Across
algorithms, consistent trade-off patterns are observed, indicating robustness of the adoption–cost relationship. Compared with static baseline policies, the RL framework explores a broader range of policy configurations. These findings demonstrate the potential of RL as a flexible tool for adaptive policy design under uncertainty.
Read the full abstract ▾
All authors Iias Faiud, Jonaid Shianifar, Michael Schukat, Karl Mason
Information Theory
Near-Field Physical-Layer Authentication Under Impersonation Attacks
Hajar El Hassani, Linda Senigagliesi and Arsenia Chorti
This paper studies physical-layer authentication (PLA) in the near-field regime under impersonation attacks. Unlike the far-field case, where the steering vector depends only on the angle of arrival (AoA), in the near field it depends on both angle and distance. We analyze the attack by minimizing the mean-square error (MSE) between the signal received from a legitimate transmitter Alice and the signal generated by an active attacker Eve. For a single-
antenna Eve, we derive the optimal scalar precoder and show that, for an inter-element spacing no larger than half a wavelength and under the standard second-order Fresnel approximation, perfect impersonation is possible only if Eve has the same angle and the same distance from Bob as Alice. We then extend the analysis to a multi-
antenna Eve and derive the optimal precoding vector. In this case, perfect impersonation is possible only if Alice steering vector belongs to the subspace spanned by Eve steering vectors. Simulation results show that, in the near field, a distance difference is sufficient to prevent a successful impersonation attack even when Alice and Eve have the same AoA, and confirm the analytical results.
Read the full abstract ▾
Cryptography and Security
ReCAST: Restoration-aware Cascaded Stage-wise Training for Obfuscated SMS Risk Classification
Jieyun Huang, Yi Shen, Kaikai Zhao et al. (+7 more)
Fraudulent messages sent via Short Message Service (SMS) are increasingly obfuscated to evade cost-conscious classifiers in production systems. In Chinese SMS, attackers can exploit a wide range of carefully crafted obfuscation strategies to hide risk-bearing phrases while preserving human readability, making direct classification brittle under real-world latency and throughput constraints. We propose ReCAST, a Restoration-aware Cascaded Stage-wise Training framework for robust obfuscated Chinese SMS classification. ReCAST distills a large teacher model's de-obfuscation ability into a smaller deployable student model by supervising obfuscated span detection, obfuscation type prediction, and text restoration, and then uses the restoration-aware student for downstream risk classification. Experiments on an internally constructed real-world Chinese SMS
benchmark show that ReCAST substantially improves classification performance over directly trained baselines under obfuscation. The results suggest that restoration-aware
distillation offers a practical path toward robust SMS risk classification with smaller deployable models under production-oriented constraints.
Read the full abstract ▾
All authors Jieyun Huang, Yi Shen, Kaikai Zhao, Jiangze Yan, Wenjing Zhang, Ping Chen, Ning Wang, Zhaoxiang Liu, Kai Wang, Shiguo Lian
Artificial Intelligence
MARLA: A Conceptual Scaffold for Regulatory Learning under the EU AI Act
Alessio Buscemi, Tom Deckenbrunnen, Imane Hmiddou et al. (+5 more)
The EU AI Act positions regulation as part of the infrastructure for safe, trustworthy and
market-ready innovation. Realising this ambition requires regulatory learning: the evidence generated during implementation must be translated into governance and legal knowledge that supports consistent interpretation, effective oversight, and adaptation as technologies evolve. Yet the actors who produce this evidence and those who rely on it operate in different professional worlds. This paper proposes MARLA (Map, Assess, Report, Learn, Adapt), a conceptual scaffold organising regulatory learning as a five-stage cycle centred on the implementation of legal requirements into socio-technical practices, situated at the Local, National and European levels of the AI Act's governance architecture. Deliberately non-prescriptive, MARLA gives technical and legal stakeholders a shared vocabulary in which each of the first three stages generates its own documentable form of regulatory learning. We illustrate the scaffold with two piloted case studies and a prospective National-to-European illustration.
Read the full abstract ▾
All authors Alessio Buscemi, Tom Deckenbrunnen, Imane Hmiddou, Marco Billi, Livio Rubino, Silvia Rizzuto Ferruzza, Daniele Pagani, Antonino Rotolo