Robotics
Size Doesn't Matter: Material-State Reinforcement Learning for Excavator Transferable Soil Manipulation
Lennart Werner, Pol Eyschen, Sean Costello et al. (+3 more)
Earthmoving tasks such as excavation, backfilling, or embankment construction require deliberate repositioning of deformable soil. For these tasks, human operators use all shovel faces, while autonomous systems so far are limited to excavation and dumping. Current methods often rely on heuristic models but do not incorporate soil mechanics. We address this shortcoming by using Reinforcement Learning in a GPU-parallelized Material Point Method particle simulation. Our controllers are conditioned on material state such as shape and compactness, enabling skills that use multiple contact faces of the tool and displace material both inside and outside of the shovel. To use the same learned weights across machines, our policies operate in a normalized end-effector space and are deployed through a calibrated machine interface. We evaluate this calibrated transfer on an 11.5t hydraulic excavator and a 500g tabletop
robot. We validate performance through autonomous construction of a 42m long, 2.1m high embankment in 45min, executing 201 individual policy strokes without failure, retry, or operator intervention. In a direct comparison, the autonomous controller matches an expert operator's progression speed and produces a higher, more consistent embankment. Additional qualitative backfilling and compaction experiments demonstrate the material-state awareness and calibrated transfer across machines.
Read the full abstract ▾
All authors Lennart Werner, Pol Eyschen, Sean Costello, Pierluigi Micarelli, Andrei Cramariuc, Marco Hutter
Numerical Analysis
An Optimal IPDG Scheme for the Biharmonic Equation
Bohua Zhang, Xia Ji and Shuo Zhang
This paper presents an optimal interior penalty discontinuous Galerkin (IPDG) scheme for the planar biharmonic equation using piecewise
polynomials of degree $k=3$ or $4$. In standard IPDG methods, large penalty parameters force the discrete solution into an overconstrained space, severely degrading accuracy—phenomenon known as numerical locking. To overcome this, our method enforces only vertex continuity and projects the jumps of the function and its normal derivative onto $\mathcal{P}^{k-3}$ and $\mathcal{P}^{k-2}$, respectively. Consequently, as the penalty parameters tend to infinity, the discrete solution is forced to lie in a constrained subspace $V_{h,\infty}^k$, which we identify as the optimal nonconforming finite element space $B_h^k$. This intrinsic connection fundamentally eliminates numerical locking. We prove optimal error estimates of $\mathcal{O}(h^{k-1})$ in a mesh-dependent energy norm and $\mathcal{O}(h^{k+1})$ in the $L^2$ norm. Numerical experiments on both convex and L-shaped domains confirm that the proposed scheme is robust and entirely locking-free, even for extremely large penalty parameters.
Read the full abstract ▾
Computation and Language
Doc2FRC: Length-Consistent Document-Level Machine Translation via Fixed-Range Chunking
Xiaotian Wang, Youyuan Lin, Zhan Shen et al. (+1 more)
Advanced large
language models (LLMs) with long context windows can substantially reduce input truncation in document-level machine translation (DocMT). However, direct Doc2Doc translation remains prone to n-gram repetition and progressive quality degradation. A common remedy is to segment the document into finer-grained chunks. Nonetheless, conventional rule-based chunking approaches fail to handle the length distribution mismatch between training and inference. To address this, we introduce Fixed-Range Chunking (FRC), utilizing dynamic programming to partition documents into chunks within a predefined length interval. By consistently applying FRC during training and inference, the input documents of any length are mapped to the same length distribution, substantially reducing train-test length mismatch. Centered on FRC, we propose a lightweight dual-boundary matching
algorithm for chunk
alignment, alongside four distinct training strategies. Experimental results show that FRC-based fine-tuning substantially improves 7B LLMs over direct Doc2Doc fine-tuning and outperforms existing DocMT methods on IWSLT2017. We further construct GlobVDoc, a 10-
language test set independent of mainstream DocMT training sources, and show that FRC improves out-of-distribution document translation.
Read the full abstract ▾
All authors Xiaotian Wang, Youyuan Lin, Zhan Shen, Hitomi Yanaka
Computational Complexity
Deterministic NC Quadratic Root Counting in Characteristic Two
Sanyam Agarwal and Gorav Jindal
Counting satisfying assignments of Boolean formulas is a basic problem in theoretical computer science, with $\#3\text{-}\SAT$ as the standard $\SharpP$-complete problem. More generally, counting the solutions of a system of
polynomial equations over $\F_2$ is $\SharpP$-complete. Here we focus on the more structured problem of counting the solutions of a single
polynomial equation. For
polynomial equations over finite fields, Ehrenfeucht and Karpinski \cite{computationalcomplexityofxorandcountingproblems1990} showed a sharp difference between degrees two and three: quadratic root counting is solvable in
polynomial time, while the degree-three problem is $\SharpP$-complete. Their quadratic
algorithm is sequential. For fixed finite fields, Ishai et al.~\cite{ishai2012randomizing} later gave deterministic parallel
algorithms in odd characteristic and randomized parallel
algorithms in characteristic two. We give a deterministic $\NC$
algorithm for exactly counting the solutions of a quadratic
polynomial equation over every fixed finite field of characteristic two. Our
algorithm separates the radical and uses the absolute trace to realize the bit distinguishing the two nondegenerate finite-field types as the Arf invariant \cite{arf1941untersuchungen} of a quadratic form over $\F_2$. It then recovers that invariant from an integral
matrix using Browder's determinant criterion \cite{browder2006complete}. This replaces the randomized canonical-form step in the
algorithm of Ishai et al.
