Peer Revenue Fully funded

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science

cs Watch Computer Science

Tuesday 8 September

Computer Vision

CoVeR: Coverage-Based Token Pruning for Multi-View 3D Reasoning in VLMs

Nhat-Tan Bui, Varshini Elangovan, Arun Reddy Anugu et al. (+7 more)

Representing a 3D scene as multi-view images allows 2D VLMs to reason in 3D by reusing priors from pre-training, sidestepping the scarcity of annotated 3D data. However, it produces thousands of redundant visual tokens whose cost grows with every view. Existing visual token pruners fall into two families, each limited in the 3D multi-view setting. Learned importance methods rank tokens by attention or encoder features; because redundancy here is fundamentally spatial, they keep near-duplicate tokens from a few prominent regions and leave most of the scene unrepresented. Voxelization methods improve spatial coverage but cannot enforce an exact token budget and saturate as multi-view observations overlap in 3D, capping retention well below the target. We show that spatial coverage is associated with 3D reasoning performance and introduce CoVeR, a deterministic, training-free selector that uses only token coordinates, with no learned signals. CoVeR selects tokens that collectively cover every region of the scene, and solves the limitations of both families: it enforces an exact per-scene budget, breaks the voxelization saturation plateau, and avoids the near-duplicate selections of learned importance. Extensive experiments show CoVeR outperforms prior SOTAs on all three 3D reasoning benchmarks and generalizes as a plug-and-play module tested across four VLMs. Notably, with only $\approx$8% of visual tokens, it preserves 93.5% of full-token performance, surpassing SOTA by 3.9 percentage points on average across benchmarks.
Read the full abstract

All authors Nhat-Tan Bui, Varshini Elangovan, Arun Reddy Anugu, Sreyas Mohan, Wei Ye, Dilin Wang, JQ Huang, Rakesh Ranjan, Aviral Chharia, Fernando De la Torre

Computer Vision

VeriScene: Reconstructing Crime Scenes from Legal Evidence via World-Model Agent

Kevin Chuanpu Fu, Yongsen Zheng, Zee Kin Yeong et al. (+1 more)

World models take multimodal inputs like text, photos, and diagrams to generate dynamic scenes in accordance with the laws of physics, thus opening a compelling application: fusing multimodal legal evidence to re-create a crime scene and re-enact how an offence could have been committed. However, feeding the raw, unorganized evidence into a world model fails in forensic use: it silently drops evidence, glosses over contradictory testimony, and produces motion that violates the evidentiary record. This paper presents VeriScene, an agent that orchestrates the world model: it reconstructs crime scenes from forensic photographs and witness statements of varying reliability, keeping every claim traceable to evidence and every motion physically plausible. VeriScene iteratively fuses the evidence into a cited narrative under an auditing loop, verifies the hypothesized dynamics via probe rollouts in the world model with corrective constraint injection, and renders the offence as a re-enactment video from a fused keyframe. On a benchmark of 25 crime scenarios across 7 physically-driven case types (139 forensic-style photographs and 65 statements with planted unreliability), VeriScene attains 0.9014 evidence coverage and 0.7217 factual consistency (0-1 scale) on the 20 test scenes, outperforming an end-to-end multimodal-LLM baseline by 20.35% in factual consistency and 34.88% in temporal coherence, while generalizing across four LLM orchestration backends at USD 1.82 per scene.
Read the full abstract

All authors Kevin Chuanpu Fu, Yongsen Zheng, Zee Kin Yeong, Kwok-Yan Lam

Machine Learning

TV-Regulated OPD: Direction Matters in On-Policy Distillation

Han Xiao, Yifan Niu, Dongyi Liu et al. (+2 more)

On-Policy Distillation (OPD) facilitates the transfer of knowledge from domain expert to student in the post-training phase of Large Language Models (LLMs). However, the supervision signals in mainstream OPD methods suffer from high variance and noise which is generally instable during training. In this work, we systematically investigated what really matters to the performance and the fundamental mechanisms behind the instability during training. We found that retaining only the sign of token-level advantages is sufficient to achieve the performance comparable to standard OPD. Meanwhile, smoother and bounded advantages can stabilize the training process without sacrificing its performance. These motivated us to shape the advantages using the Total Variation (TV) and propose a robust TV regulated On-Policy Distillation (TV-OPD) method. Benefiting from the bounded and diminished advantages, TV-OPD exhibits stable training dynamics and steady late-stage performance. We conducted comprehensive experiments and found that, across various settings, TV-OPD consistently achieved better performance and lower variance in the late-stage of training.
Read the full abstract

All authors Han Xiao, Yifan Niu, Dongyi Liu, Chang Luo, Jia Li

Robotics

RoboCousin: Build Your Own Simulation Playground for Robust Bimanual Robotic Manipulation

Jingxuan Zhu, Jingyi Li, LiangLiang Chen et al. (+3 more)

Bimanual manipulation policies require large and diverse training datasets, yet collecting demonstrations on physical robots is expensive and difficult to scale. Simulation can generate data efficiently, but existing pipelines typically operate within closed asset libraries and predefined scenes: adding a newly observed object or environment still requires substantial effort to reconstruct geometry, specify physical and semantic properties, annotate interactions, and integrate the result into executable tasks. We present RoboCousin, an extensible simulation-based data-generation platform that turns user-provided observations into reusable assets, scenes, and expert trajectories for bimanual manipulation. Built on RoboTwin~2.0, RoboCousin converts object images into simulation-ready assets with visual and collision geometry, semantic and physical metadata, and automatically generated grasp-contact candidates. It further constructs digital cousins that vary compatible objects, backgrounds, layouts, and language instructions while preserving task-relevant affordances and spatial relations. The same asset system supports tabletop and room-level scene construction, with collision-aware base control for interaction beyond a fixed workspace. We release RoboCousin-OBD, containing more than 3,000 annotated object instances and 50 background environments, and use RoboCousin to generate over one million expert trajectories across 50 tasks. Simulation and real-robot experiments show that the automatically generated interaction annotations are comparable to curated annotations, generated assets provide effective sim-to-real supervision, and tabletop cousins can improve transfer beyond training on a single reconstructed scene. RoboCousin therefore provides a practical path for expanding both the scale and coverage of synthetic bimanual manipulation data.
Read the full abstract

