Systems and Control
Construction of Control Lyapunov-Barrier Functions from CLF-CBF Pairs
Bo Wang and Miroslav Krstic
This paper studies the construction of control Lyapunov-barrier functions (CLBFs) from a given control Lyapunov function (CLF) $V$ and control barrier function (CBF) $h$. We consider functions of the form $W=F(V,h)$ that increase with the CLF value and do not increase with the barrier value, and show that the CLBF decrease condition is completely characterized by a nonnegative scalar weight that governs the relative contributions of the CLF and CBF. Because this weight depends only on $(V,h)$, the same value must satisfy the decrease condition at all states sharing the same CLF and CBF values, which leads to a joint-level-set admissibility condition. We then construct CLBFs from admissible weights through a first-order partial differential equation (PDE) and obtain explicit multiplicative and power-type families as special cases. We further identify an integrability obstruction showing that admissibility alone does not guarantee properness. Two nonlinear examples illustrate the constructions and demonstrate safe stabilization in cases where feedback based on the original CLF violates the safety constraint.
Read the full abstract ▾
Distributed Computing
Building py-kvcache: A Performance Characterization of External KV Caching for vLLM with NVMe SSDs
Joseph Kanichai, Tiziano De Matteis and Animesh Trivedi
Prefix caching can reduce the time to first
token (TTFT) of long-context LLM requests by reusing previously computed key-value (KV) states, but for short prefixes or fast GPUs, recomputation can be faster than loading from an external cache. We characterize this tradeoff in vLLM across GPU, CPU, and NVMe tiers using synthetic workloads, long-context
benchmarks, production traces, and find that cache performance depends on transfer granularity, intermediate
memory use, and when transfers enter the request schedule, not only on device bandwidth. These findings motivate py-kvcache, a vLLM KV Offload connector with asynchronous direct I/O, bounded shared staging, and scheduler-aware preloading, which starts disk reads while requests are still waiting, overlapping with compute. At 80k
tokens, py-kvcache loading from disk is 2.0x faster than LMCache, with preloading contributing 1.34x. With GPU, CPU, and disk caching enabled, it is 1.23x faster than LMCache and within approximately 4% of the native vLLM KV Offload implementation. LongBench and SCBench show that these benefits extend to irregular prefix chains and multi-turn workloads. Bailian trace replays improve TTFT on a weaker GPU, but on an H100 the average request falls below the break-even point and GPU
memory alone retains enough prefixes. External KV caching should therefore be treated as a setup specific admission decision. The py-kvcacheimplementation is available at: https://github.com/atlarge-research/py-kvcache.
Read the full abstract ▾
Computation and Language
LOCUS: Task-Aware Low-Rank Post-Training for Token-Efficient Language Generation
Dongfang Zhao
Large
language model serving costs scale directly with output sequence length, yet standard preference
alignment often inflates response verbosity without improving utility. We study whether the parameterization of post-training updates affects generation length: low-rank subspaces alter sequence length without modifying the
alignment loss. We present LOCUS, a method that selects a task-aware low-rank adaptation subspace to minimize output-
token cost subject to a utility constraint. Within this subspace, post-training retains the native preference objective with a frozen backbone. Across Anthropic HH-RLHF dialogue preferences, we evaluate two $\sim$3B decoder backbones, Pythia-2.8B and Qwen2.5-3B, against protocol-matched full-parameter DPO and DrDPO branches and the released SamPO checkpoint. LOCUS reduces continuation length by up to 39.84% on Pythia-2.8B and by 14.87–17.58% on Qwen2.5-3B while updating only 0.24–0.28% of model parameters, with no material change in the internal preference diagnostic.
Read the full abstract ▾
Multiagent Systems
Organizational Principles Enable Collective Intelligence in Embodied AI
Zhengran Ji, Jonathan Hyun and Boyuan Chen
Collective intelligence depends not only on the capabilities of individual members, but also on how those members are organized. Yet artificial multi-
agent systems are typically assembled using fixed organizational structures, even when the physical tasks they perform impose fundamentally different coordination requirements. Here we show that principles from human organization theory can be operationalized to organize large, heterogeneous collectives of
embodied artificial
agents. We introduce ORCH (Organizing Roles and Coordination Hierarchies), which constructs task-specific hierarchical organizations by combining pooled interdependence for work that can proceed concurrently with sequential interdependence for work governed by prerequisite relationships. Across 25 wildfire-response missions spanning reconnaissance, rescue, transportation, resource management, containment and suppression, we evaluated teams of up to 50 heterogeneous
agents using eight large
language models. Organizations constructed using these principles consistently outperformed four representative
embodied multi-
agent approaches across mission outcome, execution efficiency, exploration and computational resource use. Human-designed ORCH organizations improved final score by 63.97% and execution efficiency by 74.29% on average relative to the four prior frameworks. Organizations generated automatically by
language models improved these measures by 43.63% and 52.53%, respectively. These advantages persisted across missions and underlying
language models. Notably, collective performance was not monotonically determined by model scale. Analysis of long-horizon missions showed that hierarchical organization enabled teams to preserve concurrent activity within specialized groups while coordinating ordered transitions between mission phases.