Read the full abstract ▾
Computer Vision
Beyond Ambiguous Visual Cues: Studying Physiological Disruptions and Cross-Modal Inconsistencies in Deepfake Videos
Chenxi Yang, Yassine Ouzar and Larbi Boubchir
Recent deepfake detection studies increasingly suggest remote photoplethysmography (rPPG) signals as an authenticity cue. However, existing
benchmarks lack physiological ground truth, and current detectors underexplore the cross-level relationship between facial features and physiological dynamics, often relying on late fusion or rPPG features alone. In this paper, we construct high-fidelity deepfake manipulations on established real rPPG
datasets (COHFACE and UBFC-rPPG) to investigate how forgeries disrupt natural physiological signals and facial behavior at the same time. Building on this analysis, we propose a bidirectional co-
attention fusion detector that jointly models rPPG and facial behavior
tokens. This mechanism explicitly captures the cross-level dependencies between pulse dynamics and facial motion to learn a robust, joint authenticity representation. Extensive experiments using a subject-disjoint 5-fold evaluation demonstrate the superiority of our approach. Achieving a 92.80% AUC on constructed
datasets using face swapping and 96.78% AUC on motion transfer, our model outperforms both the rPPG-only single modality baseline and the best feature-level fusion methods. Furthermore, transfer-learning result of the fusion detector on Celeb-DF-v2 while keeping both feature extractors fixed achieves 91.20% accuracy and 86.08% AUC, which suggests applicability under target-domain adaptation.
Read the full abstract ▾
Information Theory · Cross-listed from quant-ph
Log-Sobolev inequality, von Neumann entropy and Entanglement of Formation
A.S.Holevo and M.E.Shirokov
We present two results derived from the sharp log-Sobolev inequality for the uniform measure on a complete graph which concern the von Neumann
entropy and the
Entanglement of Formation of a state of finite and infinite-dimensional
quantum systems. The first result is a sharp Lipschitz lower semicontinuity bound for the von Neumann
entropy at any mixed state $\rho$ with uniform positive spectrum (i.e. a state proportional to a projector) w.r.t. the fidelity deficit: the inequality $\,S(\rho)-S(\sigma)\leq C_\rho(1-F(\rho,\sigma))\,$ valid for any state $\sigma$, where $C_{\rho}$ is a constant depending on the rank of $\rho$. The second result is a sharp Lipschitz lower semicontinuity bound for the
Entanglement of Formation at any pure state $\rho$ with uniform positive spectrum of marginal states w.r.t. the fidelity deficit: the inequality $\,E_F(\rho)-E_F(\sigma)\leq C_\rho(1-\mathrm{Tr}\rho\sigma)\,$ valid for any state $\sigma$, where $C_{\rho}$ is a constant depending on the Schmidt rank of $\rho$. In both cases the optimal constant $C_\rho$ is equal to the optimal constant $K_{d}$ in the log-Sobolev inequality for the complete graph with $d$ vertices: in the first case $d=\mathrm{rank}\rho$, in the second one $d=\mathrm{rank}\rho_A=\mathrm{rank}\rho_B$. The authors are grateful to GPT 5.6 for valuable discussion and technical help in preparing this note.
Read the full abstract ▾
Sound
What Did the MLLM Hear? Token-Level Spectro-Temporal Grounding for Audio MLLM Explainability
Lucia Cascone, Valeria Fraenza, Michele Nappi et al. (+2 more)
Audio-based
Multimodal Large
Language Models (MLLMs) can generate detailed natural-
language descriptions of complex acoustic scenes, yet it remains unclear which parts of the input audio support each generated
token. This is particularly challenging because acoustic evidence is distributed across time and frequency, and concurrent sound events may overlap temporally while occupying different spectral regions. We introduce STAG, to our knowledge the first post-hoc framework for
token-level spectro-temporal grounding of captions generated by audio-based MLLMs. STAG estimates the temporal support for each generated
token using target-
token-specific vocabulary projections of the encoded audio representations, measures frequency-band relevance through controlled spectral occlusion, and combines the two signals into a spectro-temporal relevance map. We evaluate STAG against ten post-hoc explanation methods across four grounding
benchmarks, where it achieves the best event-localization performance on every
dataset, and apply it to eight audio-
language backbones without parameter updates. Counterfactual deletion further shows that removing the identified evidence selectively reduces confidence in the corresponding event and frequently removes it from the regenerated caption. These results provide behavioral support for the faithfulness and selectivity of the explanations.