All authors Jingxuan Zhu, Jingyi Li, LiangLiang Chen, Zhiyuan Jing, Jidong Zhang, Hongming Li

Robotics

A Multi-Modal Perception Pipeline for Object Detection and Tracking in Autonomous Racing

Davide Malvezzi, Michele Pestarino, Vittoria Cavicchioli et al. (+9 more)

Object detection and tracking are fundamental components of perception systems for autonomous driving. Achieving robust performance under adverse conditions such as limited visibility, sensor noise, and failures remains an open challenge, particularly in autonomous racing, where vehicles operate at very high speeds, experience strong vibrations, and interact under small safety margins. This paper presents a multi-modal late-fusion perception pipeline for object detection and tracking in the autonomous racing domain. The proposed system extends previous work by exploiting all onboard sensors through a late-fusion approach and a dedicated multi-object tracking framework. Independent detections from cameras, LiDARs, and RADARs are combined to provide timely and robust state estimates of surrounding vehicles. The tracking method explicitly compensates for detection delays and embeds in its model prior knowledge of vehicle dynamics and track layout. Experimental evaluation on real-world data across diverse critical scenarios, representative of challenging edge cases also in urban driving, confirms the effectiveness of the proposed pipeline and its suitability to support safe and adaptive planning decisions.
Read the full abstract

All authors Davide Malvezzi, Michele Pestarino, Vittoria Cavicchioli, Valentina La Gamba, Silvia Severi, Fabio Bagni, Luca Bartoli, Massimiliano Bosi, Francesco Gatti, Micaela Verucchi, Ayoub Raji, Marko Bertogna

Machine Learning

Distillation as Probability Transport: Routed On-Policy Distillation

Tianle Xia, Lingxiang Hu, Yiding Sun et al. (+6 more)

On-policy distillation (OPD) transfers teacher knowledge on student-generated trajectories, but efficient sampled objectives reduce the teacher distribution to scalar credit on individual tokens. Such credit indicates whether a token should gain or lose probability, yet leaves the corresponding redistribution unspecified. We recast OPD as teacher-guided probability transport and propose RouteOPD (Routed On-Policy Distillation), which decomposes local teacher–student disagreement into student-excess sources and teacher-deficit destinations and couples them into explicit transport pairs. RouteOPD optimizes pairwise log-odds toward jointly realizable targets obtained from a bounded teacher potential, while adapting the transport budget to the concentration of teacher demand. This formulation directs updates toward teacher-preferred destinations and controls their magnitude within a single transport operator. Experiments across four teacher–student settings and four mathematical-reasoning benchmarks demonstrate that RouteOPD consistently outperforms sampled reverse-KL OPD, with improvements accompanied by higher routing fidelity and lower background leakage. These results demonstrate the effectiveness of explicitly modeling probability transport in on-policy distillation.
Read the full abstract

All authors Tianle Xia, Lingxiang Hu, Yiding Sun, Linfang Shang, Ming Xu, Lan Xu, Ning Zheng, Wei Xu, Jie Jiang

Machine Learning · Cross-listed from physics.chem-ph

Fixed-Dimensional Latent Flow for Generating Variable-Size 3D Molecules

Weichi Yao, Cameron Gruich, Bryan R. Goldsmith et al. (+1 more)

Molecular size is intrinsically linked to composition, structure, and properties, but most 3D molecular generators treat size as a predefined input rather than a quantity to be generated. We introduce Equivariant-Free Transformer-Autoencoded Latent Flow Matching, a two-stage framework that samples a fixed-dimensional latent vector using flow matching and employs an autoregressive Transformer to determine molecular size, atom types, coordinates, and chemical attributes. Canonical atom ordering and rigid-pose alignment enable Transformers without equivariant layers, while decoded attributes guide bond reconstruction. On PCQM4Mv2, unconditional generation yields 87.9% unique and novel molecules passing sanitization and PoseBusters checks, exceeding baselines with lower end-to-end training and sampling time and higher end-to-end throughput. Across ten target HOMO-LUMO gaps, internal ranking retains 30% of the screened candidates and increases the density functional theory-verified hit rate within 0.1 eV of the target from 25.0% to 52.4%, while largely preserving novelty and diversity. These results demonstrate fixed-dimensional latent generation with autoregressive decoding as a practical approach to molecular design without prespecifying size.
Read the full abstract

All authors Weichi Yao, Cameron Gruich, Bryan R. Goldsmith, Yixin Wang

Computer Vision

EdMCGS: Event-Driven Markov Chain Gaussian Splatting for Extreme-Low-Frame-Rate Dynamic Scene Reconstruction