Read the full abstract ▾
Machine Learning · Cross-listed from quant-ph
Learning structural balance of graphs from quantum spectral features
Stefano Scali and Oleksandr Kyriienko
We develop a
quantum approach to spectral feature extraction from the density of states (DOS) of a problem-dependent
Hamiltonian, and apply it to machine learning on signed graphs. We propose to embed a signed graph as an Ising model instance with positive and negative interactions, and use the standardized moments of the Ising DOS as features for learning. We show that these moments count signed closed walks, are switching-invariant, and are size-free by construction. As a
benchmark, we target learning the frustration index, an NP-hard measure of structural balance that can be labeled exactly at moderate size. At zero field, the models can be sampled classically, allowing the
quantum extraction procedure to be certified against exact ground truth. We propose DOS-QPE, a phase estimation on a purified maximally mixed probe, which samples the spectral density with orders of magnitude fewer shots than Hadamard test-based trace sampling and feeds the resulting features directly into classically trained models. On $1.4\times10^5$ labeled graphs the exact DOS determines the frustration index, and five moments recover it with a mean error of 0.4, well below one sign flip. Beyond zero field, the underlying trace-estimation problem is DQC1-complete, providing access to spectral features for which no efficient classical sampling method is known. Our work opens routes towards
quantum applications in social network balance analysis,
spin-glass studies, correlation clustering, and
protein-interaction networks.
Read the full abstract ▾
Robotics
Reflex-Informed Neuromuscular Reinforcement Learning for Muscle-Driven Locomotion
Jian Zhou, Xingyu Zhang, Rui Ma et al. (+3 more)
Muscle-driven locomotion provides a physically grounded approach to generating realistic human movement. However, achieving both physiological plausibility and adaptability to changes in musculoskeletal capacity and external disturbances remains a fundamental challenge. To address this limitation, we propose a Reflex-Informed Neuromuscular Reinforcement Learning framework for muscle-driven locomotion. Within this framework, a fixed phase-dependent reflex controller serves as the underlying neuromuscular control mechanism, while the reinforcement learning policy produces four biomechanically meaningful residual parameters to modulate key reflex gains and thresholds associated with hip swing, knee support, and ankle propulsion according to the current state. Experimental results demonstrate that the proposed framework generates physiologically plausible locomotion with improved kinematic accuracy and dynamic consistency, as well as better bilateral
symmetry and stride-to-stride consistency under nominal walking conditions. The learned policy remains robust under muscle weakness and external perturbations without retraining.
Read the full abstract ▾
All authors Jian Zhou, Xingyu Zhang, Rui Ma, Yu Cao, Shane Xie, Zhi-qiang Zhang
Software Engineering
Reproducibility in the Age of Agentic AI: Context Engineering at the Timescale of a Codebase
Lorena A. Barba
Reproducible research practices are context engineering for AI coding
agents. I argue that
agents lower the cost of maintaining tests, commit histories, repository structure, instructions, and decision records while making their benefits immediate. Researchers remain responsible for verifying these artifacts and the scientific judgments they encode.
Read the full abstract ▾
Sound
Continuous-Time Acoustic Modelling with Neural Controlled Differential Equations
Mattias Cross, Minghui Zhao and Anton Ragni
Text-to-
speech (TTS) models commonly address text–
speech alignment by expanding phone-level encoder states to frame-level decoder inputs using predicted durations. While this length-regulation step resolves
alignment structurally, this use of duration typically changes only where and how often latent states appear, not the values of the states themselves. This paper proposes a continuous-time mechanism for duration-aware acoustic modelling in TTS using neural controlled differential equations (CDEs). We formulate the phone representation as a temporally parameterised control path and use a neural acoustic vector field to produce a continuous-time hidden state whose values evolve with phonetic content and duration-derived timing. The resulting trajectory can be sampled at discrete points and integrated into a standard acoustic decoder pipeline. Objective results contrast CDEs and typical recurrent models. Subjective results suggest that CDE-based models evaluating one phone per step can improve rank-order agreement between synthesised and reference
emotion intensity while maintaining comparable
emotion-expression quality to a strong baseline. Additional experiments with half-phone step-sizes suggest that temporal resolution changes the trade-off between style tracking and absolute calibration. These results position CDEs as a promising design space for continuous-time and duration-aware style-sensitive TTS.
Read the full abstract ▾
Computation and Language
The Eloquence submission for Task 2 of the Interspeech 2026 MLC-SLM challenge
Jordi Luque, Lorenzo Concina, Marco Matassoni et al. (+2 more)
This paper details the Eloquence team's approach to Task 2 of the 2nd MLC-SLM challenge at Interspeech 2026, which involves
multilingual Multiple-Choice Question Answering (MCQA) across 21
languages. Three approaches are explored. First, we fine-tune Voxtral-Mini-3B via LoRA with cross-lingual data augmentation, ASR transcript augmentation and timestamp-aware audio cropping, achieving 0.72 macro-accuracy on evaluation Phase 2. Second, we apply
multimodal in-context learning (ICL) to the frozen Voxtral-24B model to correct a strong label bias, reaching 0.81, our best result. Third, a training-free
retrieval system based on a three-layer voice-anchored
memory combining acoustic identity, semantic content, and a knowledge graph achieves 0.68. All three systems substantially outperform the official baseline.