Read the full abstract ▾
All authors Lucia Cascone, Valeria Fraenza, Michele Nappi, Fabio Narducci, Benedetto Simone
Robotics
Driving Context-guided Model Predictive Planning and Control for Autonomous Car Racing at the Limit and Beyond
Ayoub Raji, Federico Sacco, Nicola Musiu et al. (+1 more)
This paper presents a Model Predictive Control-based motion planning and control pipeline for autonomous car racing capable of adapting to different driving contexts, such as overtaking, nominal driving, and countersteering. A Cost Blending state machine manages the identification of different driving contexts and the selection of their predefined weights to be applied to the Model Predictive Planning (MPP) and Control (MPC) modules. The two optimization-based solutions share the same problem formulation and model, differing only in horizon length, rate, tuning, and in their open-loop versus closed-loop approach to maximize the effectiveness of their interaction. The work is validated on the fully autonomous open-wheel racecar Superformula EAV-25, with a lap time achieved that is within 2% of the best human driver reference. The results demonstrate the capability of the solution in driving at the limit of handling, smoothly executing overtaking maneuvers, and quickly reacting to high oversteering conditions to recover the vehicle stability.
Read the full abstract ▾
All authors Ayoub Raji, Federico Sacco, Nicola Musiu, Marko Bertogna
Machine Learning
LiveProBench: Can Streaming Video Models Really Interact Like Humans?
Kaixuan Du, Xin Wan, Hang Zhang et al. (+4 more)
Streaming
video understanding requires models to process continuous
multimodal input while maintaining temporal context. Existing evaluations are predominantly reactive: they query a model at a selected timestamp and therefore do not assess when it should respond. Proactive interaction instead requires monitoring a standing request, responding within an appropriate interval after the target event, and otherwise remaining silent. We introduce LiveProBench, which evaluates models at one-second stream intervals without an explicit response cue. Its six subtasks vary trigger ambiguity and timing tolerance. Event Sensitivity geometrically combines response and silence rates on the same recording; four window-based subtasks distinguish early, in-window, and missed responses; and Duplicate Counting penalizes omissions and repetitions. Premature responses outnumber missed responses for half of the evaluated models, revealing a substantial gap in the temporal decision-making required for human-like interaction.
Read the full abstract ▾
All authors Kaixuan Du, Xin Wan, Hang Zhang, Meng Cao, Dai Guan, Ming Chen, YuKun Wang
Software Engineering
Separating Engineering Reasoning from DEXPI Serialization in LLM-Based Greenfield Surface-Process Design: A Three-Case Study for Underground Gas Storage
Qingchuan Zhu, Shuyue Tong and Pengju Ren
Large
language models can produce engineering descriptions and structured process representations, but standards-level serialization can substantially increase the generation burden. This diagnostic study examines whether separating engineering
reasoning from Data Exchange in the Process Industry (DEXPI) serialization changes where representation and engineering failures occur in constrained greenfield surface-process design for underground gas storage. We compare Direct DEXPI generation with generation of a lightweight Engineering Intermediate Representation (IR) on three cases: single-pressure injection, withdrawal and export, and dual-pressure injection. All six conditions use one fixed model snapshot, qwen3.8-max-0902, with one completed hosted generation per condition. Direct outputs are XSD-valid in 2 of 3 cases, while all 3 Engineering IR outputs are structurally valid under a minimal validator. Direct
prompt inputs contain approximately 121.8k-121.9k
tokens, compared with 617-699
tokens for the Engineering IR
prompts. Engineering feasibility does not uniformly favor the IR: one IR output is rejected for an explicit cooling-state contradiction. The cases also reveal two distinct Direct DEXPI failure modes: engineering inconsistency and standards-level serialization failure. The observed comparison shows that, in these evaluated method bundles, deferring DEXPI serialization substantially reduces representation burden and helps isolate serialization failure, but reducing representation burden alone does not eliminate engineering inconsistencies.
Read the full abstract ▾
Computation and Language
LifeMem: Enabling Lifelong Experience Reuse for LLM Agents
Yuli Qiu, Yutong Li, Wei Su et al. (+5 more)
Large
language model
agents are expected to continuously adapt to new tasks and environments over their lifetime by reusing past experience. However, existing
memory-based
agents struggle to transfer reusable experience across environments and suffer from catastrophic forgetting as experience accumulated. To address these challenges, we propose LifeMem, a lifelong learning framework that enables
agents to transfer knowledge across multiple environments. During learning, LifeMem clusters accumulated interaction trajectories based on underlying workflows to extract reusable skills. When solving a new task at inference time, the
agent recalls relevant skills and trajectories to guide actions. To validate our method, we conduct experiments across 10 environments and over 13k tasks with 2k newly annotated interaction trajectories. Results show that LifeMem enables effective experience reuse in lifelong learning, achieving both reduced forgetting on learned tasks and superior cross-task transfer. Further analysis reveals that task streaming impacts learning, while consolidating structurally similar trajectories within
memory boosts performance.
Read the full abstract ▾
All authors Yuli Qiu, Yutong Li, Wei Su, Zeming Liu, Wanxiang Che, Heyan Huang, Haifeng Wang, Yuang Guo
Computation and Language
SWARM: A Multilingual Human-Annotated Dataset for Russian Propaganda Detection in Search Engine Results
Manuel Tonneau, Abhinav Dubey, Farhan Shaikh et al. (+8 more)
Russian state propaganda spreads across many
languages and online spaces. Yet, most computational work examines only one such space, usually social media, in one or two
languages, and analyses sources rather than content. We introduce SWARM (Search-Web documents Annotated for Russian propaganda,
Multilingual), a
dataset of 2,183 search engine results across nine
languages and diverse web domains (e.g., news, blogs, government sites), each annotated by trained coders for whether it supports a recurring Russian propaganda narrative. We
benchmark a source-based blocklist, supervised classifiers, and zero-shot LLMs against these labels. The blocklist misses most propaganda-supporting documents, because such content is not confined to flagged "propaganda" outlets but also appears on mainstream ones. Content-level analysis helps, though how much depends on the model: the strongest LLM reaches a positive-class F1 of 0.73, whereas the supervised classifiers reach only about 0.5, with the smaller LLMs over-predicting support, mistaking topical relevance for endorsement. Detecting search-borne propaganda thus requires per-
language, content-level evaluation, which we hope SWARM and our evaluation code enable.