Yuzhong Wang, Wenmin Wang and Xinxing Yu

We present EdMCGS (Event-driven Markov chain Gaussian Splatting), an end-to-end method for reconstructing dynamic 3D scenes from extreme-low-frame-rate RGB together with an event stream, which can then be rendered at any intermediate timestamp. Methods relying solely on RGB images generate numerous artifacts due to the lack of evidence from between consecutive frames. To supply this missing evidence, we model the scene motion as an event-driven Markov chain, in which the sparse RGB frames anchor the state at their own timestamps while the events recorded within an interval drive the transition across it. Since the transition reads the events of the current interval, it remains active at inference and produces the in-between motion of the 3D Gaussians directly from the events rather than by interpolation, which sets our method apart from prior work that uses events only as training-time supervision. The state is carried by a compact set of control points, each driven by the events sampled in the neighborhood of its own image projection, and a temporal local isometry term keeps the propagated motion locally rigid. Experiments on synthetic and real-world scenes show that EdMCGS outperforms both RGB-based and event-based baselines, while rendering in real time with far fewer Gaussians than the strongest event-based baseline. We release our source code and a new dataset at https://github.com/joseclipse/EdMCGS.
Read the full abstract

Computer Vision

Do Input-Level Defenses Transfer to Observation-Level Attacks on VideoLLMs?

Bangshuo Zhu, Wei Song, Yuxin Cao et al. (+4 more)

Video Large Language Models (VideoLLMs) are increasingly deployed in safety-critical applications such as content moderation and video analytics. To process long videos efficiently, VideoLLMs rely on frame sampling, token compression, and modality fusion, which together form an observation pipeline that reduces the raw video to a compact internal representation. Recent observation-level attacks exploit this pipeline to prevent the model from perceiving harmful content, yet no defense has been explicitly designed for this threat. We introduce DefTEval, a controlled evaluation framework that systematically assesses whether input-level adversarial defenses, which operate on the pixel content of already-sampled frames, can mitigate observation-level attacks. Across five VideoLLMs, eleven representative defenses, and five attack types, we find that input-level defenses offer limited and inconsistent protection, with harmful detection rates frequently near zero. Critically, defenses fail even against attacks that embed harmful signals in every sampled frame, indicating that the bottleneck extends beyond sampling omission to the suppression of signals that do enter the model. Token compression discards localized features, and modality fusion systematically down-weights weakened visual signals. Furthermore, defense effectiveness is dominated by model architecture rather than by the defense method itself, and detection rates vary drastically across content categories, exposing structural weaknesses in temporal reasoning. These findings demonstrate that securing VideoLLMs requires system-level robustness mechanisms spanning sampling-aware coverage guarantees, token-level preservation of safety-relevant features, and modality-balanced fusion.
Read the full abstract

All authors Bangshuo Zhu, Wei Song, Yuxin Cao, Yuezhong Wu, Zhiquan Liu, Yuekang Li, Jingling Xue

Machine Learning

EMBLEM: Enhancing Multi-script Table Detection through Masking

Dhruv Kudale, Udhay Brahmi and Ganesh Ramakrishnan

Table detection is a core task in document analysis, supporting downstream applications such as information retrieval, document reconstruction, and visual question answering. While existing deep learning models perform well on English and Chinese documents, they struggle with multilingual, multi-script documents due to script diversity and the limited availability of labeled data. To address this challenge, we introduce MANDALA (Multi-script Annotated Documents for Table Detection), a manually curated dataset of 2,323 table-containing pages spanning 18 languages and 15 scripts across diverse domains. We also propose EMBLEM, a masking-based paradigm for Multi-script Table Detection (MTD). EMBLEM generates masked images that conceal script- and font-specific details, enabling models pre-trained on abundant English documents to focus on script-agnostic page layout. Experiments across three table detection architectures show that EMBLEM consistently outperforms strong baselines on MANDALA while remaining competitive on five standard English-dominant benchmarks. Using only English masked images for fine-tuning, with no multi-script training data, EMBLEM achieves an absolute F1-score gain of 20.8% on MANDALA. We release MANDALA along with the accompanying code and models at https://github.com/IITB-LEAP-OCR/EMBLEM.git.
Read the full abstract

Software Engineering

Tool Retrievers Are Underestimated: Annotation Expansion Reveals True Capability

Yanyu Zhu, Chenheng Zhang, Shaoshen Chen et al. (+8 more)

In open-world scenarios with massive and evolving tool repositories, tool-augmented large language models rely on a retriever to surface relevant tools for a given query. Because such repositories often contain many tools that implement the same functionality, a single query can often be resolved by several distinct but functionally equivalent tool combinations, making the natural query-to-tool mapping inherently one-to-many. However, existing tool retrieval benchmarks annotate each query with a single relevant tool combination, collapsing this one-to-many mapping into a rigid one-to-one annotation and causing valid retrieved tools to be misjudged as failures. To address this, we propose ToolEX (Tool Equivalent eXpansion), a framework that automatically discovers and annotates the tool combinations functionally equivalent to the labeled ones. Applied to the 7,360-query Tool-DE benchmark, ToolEX finds that 67.9% of sub-queries admit equivalent alternatives, expanding the singular ground truth to an average of 5.3 valid combinations per query. Using the expanded benchmark ToolEQ, we re-evaluate eight base retrievers and two fine-tuned variants; metrics on ToolEQ rise substantially over Tool-DE, showing that one-to-one annotation systematically underestimates retrievers and that 30–47% of the reported fine-tuning gain is an evaluation artifact rather than genuine improvement. Applying the same pipeline to skill retrieval on SkillRet further confirms that the one-to-one problem extends beyond tool retrieval.
Read the full abstract

All authors Yanyu Zhu, Chenheng Zhang, Shaoshen Chen, Hoilam Pao, Yufei zhang, Jiajun Chai, Dongnian Wang, Zhaoyu Hu, Guojun Yin, Wei Lin, Hai-Tao Zheng