Read the full abstract ▾
All authors Jordi Luque, Lorenzo Concina, Marco Matassoni, Alessio Brutti, Filippo Vella
Computer Vision
Revisiting Avatar-As-Image: High-Fidelity Registration is All You Need
Margaret Kostyrko, Yuxuan Xue, Garvita Tiwari et al. (+1 more)
The representation of 3D clothed humans as standardized 2D UV texture and displacement maps over an underlying body model has long been studied. This compact representation is enticing as it enables pretrained image networks to process, generate, and edit 3D avatars, but is only useful if scans are accurately aligned and brought into correspondence via high-fidelity registration. This prerequisite has never been met, which we argue explains the limited quality of prior UV-based methods for clothed humans. Despite its significance, no public method produces high-fidelity SMPL(-X)+D registrations with UV texture from arbitrary clothed scans. We present AvaImg, a multi-stage optimization pipeline, to close this gap: it enforces body-inside-clothing constraint via signed winding numbers, made viable by a three-level efficiency cascade (~10x runtime reduced, ~95% storage saved), and recovers fine surface detail using coarse-to-fine displacement optimization. AvaImg outperforms all baselines in body fitting, shape estimation, and surface registration across six
datasets, yielding textured registrations near-indistinguishable from scans (PSNR=34.48dB). For validation of AvaImg's Avatar-as-Image representation as imminently compatible with image foundation models, we auto-encode our UV maps via the frozen FLUX VAE. This achieves only 0.76mm added Chamfer error relative to scan and shows that the resulting maps lie within natural-image distributions, supporting the use of 2D generative priors for 3D avatar generation. Code, data, and Singularity containers will be at https://yuxuan-xue.com/avaimg.
Read the full abstract ▾
All authors Margaret Kostyrko, Yuxuan Xue, Garvita Tiwari, Gerard Pons-Moll
Numerical Analysis · Cross-listed from math.OC
Verified Linear Programming via Tolerance-Aware Precision Boosting
Ernesto Casablanca, Martin Sidaway, Sadegh Soudjani et al. (+1 more)
Linear programming plays a fundamental role in computer science, with applications in optimization, formal verification, SMT solving, and numerous other domains. When exactness guarantees are required, numerical inaccuracies arising from floating-point arithmetic can compromise soundness, whereas exact rational arithmetic often incurs significant computational overhead. In this paper, we investigate the numerical stability of the simplex
algorithm and establish conditions under which a precision-boosting floating-point implementation provably produces the same pivot decisions and final basis as an exact rational implementation, from which the final result is reconstructed and certified exactly. Our analysis shows that correctness depends on both arithmetic precision and a careful handling of numerical tolerances. Based on these results, we develop a tolerance-aware, precision-boosting simplex
algorithm with formal correctness guarantees. Finally, we introduce a delta-complete termination criterion that allows the
algorithm to terminate once certified upper and lower bounds on the optimal objective differ by at most a user-specified threshold delta, providing a certified, user-controlled optimality gap.
Read the full abstract ▾
All authors Ernesto Casablanca, Martin Sidaway, Sadegh Soudjani, Paolo Zuliani
Computer Vision
MC-DeTra: Motion-Consistent Joint Object Detection and Socially-Aware Trajectory Forecasting in Bird's-Eye-View Images
Vladislav Diuzhev and Dmitry Yudin
Unified models for object detection and trajectory
forecasting aim to merge perception and prediction for autonomous driving, refining actor trajectories directly over shared bird's-eye-view (BEV) images rasterized from
LiDAR and high-definition maps. Their accuracy on dynamic, moving actors, however, remains the hardest part of the task, and the strongest such model, DeTra, has no public implementation. We contribute an openly released DeTra reimplementation with documented approximations, and on top of it MC-DeTra: a family of motion-consistency mechanisms that add supervision through two annotation-derived auxiliary signals – each actor's observed past motion and the occupancy of the surrounding traffic that forms its social context – and one inter-output consistency constraint that aligns an actor's predicted heading with its predicted direction of motion. Every proposed loss is train-only and inference-safe: it shapes the shared BEV representation during training and is removed at test time, adding no inference latency. On the Waymo Open
Dataset, evaluated under a strict, detection-conditioned
forecasting protocol, MC-DeTra improves dynamic, socially-situated trajectory
forecasting while preserving or improving detection accuracy; a gradient-based loss-calibration analysis exposes how the auxiliary objectives compete at the shared backbone, and our ablation identifies which signals contribute most. We release code, configurations, and evaluation tooling at https://github.com/diuzhevVlad/MC-DeTra.
Read the full abstract ▾
Machine Learning
Why Does Post-Training Quantization Work?