Read the full abstract ▾
All authors Manuel Tonneau, Abhinav Dubey, Farhan Shaikh, Ilaria Vitulano, Martha Stolze, Hale Dedeoglu, Clara Riechert, Ella Kuka, Maryna Sydorova, Mykola Makhortykh, Elizaveta Kuznetsova
Machine Learning
Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner
Kazusato Oko, Annie Ulichney, Nika Haghtalab et al. (+1 more)
While Reinforcement Learning from Human Feedback (RLHF) is the standard paradigm for aligning large
language models with human preferences, its effectiveness in pluralistic settings has been called into question. Notably, recent work by Gölz et al. (2025) demonstrated that the distortion – defined as the multiplicative gap between the average user utility of the RLHF policy and the optimal average utility – can scale exponentially with the Bradley-Terry temperature parameter $\beta$ when users have heterogeneous preferences. In this work, we present a fine-grained analysis of the distortion of RLHF with reward clipping and demonstrate that such exponential degradation is not a fundamental property of the
algorithm but rather a consequence of distribution mismatch between the distribution generating preference data ($\mu$) and the KL reference policy ($\pi_{\mathrm{ref}}$). To this end, we establish tight upper and lower bounds on the distortion of RLHF across multiple regimes of the KL regularization strength. We show that in a representative regime, under the Bradley-Terry model, the distortion is $\tilde{\Theta}(\beta B + \beta)$, where $B$ is an upper bound on the log density ratio between $\mu$ and $\pi_{\mathrm{ref}}$. In particular, when there is no distribution mismatch (i.e., $\mu = \pi_{\mathrm{ref}}$), RLHF achieves the optimal distortion of $O(\beta)$ up to a constant. Our results suggest that, to reasonably maximize average utility with RLHF, it is preferable to use on-policy sampled preference data or to fine-tune before RLHF on data from a source close to $\mu$.
Read the full abstract ▾
All authors Kazusato Oko, Annie Ulichney, Nika Haghtalab, Han Bao
Numerical Analysis
Second-Order IMEX Time-Stepping Methods for Efficient Bidomain Multi-Electrode Array Simulations of Cardiac Stem Cell Monolayers
Sofia Tonali, Sofia Botti and Luca Franco Pavarino
Multi-electrode Arrays (MEAs) enable tissue-level electrophysiological studies of human induced pluripotent stem
cell-derived cardiomyocytes (hiPSC-CMs) by recording extracellular field potentials. Recent computational models have improved MEA simulations by coupling detailed electrode descriptions with the Bidomain framework, a
parabolic-
elliptic system of nonlinear PDEs coupled with a stiff ionic model. The standard numerical strategy relies on operator splitting techniques that decouple the PDE and ODE components. In most implementations, the ionic subsystem is treated explicitly while the diffusive operator is handled implicitly, resulting in a first-order implicit-explicit (IMEX) time discretization. Although computationally convenient, this approach limits temporal accuracy and may reduce efficiency in large-scale simulations. In this work, we investigate higher-order IMEX Runge-Kutta schemes within a Strang-based operator splitting, specifically tailored for the MEA model of hiPSC-CMs monolayers. First- and second-order schemes are compared in terms of computational cost and global error against a high-fidelity reference solution obtained with a very small time step. For a fixed computational cost, the second-order schemes achieve errors that are 2-3 orders of magnitude smaller than those of the first-order method. These results demonstrate that higher-order IMEX integration significantly improves the accuracy-to-cost ratio of MEA simulations, providing a practical and reliable approach for large-scale electrophysiological studies.
Read the full abstract ▾
Networking
Hybrid Monitoring for Early Fault Detection in Cloud-Native 5G Systems
Anton Andersson, Sai Akshara Naineni, Mats Jansborg et al. (+2 more)
This paper presents the design implementation and evaluation of NetMon a hybrid network monitoring system designed for Kubernetes-based 5G packet core deployments specifically evaluated on Ericssons Access and Mobility Management Function AMF clusters NetMon combines eBPF-based passive kernel-level traffic observation with active TCP probing and centralized correlation to detect and localize network degradation within seconds The evaluation results demonstrate that the system detects faults as subtle as 10ms of added latency or 5 packet loss correctly attributes them to the affected infrastructure component and maintains this capability under application loads up to 50 simulated UE load The total resource overhead of 34 millicores CPU and 45 MiB
memory per pod suggests that the approach is promising for further validation without impacting the monitored workload The hybrid approach addresses a gap in existing monitoring tools standard health checks cannot detect partial degradation scrape-based systems introduce detection delays measured in tens of seconds and purely passive tools cannot verify idle network paths By combining these complementary techniques and centralizing the analysis the system provides the early detection and fault localization capabilities required for maintaining service quality in cloud-native 5G infrastructure.