Logic in Computer Science

Width-Bounded Equational Derivations for Finite Graph Expressions

Antonios Kalampakas

Completeness of an equational presentation guarantees an equality path but need not control the resources used along it. For finite graph expressions we measure derivational space by the largest input-output interface of an intermediate raw term. For every finite doubly ranked edge alphabet $\Sigma$, we prove that equal closed expressions of pattern width at most $k$ are joined by a derivation in which every step applies an equation in either direction and every intermediate width is at most a computable $B_\Sigma(k)$, independently of graph size. The derivation uses only the structural magmoid laws and the fifteen finite-graphoid schemes. The construction compiles each expression through protected cores and finite routing windows to an encoding-canonical representative $\operatorname{NF}_k$ of width at most $4k+4$. We call this property bounded equational coherence. We also prove $\operatorname{bw}(F)\le\operatorname{patw}(F)\le4(\operatorname{tw}(F)+1)$ whenever the underlying simple graph has at least two edges, separate layered linear from branching expressions on cliques, and provide machine-checkable witnesses and finite invariants for the first five nontrivial clique values.
Read the full abstract

Computation and Language

Tracing Stereotypes from Representation to Output in Multilingual LLMs

Ariun-Erdene Tumurchuluun, Yusser Al Ghussin, Pinzhen Chen et al. (+2 more)

Multilingual LLMs show stereotype-related behavior that varies across languages, but behavioral scores do not show where the relevant information is represented or how it affects the output. To investigate these internal mechanisms, we compare linear probing, attribution patching, sparse autoencoders (SAEs) and feature ablation in Llama-3.1-8B, Qwen3-8B, and Gemma-2-9B. Probe performance peaks substantially earlier than attribution in all three models, with a separation of 36-53% of model depth. Retained Llama-Scope features often match the social category on which they were selected and form recurring semantic families, but their lexical alignment and ablation effects vary across SAE suites. Only 6-18% of evaluated residual-stream features have language-agnostic effects under our criterion, and none are category-agnostic. Language-agnostic features have larger mean ablation effects in Llama-Scope, but this pattern does not repeat in the other SAE suites. Decodability, output influence, and cross-lingual ablation effects therefore need to be measured separately.
Read the full abstract

All authors Ariun-Erdene Tumurchuluun, Yusser Al Ghussin, Pinzhen Chen, Josef van Genabith, Koel Dutta Chowdhury

Human-Computer Interaction · Cross-listed from eess.SP

Understanding emotional and behavioural responses through physiological signal analysis recorded during theatrical performance

Aditi Site, Annariina Lohiranta and Tarmo Lipping

Among various cultural interventions, theatre offers a powerful medium for emotional expression, reflection, and social connection. Live performances can positively influence mental well-being. Therefore, studying audience responses during theatrical performances can provide valuable insights into the relationship between theatre, emotional experiences, and well-being. Physiological responses have emerged as an important modality for identifying patterns associated with different emotional states. However, in naturalistic environments such as live performances, analysing these signals is challenging due to the absence of explicit labels. In this study, we focus on understanding emotional and behavioural responses using physiological data collected during a live theatrical performance. Data were recorded from participants using multiple sensing modalities. In addition, a comprehensive analytical framework was employed, which included demographic analysis, visual exploration of physiological signals, participant-level observations, scene-wise physiological analysis, and intensity-based analysis using z-score normalization to summarize the findings. The findings revealed that dramatic, surprising, and intense scenes produced the strongest activation, often with concurrent increases in HR and SCL, whereas empathy-driven scenes showed lower and more stable responses. Humorous scenes elicited moderate-to-high activation, particularly in the presence of sudden reactions or loud sounds. Analyses also identified shared physiological response patterns across participants, with activation concentrated around key emotional events. These results indicate that combined HR and SCL measures can effectively capture variations in audience engagement during live theatrical performances.
Read the full abstract

Numerical Analysis

Compact implicit high-resolution numerical scheme for multicomponent transport problems with Langmuir sorption

Dagmar Zakova and Peter Frolkovic

We present a compact implicit high-resolution finite volume scheme for one-dimensional multicomponent transport problems with nonlinear Langmuir sorption. The method combines a second order compact implicit discretization with WENO reconstruction and a time limiter for the treatment of steep gradients and discontinuities. A special attention is given to the Courant number used in the limiter, where local characteristic Courant numbers based on the eigenvalues of the inverse retardation matrix are considered. The compact implicit schemes are compared with analogous explicit first and second order methods. Numerical experiments show second order convergence for smooth solutions and demonstrate that the implicit approach can be used with significantly larger Courant numbers. The method is also applied to a three-component displacement chromatography problem with sharp concentration fronts.
Read the full abstract

Software Engineering

AttnCompress: Dynamic Attention-Guided Trajectory Compression for Software Engineering Agents

Zhengran Zeng, Yixin Li, Rui Xie et al. (+2 more)

The transition from human-centric assistance to Autonomous Software Engineering (ASE) agents has enabled the resolution of complex real-world SE tasks. However, the trial-and-error nature of these agents generates lengthy interaction trajectories, creating severe bottlenecks in terms of context window limits and cost. While context compression offers a potential remedy, prior approaches suffer from static pruning strategies and granularity mismatches, often failing to preserve the semantic dependencies and syntactic details crucial for SE tasks. To strictly preserve critical task evidence while reducing context length, we introduce AttnCompress, a dynamic attention-guided trajectory compression framework. Unlike existing approaches, AttnCompress bridges the gap between semantic integrity and dynamic adaptability through three key mechanisms: (1) structure-aware segmentation via perplexity (PPL) spikes to preserve the syntactic structure of code and logs; (2) relevance estimation using proxy attention weights to quantify the precise relevance of historical blocks to the agent's current reasoning; and (3) a dynamic rolling window to re-evaluate and recall historical context as the task evolves. Extensive evaluation on SWE-Bench-Verified and Multi-SWE-Bench demonstrates that AttnCompress achieves a pass rate of 53.17%, outperforming prior state-of-the-art baselines while reducing token consumption by 21.6% and total costs by 33.6%. The framework proves to be model-agnostic and generalizes effectively across diverse programming languages.
Read the full abstract