Yuxiang Chen, Michael Beyer, Jun Zhu et al. (+1 more)
Post-training quantization compresses large
language models (LLMs) by storing their weights at reduced precision, and each quantized weight introduces an error into the hidden states. Naively, these errors should accumulate with depth and corrupt next-
token prediction; randomly initialized models accumulate these discrepancies rapidly, whereas quantized pretrained models accumulate much less hidden-state error and largely maintain downstream task performance, even though they were never trained with quantization noise. This raises the question we address: why does post-training quantization work? Comparing full-precision and quantized forward passes, we identify two mechanisms that characterize pretrained quantization robustness. First, the error a layer newly introduces tends to oppose the error it inherits from the layer's input. The two cancel partially such that the discrepancy between full-precision and quantized passes grows slowly. This counteracting residual interaction develops during pretraining. Our quantitative analysis identifies it as a major factor slowing hidden-error growth. Second, LM-head geometry preferentially preserves the scores and probabilities of high-ranked
tokens, which typically represent the model's most confident predictions. Together, these mechanisms explain why quantization error that passes through numerous layers can still produce only small output changes, and we verify the findings across models and quantization settings.
Read the full abstract ▾
All authors Yuxiang Chen, Michael Beyer, Jun Zhu, Jianfei Chen
Emerging Technologies
A Time-Based Readout for Vector-Matrix Multiplication in Fully Analog Memristive SNNs
Elia Mateu-Barriendos, Álvaro Gómez-Pau, Josep Rius et al. (+3 more)
Artificial neural networks rely on vector-
matrix multiplications (VMMs), whose implementation in von Neumann architectures is dominated by costly data movement between
memory and processing units. Spiking neural networks (SNNs) mitigate this bottleneck by performing in-
memory, analog VMMs using memristive crossbar arrays. However, conventional current-mode readout
circuits incur significant area and power overhead. This work proposes a fully analog readout architecture based on voltage-to-time conversion of the VMM output. By sensing the column voltage, the proposed approach avoids current-mode summing and scaling circuitry, improving area and energy efficiency. Post-layout simulations of a 10x1 SNN implemented in a 130 nm CMOS technology validate the proposed architecture, while application to a trained 64x10 SNN for digit classification further demonstrates its feasibility for SNN inference.
Read the full abstract ▾
All authors Elia Mateu-Barriendos, Álvaro Gómez-Pau, Josep Rius, Daniel Arumí, Rosa Rodríguez-Montañés, Salvador Manich
Machine Learning · Cross-listed from stat.ML
Generalization Analysis of Distributed Kernel-based Robust Gradient Descent Algorithms
Jun-Yi Meng, Zheng-Chu Guo and Yuan Mao
In this paper, we investigate the generalization performance of distributed gradient descent
algorithms in a reproducing kernel
Hilbert space under a robust loss function $l_{\sigma}$. By exploiting the spectral characterization of gradient descent together with the intrinsic properties of robust loss functions, we establish optimal learning rates for the distributed kernel-based robust gradient descent (DKRGD)
algorithm with an appropriately chosen scale parameter $\sigma$. The proposed parameter choice of $\sigma$ simultaneously alleviates the saturation phenomenon and guarantees statistical robustness. A key technical contribution is a novel error analysis that provides substantially sharper bounds for products of operators, thereby significantly relaxing existing restrictions on the maximum number of local machines while retaining optimal learning rates. Finally, we develop a communication-efficient strategy that further improves the convergence performance of DKRGD.
Read the full abstract ▾
Systems and Control
Mixed-integer optimization for multi-year military aircraft fleet management
Matteo Vescovi, Raffaele Giuseppe Cestari, Roberto Valdambrini et al. (+3 more)
While existing strategies for Flight and Maintenance Planning for the defense sector generally address idealized conditions, real-world planning often involve non-nominal initial fleet states and complex inspection schemes. To address these challenges, we propose a multi-year planning strategy that maximizes long-term fleet availability. The formulation incorporates multiple competing objectives capturing the tight coupling between aircraft usage and maintenance, while enforcing cyclic inspection requirements and limited maintenance-dock capacity. To enhance practical applicability, we introduce three complementary strategies: a receding-horizon approach to reduce computational burden, a
Bayesian Optimization routine to automatically tune cost-function coefficients, and an aircraft-specific weighting scheme that adapts model priorities based on fleet-condition statistics. Results on three realistic
benchmark scenarios demonstrate the effectiveness of the approach.