Read the full abstract ▾
All authors Anton Andersson, Sai Akshara Naineni, Mats Jansborg, Yixing Zhang, Romaric Duvignau
Discrete Mathematics · Cross-listed from math.PR
Local Search for Almost-Spanning Square Grids in Erdős–Rényi Random Graphs
Dávid Ferenczi and Alexander Grigoriev
Finding large
lattice subgraphs in sparse Erdős–Rényi random graphs is a classical problem at the interface of random graph theory and
algorithms. General bounded-degree
embedding and universality
theorems give powerful results for broad graph families, but when specialized to square grids they operate at densities substantially larger than the grid-emergence scale. In this paper we exploit the specific geometry of the square grid. We introduce the Quarantined Local Search, a three-phase local
algorithm that separates the construction of an initial boundary from the later corner-closure process and controls adaptive negative exposure through bounded pair-test histories. We prove that, for every fixed $\delta\in (0,1)$, there exists $C_\delta>0$ such that the
algorithm embeds a $k\times k$ square grid with $k^2\le (1-\delta)n$ in $G(n,p)$ with high probability whenever $p\ge C_\delta\sqrt{\ln k/n}$. Thus, for $k^2=\Theta(n)$, a density of order $\sqrt{\log n/n}$ is sufficient, a factor of order $\sqrt{\log n}$ above the corresponding $n^{-1/2}$ emergence scale.
Read the full abstract ▾
Cryptography and Security
Poster: Towards Selecting Threat Appropriate Industrial Intrusion Detection Systems
Stefan Lenz, Johannes Weidmann and Martin Henze
As the threat landscape against industrial control systems is dynamic, effective security requires detection strategies capable of timely reactions to these evolving threat situations. To address this problem, we propose the idea of a counter-threat intelligence sharing based mechanism to select appropriate detectors for current circumstances. To highlight the potential of this mechanism, we conduct attack-level performance evaluations of various intrusion detection systems. Results show the variance of intrusion detection performance depending on the attack scenario, emphasizing the benefits of such a mechanism for industrial control system security
Read the full abstract ▾
Human-Computer Interaction
Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments
Jun Fang, Jiajin Li, Yuntao Wang et al. (+5 more)
Positive everyday moments often escape deliberate recording, while continuous self-tracking can generate extensive records that are difficult to revisit. We explore laughter as a naturally occurring, sparse index for constructing contextualized personal records to support later reconstruction and reflection. A formative study with 12 participants characterized laughter as an affective but semantically incomplete index and informed LaughAnchor, a mobile and wearable self-tracking system. During participant-initiated recording, the system assembles detected laughter and aligned context into candidate moments for later reconstruction and reflection, with layered context disclosure, user-controlled curation, and near-term and long-term resurfacing. In a three-week field deployment with 12 participants, passive indexing preserved moments they considered unlikely to record deliberately but valued retrospectively. During resurfacing, participants attributed affective re-experiencing to laughter and used additional context both to reconstruct episodes and to explore already-recalled experiences. Across moments and reviews, resurfacing supported rediscovery and broader awareness of relationships, routines, and emotional states. These findings inform self-tracking designs that use sparse affective indices to organize contextual records for reconstruction and reflection, while keeping interpretation and retention under user control.
Read the full abstract ▾
All authors Jun Fang, Jiajin Li, Yuntao Wang, Kexin Miao, Susu Wang, Xiaoyu Xie, Kaixin Ji, Yuanchun Shi
Robotics
Breaking the Vision-Action Shortcut: Latent Interface Training for Generalizable Robotics Foundation Models
Jianman Lin, Shailesh Shailesh, Zhongyi Luo et al. (+1 more)
Robot foundation models achieve strong in-distribution performance but often degrade under visual distribution shifts. When learning to generate actions from pretrained visual representations, models may exploit task-irrelevant visual cues that correlate with demonstrated actions within the training distribution. Such
vision-action shortcuts can undermine generalization when these correlations change under distribution shifts. Mitigating these shortcuts requires constraining how visual information is used for action generation while preserving task-relevant spatial information. We propose Latent Interface Training (LIT), a framework-agnostic two-stage strategy that first establishes a spatial-goal-conditioned action prior without images, then constrains visual conditioning through a pose-supervised latent interface. Stage 1 trains the action expert to generate action chunks conditioned on
language,
robot state, and each demonstrated chunk's terminal SE(3) end-effector pose, learning goal-directed action generation independently of visual cues. Stage 2 introduces a latent interface that aggregates visual and semantic representations and serves as the pretrained action expert's only visual conditioning pathway. The interface is supervised to reconstruct the terminal pose previously used to condition Stage 1, encouraging it to retain the goal-relevant spatial information needed for action generation. Across four
vision-
language-action and world-action architectures (Pi0.5, MolmoAct2, FAST-WAM, and ImageWAM), LIT improves overall LIBERO-Plus success by 3.87-10.70 percentage points while preserving or improving average LIBERO success. Real-world evaluations show 13.30-16.70 percentage-point gains in success aggregated across three tasks under unseen camera configurations, lighting variations, and distractors.