All authors Zhengran Zeng, Yixin Li, Rui Xie, Wei Ye, Shikun Zhang

Computer Vision

Supervised Cross-Modal Feature Alignment for Zero-Wearable Freezing of Gait Detection in Parkinsonism

Aryan Singh and Chandan Biswas

Objective assessment of Freezing of Gait (FoG) in Parkinson's disease (PD) relies predominantly on wearable Inertial Measurement Units (IMUs). While IMUs provide optimal kinematic precision, mandatory sensor attachment restricts continuous clinical deployment. Conversely, unobtrusive vision-based alternatives suffer substantial classification errors during turning-in-place tasks, where geometric self-occlusion degrades deterministic skeletal coordinates and obscures the high-frequency precursors required for FoG detection. To resolve these physical observation limits, we propose a supervised cross-modal subspace distillation framework. During optimisation, pre-trained kinematic data from IMU sensors and contextual clinical metadata act as oracles to guide a deployable visual architecture. By incorporating joint velocity and acceleration derivatives, utilising a confidence-based gating mechanism, the visual model mitigates some of the tracking errors during occlusion events. Empirical evaluations confirm this latent alignment transfers the predictive fidelity of hardware sensors directly into the visual representation, yielding $85.5\%$ accuracy, and $82.4\%$ balanced accuracy. All the while maintaining a vision only model at inference.
Read the full abstract

Computer Vision

From Glance to Scrutiny: Progressive Distortion Reasoning for Fine-Grained Image Quality Assessment

Aoting Zhang, Mingze Gao, Dongbao Yang et al. (+5 more)

Multi-modal large language models (MLLMs) have demonstrated significant potential in image quality assessment (IQA) by bridging visual perception with descriptive evaluations. However, existing approaches mainly focus on holistic quality prediction, often functioning as black boxes that provide limited insight into where distortions occur and how they affect perceived quality, hindering fine-grained analysis of localized and heterogeneous degradations. We propose GS-IQA, a framework that reformulates IQA as a progressive Where–What–How diagnosis, emulating the human perceptual process from an initial glance to closer scrutiny. Since a severity judgment is meaningful only for a correctly localized and recognized region, we realize this progression through a two-stage reinforcement learning paradigm that respects such dependencies: the glance stage uses a perception-gated reward to establish where degradations lie and what they are, activating severity feedback only once both are correct, while the scrutiny stage introduces online reward-conditioned degradation generation to synthesize hard examples targeted at the model's perceptual bottlenecks, sharpening its discrimination of subtle severity variations. To enable systematic evaluation, we construct Diag-Bench, a region-level IQA benchmark of about 25K curated samples spanning 12 distortion types and five ordinal severity levels. Extensive experiments show that GS-IQA consistently surpasses state-of-the-art methods in distortion localization, recognition, and severity estimation, and that its diagnostic representations transfer effectively to conventional global quality prediction across diverse external benchmarks. Code and data will be released.
Read the full abstract

All authors Aoting Zhang, Mingze Gao, Dongbao Yang, Longyi Chen, Daoxin Zhang, Yi Wu, Yao Hu, Yu Zhou

Numerical Analysis

A discontinuous finite element method for the hydroelastic analysis of submerged structures

Zachary J. Wegert, Oriol Colomés, Michael H. Meylan et al. (+2 more)

This paper develops a finite element method for wave-structure interaction problems arising in the hydroelastic modelling of submerged elastic plates. The approach is formulated for both two-dimensional settings and three-dimensional channels. For the fluid part of the problem, we use a symmetric discontinuous Galerkin scheme that captures the discontinuity in the potential across the plate. To impose the Sommerfeld radiation conditions, we take inspiration from the Dirichlet-to-Neumann map approach and use the analytic solution in the semi-infinite domains to construct appropriate non-local boundary conditions. For the plate, we use a continuous/discontinuous Galerkin method to resolve the 4$^{\rm th}$ order operator without using continuously differentiable finite elements. We prove well-posedness of the method, showing that the resulting sesquilinear form is bounded and satisfies the Gårding inequality, leading to a unique and stable solution. We implement our formulation as an open-source tool using the Gridap package in Julia. We validate our method against a hypersingular boundary integral method, finding excellent agreement. The flexibility of the methodology is demonstrated through simulations of a two-dimensional slanted plate and three-dimensional plate geometries including centred rectangular plates, offset rectangular plates, and annulus-shaped plates. The finite element method presented here can be readily applied to model wave energy conversion of piezoelectric bimorphs. Furthermore, the method could easily be extended to consider wave scattering from fixed or rigid structures or from variable bottom topography, and is not limited to hydroelastic applications.
Read the full abstract

All authors Zachary J. Wegert, Oriol Colomés, Michael H. Meylan, Ngamta Thamwattana, Vivien J. Challis

Information Theory

Non-Coherent Over-the-Air Federated Learning: Protocol, Convergence, and Device Scheduling

Haifeng Wen, Nicolò Michelusi, Osvaldo Simeone et al. (+2 more)