Read the full abstract ▾
All authors Matteo Vescovi, Raffaele Giuseppe Cestari, Roberto Valdambrini, Andrea Mercurio, Valentina Breschi, Mara Tanelli
Artificial Intelligence
When Agents Disagree: Bayesian Backward Reasoning as a Label-Free Anchor for Multi-Agent Collective Decision-Making
Ken Chen, Wei Wang, Sachith Seneviratne et al. (+2 more)
When multiple LLM
agents yield conflicting answers, the decision-making process dictates whether
agent diversity improves performance or merely compounds shared errors. Existing collective decision-making methods, including voting, electoral rules, and LLM judges, rely on forward
reasoning: they map evidence to labels in one direction. Although these methods can combine diverse forward traces, they still aggregate estimates that share this evidence-to-label factorization and can inherit correlated errors within the forward pool. We therefore construct a reverse posterior for each instance through
Bayesian backward
reasoning from an explicit likelihood. The forward and reverse posteriors provide differently factorized approximations of the underlying posterior. Because estimates from different factorizations may tend to share the same error less often, we use Jensen-Shannon divergence to rank
agents by cross-path consistency. This cross-path consistency signal underlies three strategies: hard selection (MinJS), soft reweighting (FwdJS), and log-linear fusion (LogLin). Evaluated on DDXPlus across five LLM backbones, our proposed strategies show consistent improvements: MinJS outperforms random selection across all backbones, FwdJS generally improves over the strongest baseline, and LogLin achieves the best performance among the evaluated methods, with its largest gains on the subset where the
agents disagree. Despite its weaker standalone accuracy, the reverse posterior serves as a more useful anchor than forward-only alternatives, providing complementary information for collective decision-making. When labeled data are available, a lightweight two-stage calibration can further refine the reverse anchor and improve aggregation performance.
Read the full abstract ▾
All authors Ken Chen, Wei Wang, Sachith Seneviratne, Hansani Weeratunge, Saman Halgamuge
Computer Vision
Language-Augmented Semantic Priors for B-Spline Surface Fitting
Yunzhong Lou, Yusheng Luo, Jiahao Li et al. (+2 more)
The use of B-splines and Non-Uniform Rational B-Splines surfaces constitutes the mathematical foundation of contemporary computer-aided design (CAD) systems. Despite long-term progress,
geometric kernels in traditional CAD still rely heavily on predetermined heuristic initialization for surface fitting and parameterization. Meanwhile, the procedural semantics and design intent encoded in modeling histories are largely ignored during geometry generation. This disconnect creates a gap between high-level design intent and solver-executable
geometric configuration, often leading to suboptimal and semantically inconsistent fitting results. To bridge this gap, we introduce LASP, a
Language-Augmented Semantic Priors framework that leverages large
language models (LLMs) to infer structured, solver-usable B-spline priors from procedural modeling histories. Rather than modifying the
geometric kernel itself, LASP operates as a semantic
reasoning layer above existing solvers. It first translates modeling histories into rich textual descriptions that capture design intent,
geometric context, and functional relationships, and then uses a fine-tuned LLM to predict structured B-spline prior parameters. LASP is trained through a two-stage scheme that combines local
geometric regularities with long-range contextual dependencies, producing priors that are both interpretable and semantically coherent. This approach furnishes inductive signals that direct the conventional B-spline fitting process toward solutions that more accurately encapsulate the intended design objectives and demonstrate heightened semantic coherence. Compared to traditional machine learning schemes, the experiments demonstrate that
language-driven
reasoning can serve as a powerful inductive bias for
geometric solving, establishing a new paradigm of
language-guided
geometric optimization in modern CAD systems.
Read the full abstract ▾
All authors Yunzhong Lou, Yusheng Luo, Jiahao Li, Yu Song, Xiangdong Zhou
Human-Computer Interaction
Bigger than the EAR BOX: A Theory-Grounded Review of XR Accessibility Research for Deaf and Hard of Hearing Communities
Shuxu Huffman, Michaela Okosi, Jose Merino et al. (+4 more)
Extended Reality (XR) technologies have received growing
attention in accessibility research involving Deaf and Hard of Hearing (DHH) communities. Yet less
attention has been given to the assumptions shaping this work. We present a theory-grounded review of XR research involving DHH users. Drawing on Disability Studies, Deaf Studies, and DeafSpace, we develop a theoretical framework with four analytical dimensions: orientation toward access, distribution of responsibility, conceptualization of DHH communities, and spatial and perceptual assumptions. We apply this framework to 53 XR studies involving DHH users, identified through a search and screening of ACM publications from 2015 to 2025. Our analysis shows that XR accessibility is frequently framed as supporting communication within hearing-default environments, while overlooking the diversity of DHH communities. We identify directions for redistributing accessibility labor and reconfiguring space, and offer our framework as a tool for future XR research involving DHH communities.