Read the full abstract ▾
All authors Jianman Lin, Shailesh Shailesh, Zhongyi Luo, Jiafei Duan
Machine Learning
Explaining Time Series Forecasting with Horizon-Resolved Attribution
Seunghan Lee, Jun Seo, Jaehoon Lee et al. (+9 more)
Recent advances in explaining time series (TS) models have produced methods that identify which past values a prediction depends on. However, most existing methods return a single importance vector, assuming that every predicted step depends on the same past values. In this paper, we show that this assumption does not hold, as different forecast steps depend on different past values. Motivated by this observation, we propose Horizon-Resolved eXplanation (HRX), which adds a horizon axis to the explanation, so that every forecast step receives its own importance map. HRX is a simple yet effective plug-in framework with three components: 1) an
estimator that reads these maps out of any differentiable forecaster without modifying the TS backbone, 2) an evaluation protocol that validates the horizon axis by measuring how much a single forecast step changes when the inputs an importance map ranks highest are removed, and 3) a rank criterion that predicts in advance whether the axis is worth resolving on a given TS. We further show that this step-wise dependence is low-dimensional, as the explanations of all steps are built from a few shared maps whose number does not grow with the forecast length. Extensive experiments across various backbones and
datasets show that the improvement comes from the horizon axis and holds for
estimators of previous explanation methods. Code is available at https://github.com/seunghan96/HRX.
Read the full abstract ▾
All authors Seunghan Lee, Jun Seo, Jaehoon Lee, Junhyeok Kang, Sangjun Han, Sungdong Yoo, Minjae Kim, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Soonyoung Lee, Wonbin Ahn
Robotics
Online Material Estimation for Conditioned Diffusion Policy in Shaping Deformable Linear Objects
Ryunosuke Yamada, Tomohiro Motoda, Yukiyasu Domae et al. (+1 more)
Shape control of deformable linear objects (DLOs) is challenging for imitation learning because deformation behavior varies with material properties such as stiffness and elasticity, so a single policy must generate different action sequences for different objects even when the goal shape is identical. We propose a
diffusion policy conditioned on material labels that are estimated online during manipulation. A recurrent estimation network predicts the material label of the grasped object from the time series of multi-view images and
robot joint states, and the predicted label conditions the
diffusion policy at every inference step. We collected 480 real-
robot demonstrations covering four DLO materials and three groove-placement tasks, and compared per-material specialist policies, a task-conditioned policy without material labels, a policy conditioned on ground-truth material labels, and the proposed policy. Conditioning on ground-truth material labels improved the average success rate from 45.8% to 60.0% over the task-only policy, and the proposed policy reached 60.8% without any prior material information, matching the policy given ground-truth labels. A post-hoc analysis shows that the
estimator extracts material-related information from the manipulation observations and that the
diffusion policy responds to the resulting conditioning signal, while the one pronounced failure case is associated with persistent confusion between two similar materials.
Read the full abstract ▾
All authors Ryunosuke Yamada, Tomohiro Motoda, Yukiyasu Domae, Tokuo Tsuji
Information Theory
Confusion-Erasure Bounds of Error-Bounded Decoders under QAM
Wenwen Chen, Bin Han and Hans D. Schotten
6G is expected to push ultra-reliable low-latency communication (URLLC) toward stringent residual-error tar- gets for mission-critical services, where undetected errors and erasures carry fundamentally different costs. Block error rate (BLER) conflates block confusions (undetected errors) and block erasures, which have fundamentally different impacts on system reliability. To characterize the confusion rate analyt- ically, we consider an error-bounded decoder with spherical decision regions. We focus on square quadrature amplitude modulation (QAM) constellations in the finite blocklength (FBL) regime. The confusion rate is governed by the Euclidean distances between modulated codewords, whereas the code structure constrains their Hamming distances. To bridge the two metrics, we introduce a typical distance, which represents the Euclidean distance of a codeword pair at a given Hamming distance on average, and relate it to the Hamming distance through a per-symbol squared-distance coefficient. On this basis, we derive a lower bound on the block confusion rate for random codes and an upper bound for codebooks whose minimum Euclidean distance reaches the typical distance at the non-overlapping threshold. We further characterize how both bounds evolve with the average symbol energy and the block- length. Numerical results, validated by an exact computation for short codes, confirm that as the signal-to-noise ratio (SNR) or redundancy increases, the confusion rate falls many orders of magnitude below the reliability target, leaving detectable erasures as the dominant residual error.