To mitigate the scalability bottleneck in the radio access network (RAN) in federated edge learning (FEEL), over-the-air federated learning (AirFL) exploits waveform superposition over multiple-access channels (MACs) for analog model aggregation. However, coherent AirFL typically relies on stringent PHY-layer conditions such as accurate channel state information (CSI), tight time/frequency synchronization, and frequent transceiver calibration for signal alignment. However, these requirements, if not impossible to be met, incur substantial communication and computation overhead. In this paper, we propose a non-coherent AirFL (NCAirFL) protocol over a broadband single-antenna MAC, leveraging binary dithering, unbiased non-coherent detection, and long-term error feedback to waive the need for instantaneous CSI. For NCAirFL with general smooth non-convex objectives and a constant learning rate, we establish a convergence bound achieving the convergence rate in the same order of $\mathcal{O}(1/\sqrt{T})$ as communication-ideal FedAvg, where $T$ is the total number of communication rounds. To further improve communication efficiency under data and wireless resource heterogeneity, we also derive a lower bound on the expected single-round objective decrease in the global loss conditioned on device scheduling, building upon which a surrogate objective function is obtained for jointly optimal device selection and power control. Experimental results on MNIST and CIFAR-10 corroborate that NCAirFL achieves learning performance close to FedAvg in practical settings, with the proposed device scheduling policy substantially accelerating convergence.
Read the full abstract

All authors Haifeng Wen, Nicolò Michelusi, Osvaldo Simeone, Yang Yang, Hong Xing

Human-Computer Interaction

StitchOver: Technical Embroidery on Seamed Fabrics

Zekun Chang, Tianhong Catherine Yu, Yixuan Gao et al. (+1 more)

Smart textiles embed interactivity into everyday garments, supporting use cases like always-available sensing for medical applications or sports. Machine embroidery allows integrating functionalities into existing textiles. However, embroidering onto real-world textile goods remains challenging. Textile goods are rarely made of a single homogeneous substrate of fabric, and embroidery with functional materials such as conductive threads requires machines to be more tightly calibrated than for decorative embroidery. In particular, seams, which bring together different substrates, along with machine variability, cause shifts in tension and friction between the functional thread and the textile substrate that frequently lead to defects (70% of samples in our evaluation). We present a technique to reliably embroider on seamed fabric even when using functional threads. Our software tool automatically digitizes user-defined stitch patterns by introducing what we call "JumpStitches" to bypass seam interference. We evaluated our approach under varying machine states (under-tensioned, well-calibrated, and over-tensioned), and across multiple seam and pattern configurations. Our results show that the JumpStitch mechanism eliminates defects, while maintaining conductivity compared to 70% defects without JumpStitches, and even in poorly calibrated machine states continues to work well.
Read the full abstract

All authors Zekun Chang, Tianhong Catherine Yu, Yixuan Gao, Thijs Roumen

Information Retrieval

Exploring Bottom-Up Clustering for Creating Semantic IDs

Leah Woldemariam, Sudhanshu Garg, Taha Belkhouja et al. (+2 more)

The success of generative retrieval has largely been attributed to the use of Semantic IDs, which improve over arbitrary item-level identifiers such as hashes by capturing the semantics of items. The main challenges faced when constructing Semantic IDs, however, is in mapping each identifier to a unique product and capturing information valuable to downstream tasks. Past works have appended additional codewords to de-duplicate item identifiers and utilized residual quantization to create hierarchical clusters. In this work, we present an algorithm for generating Semantic IDs that ensure the identifiers are both unique and preserve the structure of the original embedding. Key to our work is the use of bottom-up clustering to preserve local structure in the embedding space, improving the clustering quality of the resulting Semantic IDs and their utility for downstream generative retrieval.
Read the full abstract

All authors Leah Woldemariam, Sudhanshu Garg, Taha Belkhouja, Charles Kim-Yip, Ali Sahami

Distributed Computing

A Measurement Study of LLM Inference Trade-offs Across Edge Continuum Hardware

Maysam Khatib, Moysis Symeonides, Demetris Trihinas et al. (+2 more)

Large language models (LLMs) are increasingly used as backends for intelligent web services, but serving them across the edge continuum requires balancing quality, latency, model footprint, and energy. This paper presents a controlled measurement study of self-hosted LLM inference across edge and near-edge deployment nodes: an NVIDIA Jetson AGX Orin and a near-edge server with CPU-only and GPU-enabled inference modes. We evaluate multiple open-weight LLMs and quantization variants using a fixed question-answering workload, and compare them against GPT-4o as a cloud-hosted accuracy and latency reference. Our benchmarking pipeline reports accuracy, model footprint, per-token decoding latency, prefill latency, and overall execution energy. The results show that GPU-enabled server execution provides the lowest compute-side latency, while Jetson Orin shows lower measured energy, consistent with its lower platform power under our setup. CPU-only execution is consistently dominated in latency for our workload and shows higher measured energy. We also show that parameter count and downloaded weight-file size alone do not reliably predict observed accuracy or latency. Finally, using Pareto-frontier analysis, we study how deployment decisions may change under possible streamed-token delivery overheads, highlighting that compute-side inference metrics alone can lead to suboptimal placement for latency-sensitive interactive web services.
Read the full abstract

All authors Maysam Khatib, Moysis Symeonides, Demetris Trihinas, George Pallis, Marios D. Dikaiakos