Read the full abstract ▾
All authors Shuxu Huffman, Michaela Okosi, Jose Merino, Tifanie Bouchara, Abraham Glasser, Christian Vogler, Raja Kushalnagar
Computer Vision
Spectral Adapters for Segment Anything Model-based Segmentation of Colorectal Liver Metastases in Computed Tomography
Ramtin Mojtahedi, Mohammad Hamghalam, Jacob J. Peoples et al. (+6 more)
Accurate
segmentation of colorectal liver metastases (CRLM) in contrast-enhanced computed
tomography (CT) is important for response assessment,
surgical planning, and follow-up. We propose two parameter-efficient spectral adapters for the Segment Anything Model (SAM): the Directional Spectral Adapter (DiSECT) and Spectral Instance-Guided Adapter (SiGA). DiSECT uses singular value decomposition of frozen weights to constrain residual updates to leading spectral directions, while SiGA adds global and input-conditioned gating through a multilayer perceptron. We evaluate these methods on 446 contrast-enhanced CT volumes (355 training, 91 testing) and compare them with LoRA, QLoRA, convolutional adapters (CAD), and a 3D nnU-Net baseline. Experiments consider single-point, three-point, bounding-box, and no-
prompt regimes. SiGA achieves the best single-point performance with a Dice score of 0.77, IoU of 0.69, and HD95 of 35.39 mm. Under no-
prompt inference, SiGA reaches 0.76 Dice, 0.68 IoU, and 46.76 mm HD95, comparable to the nnU-Net baseline (0.758 Dice). DiSECT uses only 0.14 million trainable parameters. These results show that spectral adapters can efficiently adapt SAM for CRLM
segmentation while retaining strong accuracy with limited trainable parameters.
Read the full abstract ▾
All authors Ramtin Mojtahedi, Mohammad Hamghalam, Jacob J. Peoples, Natalie Gangai, Mithat Gonen, Yun Shin Chun, HyunSeon Christine Kang, Richard K. G. Do, Amber L. Simpson
Numerical Analysis · Cross-listed from eess.SP
$hp$-adaptive trees for graph signal approximation
Giacomo Elefante, Wolfgang Erb and Michael Multerer
Tree-encoded partitionings of graphs are fundamental tools for the decomposition and approximation of graph signals. For the efficient approximation of such graph signals, we develop strategies based on $hp$-refinement by combining domain decomposition with an improved local approximation using
polynomials of higher degree. In this way, from a given graph partitioning tree, a more efficient subtree is extracted in which the cost of the signal approximation is considerably reduced by still maintaining the same total error. To this end, we interpret the refinement process as a binary knapsack problem to determine an enhanced partitioning tree. We further study an a-posteriori strategy which prunes the partitioning tree by optimizing the
polynomial degrees over the subdomains. To make
polynomial basis systems accessible for general graphs or high-dimensional data, we propose local
embeddings of graphs into low dimensional Euclidean spaces. We underpin the efficiency of our
algorithms with extensive numerical tests which carefully assess the impact of the applied refinements and optimization strategies.
Read the full abstract ▾
Computation and Language
Negative Self-Distillation: Learning to Reason by Avoiding Flaws
Rongcan Pei, Zhepei Wei, Shuyao Xu et al. (+3 more)
On-Policy Self-
Distillation (OPSD) has emerged as a popular paradigm for large
language model (LLM) self-improvement, allowing models to act as their own teachers by leveraging privileged information such as ground-truth solutions. However, recent findings indicate that OPSD can severely degrade the performance of LLMs on complex
reasoning tasks: By forcing the student to imitate an artificially confident
reasoning trace conditioned on privileged information, OPSD inadvertently suppresses expressions of uncertainty and penalizes the exploratory, self-corrective behaviors required to solve challenging problems. To address this, we introduce Negative Self-
Distillation (NSD), a new framework that optimizes LLMs by diverging from flawed
reasoning rather than imitating privileged solutions. Instead of relying on ground-truth answers or external supervision, NSD uses the model itself to generate a question-specific negative condition (eg, acting as a “careless reasoner”) and pushes the student's distribution away from this self-generated negative teacher. Naively applying unlearning objectives to achieve this divergence is problematic, as flawed
reasoning tokens are confounded with basic linguistic
tokens; indiscriminately penalizing both risks catastrophically degrading the model's foundational
language capabilities. We resolve this by designing a dynamic gating mechanism that automatically identifies and isolates
reasoning-critical
tokens, ensuring gradient updates target only behavioral flaws while preserving the model's linguistic priors. Empirically, NSD consistently outperforms OPSD and other label-free, self-bootstrapping reinforcement learning (RL) baselines.
Read the full abstract ▾
All authors Rongcan Pei, Zhepei Wei, Shuyao Xu, Xinyu Zhu, Wei-Lin Chen, Yu Meng
Robotics
Aerodynamic Prior-Free Coordinated Trajectory Generation and Tracking Control for a Tail-Sitter UAV
Erchao Rong, Zihao Liu, Junning Liang et al. (+5 more)
This paper presents a coordinated trajectory generation and tracking control framework for a tail-sitter unmanned aerial vehicle (
UAV), which does not require aerodynamic priors identified for a specific airframe while addressing the challenge of flight control under highly nonlinear aerodynamics across the full flight envelope. The core innovation lies in employing phase-specific aerodynamic modeling strategies for planning and tracking, tailored to their distinct functional characteristics, without requiring airframe-specific aerodynamic priors. Specifically, the phi-theory model under coordinated flight is employed to derive an analytic differential flatness mapping, and a simplified but locally accurate model is established for predictive control to enable real-time aerodynamic parameter estimation. The proposed framework is evaluated extensively through both simulation and challenging real-world flight tests under mild wind conditions, showing high-precision tracking and adaptability across the tested aerodynamic conditions. To the best of our knowledge, this is the first real-world demonstration of accurate trajectory tracking over tested flight regimes spanning the full envelope of a tail-sitter
UAV without relying on aerodynamic identification campaigns. The source code of our framework is available at: https://github.com/SYSU-HILAB/AP-PnC.