Read the full abstract ▾
Machine Learning
Geometric-to-Semantic Spherical Transfer Learning for Cortical Sulci Labeling
Saeb Tounsi, Joël Chavas, Pietro Gori et al. (+3 more)
Deep learning on cortical surfaces faces a dilemma: capturing the complex
topology of over 60 nomenclature-dependent sulci per hemisphere requires high-capacity models, yet the extreme scarcity of expert annotations ($N=62$ subjects) inevitably causes overfitting. Standard supervised approaches fail to generalize in this data-scarce regime, particularly for variable and small sulci where topological ambiguity is high. To overcome this limitation, we introduce a
Geometric-to-Semantic Spherical Transfer Learning framework. First, we leverage massive unlabeled data (UK Biobank, $\approx$30,000 subjects) to pre-train a spherical encoder using a locally-optimized strategy. By relying solely on continuous surface features (
curvature and depth), the relevance of this pre-training is confirmed by the model's ability to detect localized and rare topological traits, such as sulcal interruptions. The downstream labeling task, however, introduces extracted sulcal fundi (lines) as an explicit semantic input. To bridge this dimensional domain gap (from purely
geometric to semantic) without causing catastrophic forgetting, these anatomical lines are integrated into the pre-trained backbone via a soft-initialized Topological Prior Injector. Our experiments demonstrate that this approach outperforms fully supervised baselines trained from scratch, achieving a mean Dice of 0.77. Crucially, a local analysis reveals that the self-supervised
geometric priors yield the largest performance gains on variable and tertiary sulci (up to 14.8%), confirming that learning the cortex shape is highly beneficial for identifying its rarest parts.
Read the full abstract ▾
All authors Saeb Tounsi, Joël Chavas, Pietro Gori, Vincent Frouin, Denis Rivière, Jean-François Mangin
Cryptography and Security
Subgroup Packing for Batched PASTA Transciphering
Mugurel Barcau, Vicenţiu Paşol and George C. Ţurcaş
With transciphering, a server converts symmetrically encrypted records into homomorphic ciphertexts without learning the records or the symmetric key. For the PASTA cipher, this conversion involves dense linear maps whose implementation depends on how record words are arranged in the ciphertext. We ask whether rearranging a fixed batch can reduce its conversion cost. Instead of storing each record's words in a contiguous block, our layout interleaves records so that cyclic word shifts preserve each record's positions, which form a coset of a cyclic subgroup. For direct evaluation as a sum of masked translations, we characterize the required displacements and relate their counts, 255 for the contiguous layout and 128 for the subgroup layout, to a prior transversal-difference invariant. We implement three equally batched schedules for complete PASTA-3 conversion and subsequent public subset-sum queries in HElib. Across twelve paired corpora under six homomorphic keys, all 24 direct and subgroup conversions and all 48 subsequent queries return the expected values and required zeroes. The median paired ratio of direct to subgroup server cost is 1.60, including fresh public generation, conversion and two queries. The reduction comes with less remaining noise capacity. These results establish a packing-dependent cost–noise tradeoff.
Read the full abstract ▾
Artificial Intelligence
SteerDuplex: Steerable Duplex Speech Dialogue Models
Utkarsh Tyagi, Ramaneswaran Selvakumar, Advait Gosai et al. (+13 more)
Full-duplex spoken dialogue models support low-latency turn taking, interruption handling, and backchanneling, yet a key capability remains underexplored: steerability, the ability to reliably shift conversational behavior along attributes such as tone, persona, speaking rate, and voice style in response to user instructions. We introduce a taxonomy of text- and audio-based steerability that identifies substantial gaps in current full-duplex models. To address this gap, we introduce SteerDuplex, a Moshi-based full-duplex
speech model fine-tuned on natural conversations and synthetic dialogues targeting instruction following, vocal delivery,
reasoning, and duplex interaction. We further apply two-stage reinforcement learning (RL) with hybrid rewards, combining verifiable interaction checks and judge-based semantic feedback to improve timing and response continuity. To evaluate full-duplex spoken steerability, we introduce SteerBench, a
benchmark with 390 spoken
prompts and 1,067 human-authored binary audio and text rubrics spanning tone, persona, style/accent, and speed/length. On SteerBench, supervised training improves audio-steering average pass rate by 44.5 percentage points over the strongest evaluated open baseline. On Audio MultiChallenge, task average pass rate improves by 7 points over its strongest evaluated open baseline. RL further raises source-clean interruption response from 72.5% to 82.5% and reduces synthetic pause barge-in from 26.5% to 9%. Steering and aggregate task scores remain comparable or higher, while reward probes reveal reward hacking through incomplete responses. Our model and
benchmark support systematic research on spoken steerability, with reward analysis showing why timing gains must be evaluated alongside response completeness.
Read the full abstract ▾
All authors Utkarsh Tyagi, Ramaneswaran Selvakumar, Advait Gosai, Sonal Kumar, Nikhil Barhate, Isabell Sagar, Steven Li, Miheer Bavare, Daniel Quigley, Fabiola Tapia Carrillo, Jose M Patron E, Diego Macías Gutiérrez, Paul Song, Ramani Duraiswami, Dinesh Manocha, Yunzhong He
Computer Vision
RA-SOD: Reliability-Aware RGB-T Salient Object Detection under Modality Degradation
Hongbo Gao, Zhengyu Li, Xueru Nie et al. (+4 more)
RGB-Thermal (RGB-T) salient object detection leverages complementary cues from visible and thermal modalities to improve robustness in challenging environments. However, in real-world scenarios, the reliability of each modality is inherently unstable: RGB images degrade under low illumination, motion blur, and noise, while thermal imagery often suffers from contrast compression and sensor artifacts. Such degradation introduces unreliable perceptual evidence that can mislead cross-modal fusion and significantly deteriorate detection performance. To address this challenge, we propose RA-SOD, a reliability-aware RGB-T salient object detection framework that explicitly models modality reliability and integrates it into feature learning and cross-modal fusion. First, we introduce a reliability-conditioned representation that adaptively compensates degraded modality features while preserving structural cues. Second, an uncertainty-guided dual-stream refinement strategy progressively corrects cross-modal representations while suppressing unreliable evidence. Finally, we propose a pixel-wise modality competition mechanism that dynamically selects modality cues according to spatial reliability for fine-grained fusion. Extensive experiments on four
benchmarks (VT821, VT1000, VT5000, and VT-IMAG) demonstrate that RA-SOD achieves state-of-the-art performance and exhibits strong robustness under severe modality degradation. Code and models are available at https://github.com/zaoxienian/RA-SOD.