Cryptography and Security

HoneyRoute: Honeypot-Model Routing for Adversarial LLM Serving

Han Jin

We introduce HoneyRoute, an inference-serving layer that detects whether an incoming request is malicious and, if so, routes it to a dedicated honeypot model, shielding production while the adversary's interaction is continuously harvested for intelligence. Existing defenses embed traps inside model memory or rebuild deception at the protocol layer, leaving the serving tier unprotected and feeding nothing back into detection. HoneyRoute couples (i) a streaming router (a frozen 0.8B-embedding backbone with per-domain MLP heads), (ii) a dual-implementation honeypot (a rule/prompt-engineered code honeypot or a dedicated same-family replica), and (iii) an analysis loop that converts trapped interactions into attacker fingerprints for router retraining. On a production trace plus a seven-domain attack corpus, the router reaches F1=.911 at 38 ms median added latency, matching 96% of a two-tier guard-LLM cascade's F1 at 1/385 of its latency with 0% evasion under 13 adversarial transformations; diverting the malicious share cuts production-model token consumption under concurrent flooding with real GCG-suffix payloads by 97.8%; the trained replica agrees with the production model on 92.9% of benign holdout requests, while naive unconditional bait injection collapses to 7.6% and selective camouflaged injection recovers to 88.9%, mapping the recoverable fidelity-traceability frontier; and a loop-trained correction head cuts misrouting of legitimate security research 9x while raising detection F1 to .933.
Read the full abstract

Computer Vision

FPicker: Topology-Guided Evolution for Filament Tracing in Low-SNR Microscopy

Tingyin Zhao, Mingtao Huang and Yuan Shen

Automating filament tracing in Cryo-Electron Microscopy (Cryo-EM) is essential for 3D helical reconstruction but challenged by intersecting topologies and extremely low Signal-to-Noise Ratios ($\text{SNR} = \sigma_s^2/\sigma_n^2$ < 0.1 or -10 dB). Existing paradigms fail: pixel-wise segmenters suffer from severe topological fracturing, box-based detectors face ghost center drift, sequential trackers derail due to error accumulation, and traditional active contours collapse under artificial closed-curve constraints. To resolve these bottlenecks, we present FPicker, the first topology-guided framework reconciling these incompatibilities. It unifies perception via a center-endpoint representation and an open-curve evolution module to explicitly model non-cyclic connectivity. On simulated benchmarks, FPicker outperforms top baselines by over $40\%$ relative gain in mean spatio-angular precision (mSAP) and reduces topological gap rates by over $60\%$ under extreme noise ($-20\text{ dB}$). By learning intrinsic physical geometry rather than local texture, FPicker demonstrates strong potential as a resilient geometric backbone. Its zero-shot performance on the real-world EMPIAR dataset exhibits robust topological resistance, achieving a state-of-the-art 82.9% mSAP upon fine-tuning. Our results also suggest modeling physical priors is a highly robust path toward bridging the sim-to-real gap in signal-starved scientific imaging. The code is publicly available at: https://github.com/tomzhaosky/FPicker.
Read the full abstract

Computer Vision

TRIUNE-Net: Harmonizing Scale, Shape, and Efficiency in Pancreatic Tumor Segmentation

Amir Hossein Saleknia, Alireza Kheyrkhah, Sanaz Karimijafarbigloo et al. (+5 more)

Pancreatic tumor segmentation in 3D CT volumes is challenged by extreme scale variability across both the pancreas and tumor, and highly irregular tumor morphology. While recent advances have pushed segmentation performance, existing methods do not explicitly address these challenges and come at the cost of excessive computational complexity, limiting their practicality in resource-constrained clinical environments. We propose TRIUNE-Net, a lightweight unified architecture that harmonizes scale, shape, and efficiency through three synergistic innovations. A multi-scale context aggregation module with stage-adaptive dilated convolutions enables the model to reason across the broad range of anatomical scales present in both organs. A serial linear-deformable attention mechanism combines large effective receptive fields with shapeadaptive deformable convolutions to capture irregular, non-convex tumor morphologies. Finally, an information-preserving downsampling module replaces conventional max pooling entirely, retaining all spatial information while adding negligible parameters, preventing small tumors from being discarded before they can be recognized. On both the MSD Pancreas and NVD Pancreas datasets, TRIUNE-Net achieves state-of-theart results with only 5.86 M parameters and no external pre-training, outperforming all baselines across all key tumor metrics. Specifically, it surpasses the next-best model by 0.45% in tumor Dice, 6.0 points in F1 score, 6.6 points in sensitivity, and 3.4 points in precision, simultaneously reflecting its ability to suppress both missed tumors and false alarms in clinically realistic conditions. Our code is available at: https://github.com/abdora-ai/TRIUNE-Net
Read the full abstract

All authors Amir Hossein Saleknia, Alireza Kheyrkhah, Sanaz Karimijafarbigloo, Reza Azad, Sina Houshmand, Ulas Bagci, Dorit Merhof, Alaa Sulaiman

Software Engineering

Agent ATO: Visualizing Agent Interaction Timelines from Logs

Takuto Kawamoto, Yoshiki Higo and Raula Gaikovina Kula

AI coding agents are becoming part of developers' workflows, but their behavior is difficult to understand from final code changes alone. During a task, agents interact with software repositories through sequences of actions such as searching for files, reading code, editing programs, and running tests or build commands. These interactions, together with token usage, are often recorded in console logs, but raw logs are difficult for developers to inspect. In this paper, we propose Agent ATO (Agentic Trajectory Observer), a tool for visualizing AI coding agent interaction timelines from console logs. Agent ATO extracts agent interactions, classifies them by command or tool type, and visualizes them as timelines. In addition to an all-interaction timeline, Agent ATO provides filtered timelines that emphasize file discovery, file reading, file editing, and execution while preserving surrounding context. We illustrate how Agent ATO may help developers inspect and compare agent actions using selected runs from two repair tasks. Future work will apply Agent ATO to more agents, tasks, and development environments, and will evaluate whether it reduces the effort needed to compare trajectories.
Read the full abstract