Read the full abstract ▾
All authors Erchao Rong, Zihao Liu, Junning Liang, Jianguo Wang, Xiao Jie, Haoran Fu, Ziliang Chen, Ximin Lyu
Robotics
ActSafeGuard: Differentiable and Training-Aligned Constraint Enforcement for Flow-Matching Policies
Jianming Ma, Rongjun Jin, Xiaxi Si et al. (+3 more)
Vision-
Language-Action (VLA) and World-Action Models (WAMs) have demonstrated strong capabilities in general-purpose robotic manipulation, yet their generated actions may violate hard physical constraints and therefore be unsafe or infeasible for deployment. Existing safety approaches either optimize statistical safety objectives without deterministic per-step guarantees or correct unsafe actions only during inference, creating a mismatch between policy training and execution. We introduce ActSafeGuard, a differentiable and training-aligned safeguard layer for flow-matching based policies. ActSafeGuard integrates hard action feasibility into policy learning, not merely treating safety as an inference-time external component. Through an analytical ray-scaling operator design, ActSafeGuard enables boundary-aware gradients to guide the model to naturally learn constrained
manifolds. Extensive experiments on multiple standard foundation backbones ($\pi_{0.5}$ and Fast-WAM) across various tasks demonstrate that ActSafeGuard consistently achieves a $100\%$ step safety rate while fully preserving or even boosting task success rates, providing a scalable and minimally invasive solution for safe
embodied AI deployment.
Read the full abstract ▾
All authors Jianming Ma, Rongjun Jin, Xiaxi Si, Yang Zhang, Yiheng Li, Yue Gao
Networking
Electroencephalography Signal Analysis for Human Activities Classification: A Solution Based on Machine Learning and Motor Imagery
Tarciana C de Brito Guerra, Taline Nóbrega, Edgard Morya et al. (+2 more)
Electroencephalography (
EEG) is a fundamental tool for understanding the
brain's electrical activity related to human motor activities.
Brain-Computer Interface (BCI) uses such electrical activity to develop assistive technologies, especially those directed at people with physical disabilities. However, extracting signal features and patterns is still complex, sometimes delegated to machine learning (ML)
algorithms. Therefore, this work aims to develop a ML based on the Random Forest
algorithm to classify
EEG signals from subjects performing real and imagery motor activities. The interpretation and correct classification of
EEG signals allow the development of tools controlled by cognitive processes. We evaluated our ML Random Forest
algorithm using a consumer and a research-grade
EEG system. Random Forest efficiently distinguishes imagery and real activities and defines the related body part, even with consumer-grade
EEG. However, interpersonal variability of the
EEG signals negatively affects the classification process.
Read the full abstract ▾
All authors Tarciana C de Brito Guerra, Taline Nóbrega, Edgard Morya, Allan de M. Martins, Vicente A de Sousa Jr
Numerical Analysis · Cross-listed from math.AP
Numerical Study of Reported Multiple Solutions to a Cauchy Problem for the 2d Incompressible Euler Equations
Allen Tesdall and Richard Sanders
Bressan and Shen (2021) jointly with Bressan and Murray (2020) proposed a simple example of a Cauchy problem on $\mathbb{R}^2$, characterized by two key parameters, intended to yield nonunique weak solutions of the incompressible Euler equations. They suggest that a limiting process applied to two specific approximations of their example initial data yields two distinct solutions. Here we provide evidence coming from a carefully designed numerical procedure to support if, or when, this intriguing claim is valid. Some of the numerical challenges encountered in the present study are: (i) the proposed initial data is unbounded in a neighborhood of the origin, (ii) the proposed problem is posed on all of $\mathbb{R}^2$, (iii) several essential
symmetries present in the true solution must be captured, (iv) the proposed example depends on two parameters which may be critically linked. We demonstrate that an extremely large computational domain is mathematically necessary to capture the core
symmetries of the exact free-space problem. Accordingly we implement a nested-grid strategy specifically designed for its particular suitability and efficacy for the problem at hand. Our strategy also allows us to achieve extreme grid refinement in the region where solution accuracy is most critical. Within our nested-grid framework, we design an underlying discretization scheme that is high-order, robust, and fast. We consider a broad collection of allowable parameter pairs, but among all of them we find only one that may satisfy the suggested claim.
Read the full abstract ▾
Discrete Mathematics
An FPTAS for Two-Machine Open-Shop Scheduling with a Single Unavailability Interval
Hao Lu, Yuan Yuan, Xingwu Liu et al. (+2 more)
We consider the two-machine open-shop scheduling problem in which one machine is unavailable during a fixed interval. We study the resumable setting: an operation interrupted by the unavailability interval may resume, without penalty, when the machine becomes available. The objective is to minimize the makespan. Although the problem is NP-hard and several approximation
algorithms are known, whether it admits a fully
polynomial-time approximation scheme (FPTAS) has remained open for two decades. We resolve this question affirmatively by giving the first FPTAS, thereby strengthening the previously known
polynomial-time approximation scheme (PTAS). As an intermediate result, we develop a new pseudo-
polynomial dynamic program with seven state dimensions, improving on the ten-dimensional formulation in the literature.