Read the full abstract ▾
All authors Hongbo Gao, Zhengyu Li, Xueru Nie, Dihao Zhu, Lijun Zhao, Yunke Wang, Chang Xu
Game Theory
Universally truthful mechanisms for scheduling
Georgios Anastasiadis, George Christodoulou, Elias Koutsoupias et al. (+2 more)
We consider universally truthful randomized mechanisms for the problem of scheduling $m$ jobs on $n$ unrelated machines. We prove a lower bound on the expected approximation ratio of every such mechanism whose probability distribution has discrete support. We show that no universally truthful randomized mechanism in this class can achieve approximation ratio smaller than $n/12 - o(n)$ with respect to the optimal makespan. We match this, up to a constant factor, by a mechanism with approximation ratio $n/2 + o(n)$.
Read the full abstract ▾
All authors Georgios Anastasiadis, George Christodoulou, Elias Koutsoupias, Annamaria Kovacs, Conrad Schecker
Machine Learning
Correlation-Guided Fast Machine Unlearning via Hessian Analysis
Ayushi Thakur, Ruchir Gupta, Amit Kumar Jaiswal et al. (+1 more)
The increasing adoption of machine learning in network and distributed security systems has created an urgent need for mechanisms that can selectively and efficiently remove the influence of specific training data to eliminate compromised or
adversarial data points from production models.
Privacy regulations such as GDPR's right to be forgotten also pose similar requirements. However, existing approximate unlearning techniques remain computationally prohibitive for deployment in real-world security systems, as they require repeated expensive Hessian-inverse-vector computations for each data point removal, creating a bottleneck when processing multiple related requests in scenarios such as intrusion detection systems, spam filters, and threat intelligence platforms. Thus, we introduce a computationally efficient unlearning framework that identifies correlated data points in the training set and applies a theoretically derived closed-form parameter update rule, achieving an $82\times$ wall-clock speedup over standard influence function unlearning while preserving model utility with a $10^{-2}$ improvement in accuracy over state-of-the-art baselines. Our method establishes theoretical guarantees and ensures numerical stability through Hessian damping. Our evaluation across seven diverse
dataset architecture combinations, including large-scale CIFAR-100 with ResNet-50, demonstrates superior forgetting effectiveness, with membership inference attack success rates of 0.660 and tug-of-war scores of 0.950.
Read the full abstract ▾
All authors Ayushi Thakur, Ruchir Gupta, Amit Kumar Jaiswal, Prayag Tiwari
Discrete Mathematics
Fault-tolerant Hamiltonian connectivity of Johnson graphs
Huazhong Lü and Jinhao Liu
Johnson graphs $J(n,k)$ are a classical family of highly symmetric networks known to be
Hamiltonian-connected in the fault-free setting. In this paper, we investigate their
Hamiltonian connectivity under three failure models, namely general edge faults, matching faults, and vertex faults. For general edge faults, we prove that $J(n,k)$ remains
Hamiltonian-connected after the deletion of any set of at most $k(n-k)-3$ edges for $n\geq4$. Since $J(n,k)$ is $k(n-k)$-regular, this attains the natural degree-based upper bound for
Hamiltonian connectivity. We then consider matching faults, which exclude the concentration of multiple faulty links at a single vertex and permit substantially larger fault sets. We show that $J(n,k)$ remains
Hamiltonian-connected after the deletion of an arbitrary matching for $n\geq5$, including a perfect matching whenever one exists. For vertex failures, we prove that $J(n,k)$ is $(n-2)$-vertex-fault-tolerant
Hamiltonian-connected for $n\geq5$. All three results are constructive and lead to recursive fault-tolerant
Hamiltonian routing
algorithms. Simulation results on Johnson graphs with up to $12{,}870$ vertices further show that the routing
algorithms successfully construct fault-free
Hamiltonian paths for all tested source-destination pairs, with measured execution times exhibiting near-linear growth with network size. These results establish a unified fault-tolerant
Hamiltonian-connectivity framework for Johnson graphs under different failure patterns.
Read the full abstract ▾
Numerical Analysis
A parameterized block LU preconditioning for double saddle point problems
Hamed Aslani, Mehdi Najafi Kalyani and Davod Khojasteh Salkuyeh
We present a biparametric preconditioning technique for large, sparse, non-symmetric, and non-singular double saddle point problems. The preconditioner is induced using a stationary iteration method. It is also based on a two-parameterized block LU factorization of the coefficient
matrix. To begin with, some convergence results of the iteration method are derived. Moreover, the spectral features exhibited by the preconditioned
matrix are given. Finally, the presented preconditioner is conjuncted with the GMRES method. Numerical results demonstrate the satisfactory performance of the preconditioner.
Read the full abstract ▾