Computer Vision

Human-Centric Image Captioning with Subject-Centered Spatial Understanding

Bozhou Li, Jiahang Zhang, Yue Ding et al. (+11 more)

While multimodal large language models (MLLMs) achieve remarkable performance on generic image captioning, they frequently suffer from structural hallucinations in human-centric scenarios. Accurately modeling human subjects is foundational for critical downstream applications, such as accurate avatar/video/image generation and fine-grained human action understanding. However, these tasks require highly precise subject-centered spatial grounding, such as distinguishing egocentric left/right laterality and maintaining correct anatomical-object bindings. Although catastrophic for structural integrity, these localized spatial inversions are often overshadowed by overall descriptive metrics in existing benchmarks. To systematically expose and quantify this bottleneck, we introduce SPACE (Subject-centric Poses, Appearance, and Characteristics Evaluation), a benchmark designed to evaluate subject-centered spatial understanding. On SPACE, we reveal that despite strong generic perception, current MLLMs consistently fail to ground descriptions in the subject's intrinsic frame of reference. To bridge this gap, we propose a specialized data construction and alignment pipeline. We first extract structured spatial hints from fine-grained body-part localization to guide a two-stage caption rewriting process, yielding highly spatially-faithful training data. Furthermore, we design a rubric-based reward for Group Relative Policy Optimization (GRPO) that explicitly penalizes structurally critical spatial errors during alignment. Extensive experiments on SPACE demonstrate our framework significantly improves human-centric caption quality, particularly in subject-centered spatial reasoning, achieving performance competitive with strong closed-source models. Our benchmark and code are available at https://github.com/JHang2020/SPACE-Eval.
Read the full abstract

All authors Bozhou Li, Jiahang Zhang, Yue Ding, Yushuo Guan, Bohan Zeng, Yiyan Ji, Xinlong Chen, Yang Shi, Yifan Dai, Yuran Wang, Chengzhuo Tong, Pengfei Wan, Yuanxing Zhang, Wentao Zhang

Information Theory

Convolutional Codes from Cyclic Codes with Guaranteed Free and Local Minimum Distances

Khaled Abdel-Ghaffar, Daniel J. Costello Jr., Juane Li et al. (+1 more)

This paper presents an algebraic method to construct convolutional codes with guaranteed free and local minimum distances without limit based on cyclic codes of odd lengths. The constructions are simple but effective, and no computer search is needed. For any two positive integers $r$ and $t$ with $1 \leq r < t$, a rate-$r/t$ convolutional code $\mathcal{C}_{\text{convol}}$ can be constructed by using a chain of $r$ cyclic codes $\mathcal{C}_0, \mathcal{C}_1, \ldots, \mathcal{C}_{r-1}$ of the same length $n$ which satisfy the inclusion condition, $\mathcal{C}_0 \supset \mathcal{C}_1 \supset \ldots \supset \mathcal{C}_{r-1}$. Such a convolutional code $\mathcal{C}_{\text{convol}}$ is composed of a semi-infinite chain of identical local codes confined in a diagonal band of width $n$. Each local code $\mathcal{C}_{\text{local}}$ of $\mathcal{C}_{\text{convol}}$ is formed from the $r$ cyclic codes in the code chain and is a specially localized subcode of the mother code $\mathcal{C}_0$ in the code chain. The minimum distance $d_{\text{local}}$ of each local code of $\mathcal{C}_{\text{convol}}$ is lower bounded by the minimum distance $d_0$ of the mother code $\mathcal{C}_0$ in the code chain. The local structure of $\mathcal{C}_{\text{convol}}$ allows it to be decoded based on a designed parity-check matrix of the mother code $\mathcal{C}_0$ using a sliding window decoding scheme.
Read the full abstract

All authors Khaled Abdel-Ghaffar, Daniel J. Costello Jr., Juane Li, Shu Lin

Robotics

EvoNav-Bench: Benchmarking Lifelong Navigation in Evolving Environments

Xilin Wang, Guoxi Zhang, Hongming Xu et al. (+3 more)

Lifelong navigation (LN) requires an embodied agent to solve a sequence of navigation subtasks in the same environment. Since solving each subtask from scratch incurs redundant exploration, an LN agent must consolidate experience from earlier stages and reuse it in later stages, often through persistent scene representations such as scene graphs or visual snapshots. However, existing approaches typically assume a stationary environment, whereas in real-world LN settings, human activities can cause the environment to evolve. With the stationary assumption violated, existing methods may fuse outdated prior observations with new observations, yet current benchmarks cannot reveal this failure mode. In this paper, we present EvoNav-Bench, which extends the GOAT-Bench style LN formulation in the context of evolving environments. Built on the ProcTHOR framework, EvoNav-Bench introduces environment modifications between navigation tasks, making prior experience useful but not fully reliable. This design enables controlled evaluation of how environment evolution affects LN agents that reuse prior scene observations. Using EvoNav-Bench, we benchmark three recent methods that build and reuse scene representations for navigation. We also compare three simple heuristic strategies for handling environment evolution: Frontier-Update, Fail-then-Update, and Stage-Reset. Our results show that existing methods are brittle under environment evolution, while the heuristic strategies enable a controlled analysis of how agents can adapt to scene changes and mitigate their impact.
Read the full abstract

All authors Xilin Wang, Guoxi Zhang, Hongming Xu, Zhuofan Zhang, Tianxu Wang, Lifeng Fan