Read the full abstract ▾
All authors Hao Lu, Yuan Yuan, Xingwu Liu, Xin Han, Yong Zhou
Machine Learning · Cross-listed from stat.AP
Geospatial Foundation Models Capture Health-Relevant Dimensions of Place Beyond Conventional Social Risk Indices
Nathaniel Hendrix, Carl Y. Zhang, Chris Heitzig et al. (+2 more)
Area-based social risk indices summarize residents' socioeconomic conditions but incompletely capture physical features of place that may affect health. We evaluated whether numerical representations of physical place produced by four geospatial foundation model families from 2022 satellite data explained residual variance in tract-level associations between the Area Deprivation Index, Social Deprivation Index, and Social Vulnerability Index with health outcomes. We used LightGBM to predict variables from the American Community Survey and 40 chronic disease and health-behavior outcomes from CDC PLACES across 82,646 census tracts in the contiguous United States, evaluating performance across 10 held-out states. Among survey variables, models were moderately predictive of some variables including housing type (R-squared up to 0.54) but weak for disability, unemployment, and income disparity. For health outcomes, models explained up to 54% of variance left unexplained by social risk indices, with the largest gains for annual checkups, arthritis, and high blood pressure. Mean total variance explained by geospatial foundation models across the 40 health-related outcomes increased from 0.31 in the smallest tract-size decile to 0.39 in the largest. Geospatial foundation models capture health-relevant features of place not represented by conventional social risk indices and may usefully augment them in epidemiological analyses.
Read the full abstract ▾
All authors Nathaniel Hendrix, Carl Y. Zhang, Chris Heitzig, Andrew Bazemore, David H. Rehkopf
Computation and Language
Structured Transforms for Low-Overhead Quantization of Language Models
Daria Cherniuk, Alexander Rudikov, Boris Kashin et al. (+1 more)
We revisit Kashin-decomposition-based weight quantization for large
language models and propose an improved
algorithm with stronger convergence properties and structured, efficient orthogonal transforms. The method retains the core factorization of each weight into two components – one with bounded infinity norm and the other with bounded infinity norm after an orthogonal transformation – but replaces the dense random orthogonal
matrix with a sign-randomized Discrete Cosine Transform (DCT), reducing the per-iteration cost from $\mathcal{O}(N^2)$ to $\mathcal{O}(N \log N)$. The proposed greedy
algorithm with alternating updates guarantees the four-peak distribution required for stable 2-bit clustering of each factor and admits closed-form initialization of cluster centers, removing the multi-restart k-means bottleneck of prior work. Composed with OPTQ-style sequential error compensation and QuIP-style incoherence preprocessing, the resulting JAX pipeline is competitive with OPTQ, QuIP, QuIP-RG and a fine-tuning- and vector-quantization-free variant of QuIP# at 4-bit per channel on OPT, Llama-2 and Pythia, with favorable wall-clock scaling. The bounded-$\ell_\infty$ factorization is also notably robust: on stress configurations where QuIP variants diverge to four-digit perplexity (Pythia-6.9B) or abort with NaNs in LDL back-substitution (Mistral-7B), Kashin-DCT remains numerically stable and stays close to FP16 baseline. At inference time, each weight decomposes into two 2-bit factor codes per channel that are structurally suited to native-2-bit hardware.
Read the full abstract ▾
All authors Daria Cherniuk, Alexander Rudikov, Boris Kashin, Ivan Oseledets
Artificial Intelligence
COBRA-Skills: Contextual Bandit-Guided Evolution for Agent Skill Optimization
Pingchen Lu, Xiangyi Wang, Xiang Li et al. (+6 more)
Large
language model (LLM)
agents can benefit from reusable skills distilled from prior task experience, yet existing skill optimization methods often rely on costly execution-based evaluation and substantial task data. We introduce COBRA-Skills, an efficient framework that formulates skill optimization as budgeted sequential optimization over a dynamically evolving candidate space. COBRA-Skills couples contextual-bandit-guided prioritization with evidence-grounded skill
evolution, selectively allocating evaluations to promising or informative candidates while continually refining the skill population from execution feedback. Across six heterogeneous
agent benchmarks and three target models, COBRA-Skills consistently achieves the strongest average performance among compared methods, while reducing optimization cost by 55–58% relative to SkillOpt and using only 50 unique optimization examples per
benchmark. Further analyses show that COBRA-Skills remains robust to changes in the
agent harness and performs effectively when the target model itself is used for skill generation and refinement.
Read the full abstract ▾
All authors Pingchen Lu, Xiangyi Wang, Xiang Li, Jie Mao, Zikun Qu, Junfeng Luo, Yao Shu, Bryan Kian Hsiang Low, Zhongxiang Dai