Information Theory
A simple proof of the group-theoretic Zhang-Yeung inequality
Harold Nieuwboer and Lubashan Pathirana
Zhang and Yeung (IEEE Trans. Inf. Theory, 1998) established the first non-Shannon-type inequality that holds for all entropic vectors. Chan and Yeung (IEEE Trans. Inf. Theory, 2002) showed that there is a one-to-one relation between linear
entropy inequalities and multiplicative inequalities involving cardinalities of subgroups of a finite group. The Shannon inequality admits a simple proof in the group theoretic setting, but a direct proof of the translation of Zhang–Yeung's inequality to the group-theoretic setting remained elusive. We resolve this open problem, giving a direct proof of the group-theoretic Zhang–Yeung inequality for cardinalities of subgroups of finite groups, using elementary group-theoretic and counting arguments.
Read the full abstract ▾
Robotics
Uni-VLaT: Whole-Body Tactile Adaptation of VLA Policies for Humanoid Loco-Manipulation
Zihao Wang, Shutong Liu, Siqi Zheng et al. (+5 more)
Physical contact often determines how a
humanoid should respond during loco-manipulation, yet
vision and proprioception alone are often insufficient to characterize physical interaction, especially when the contact region is occluded. Unlike sparse force or torque measurements at predefined regions, distributed tactile sensing preserves spatially resolved contact patterns across the
robot body. We therefore study how to integrate such whole-body tactile information into
vision-
language-action (VLA) policies for contact-rich control. Our approach, Uni-VLaT, introduces a tactile pathway whose latent state is trained not only for action generation, but also to predict future tactile, proprioceptive, and visual representations. This predictive objective builds a tactile-anchored
multimodal context, encouraging a more structured understanding of the physical world. We evaluate Uni-VLaT on five real-
robot tasks covering tactile-triggered locomotion, sustained physical interaction, human-
robot contact, and loco-manipulation. Uni-VLaT achieves a 75% average success rate, outperforming a baseline without tactile input by 43 points and a tactile-input baseline without predictive supervision by 7 points. Across two pretrained VLA backbones, our method improves Table Sweeping by 30 points on both backbones and Back-Tap Walking by 85-90 points. Ablations further show that contextualized tactile prediction and absolute future targets are critical to performance. These results indicate that predictive tactile learning provides an effective route for extending pretrained VLA policies to whole-body physical interaction.
Read the full abstract ▾
All authors Zihao Wang, Shutong Liu, Siqi Zheng, Liu Cao, Ruoqu Chen, Rundong Liu, Yanchao Yang, Mengdi Xu
Computer Vision
From internal representations to model improvement through prediction errors
Yushi Nakaya, Kenichi Higuchi and Shuichi Ishida
With limited annotation budgets, choosing which images to label determines how much a model improves. Data-selection methods that use features from a separately trained model, or scene descriptions written by
vision-
language models, have been successful, but those signals do not directly capture changes in the model being improved. The target model's own internal features reflect what it has learned so far and change with retraining, making them a natural cue for choosing the next training data. However, feature rarity alone does not reveal the errors that matter for performance. Here we link internal features to prediction errors and their expected impact on performance and select images for labeling and retraining without using labels for candidate images. We evaluated the method with an object detector on two
datasets and two pairs of random seeds. Adding internal features improved the identification of prediction errors in 15 of 16 conditions. When performance was averaged over successive labeling rounds, the method outperformed selection based only on feature rarity in all four evaluation settings and ranked among the top two of six methods. With other conditions held fixed, performance after retraining was again higher than with rarity-based selection, even though the latter collected more errors. With longer retraining, the proposed method ranked first among six methods. These results suggest that linking a model's internal features to its errors and their effects on performance may help select training images that improve performance, thereby allowing the model's current state to guide which images are labeled next.
Read the full abstract ▾
Computers and Society · Cross-listed from econ.GN
AI-based matching improves refugee employment in a double-blind randomized trial
Kirk Bansak, Jens Hainmueller, Dominik Hangartner et al. (+8 more)
Refugee integration is a central policy challenge for host countries, and where governments initially place refugees shapes their integration trajectories. Yet placement officers often have limited information about where each case is most likely to succeed. Algorithmic refugee matching uses administrative data, machine learning, and constrained optimization to recommend employment-optimized placements in real time as cases arrive, with human placement officers retaining final authority. Between January 2020 and June 2023, the Swiss State Secretariat for Migration randomly assigned about 2,000 refugee cases to receive a canton recommendation either algorithmically optimized for employment or drawn to approximate existing procedures, with placement officers and refugees blinded to assignment. The two arms used identical but separate canton and origin-group quotas, so gains reflect better refugee-canton matching rather than reallocation toward stronger labor
markets. The trial began just before the COVID-19 pandemic shifted labor-
market conditions. For the pre-registered primary outcome – the share of months employed during the first three years – the pooled intention-to-treat (ITT) estimate across the 2020-2023 placement cohorts was +2.2 percentage points (about 10% of the 22.3% control mean; 95% CI [+0.05, +4.33]), rising to +3.9 pp (about 17%; [+1.11, +6.68]) for the post-COVID 2022-2023 cohorts. Effects grew over time: at 36 months, the pooled ITT on the employment rate was +5.2 pp (about 11%; 95% CI [+1.10, +9.25]) – comparable to the gains from hundreds of hours of intensive
language training. Overall, the results provide rare field evidence that AI-based decision support can improve high-stakes public-sector allocation, offering a scalable, low-cost way to raise refugee employment.
Read the full abstract ▾
All authors Kirk Bansak, Jens Hainmueller, Dominik Hangartner, Jeremy Ferwerda, Elisabeth Paulson, Angie Delevoye, Nicholas Adams-Cohen, Ashwin Ramaswami, Selina Kurer, Joelle Pianzola, Michael Hotard
Machine Learning
NeuronSifter: An Agent-in-Twin for Prioritizing Central Nervous System Interventions with World Action Models
Haowei Xu, Wanyi Fu, Hongbin Han et al. (+1 more)
Effective central nervous system (CNS) intervention requires more than predicting biological responses: a scientific system must determine which intervention, observation, or additional evidence can improve a decision under uncertainty. However, in partially observed neuronal microenvironments, identical predicted outcomes may correspond to different exposure states, biological mechanisms, and intervention risks. We propose NeuronSifter, a decision-centric
Agent-in-Twin framework, where an AI
agent operates with an evolving biological digital twin rather than a fixed predictive model. NeuronSifter compiles interventions into state-conditional target-occupancy representations, propagates them through a mechanism-grounded World Action Model (WAM), and selects informative measurements based on their expected reduction in decision uncertainty. A shared probabilistic belief state couples intervention simulation, observation selection, and outcome assimilation, enabling the digital twin to continuously refine its decision-relevant representations through externally verified evidence. We evaluate NeuronSifter in Alzheimer's disease intervention scenarios using mechanistic neuronal microenvironment simulations and retrospective clinical endpoint references. The evaluation framework examines adaptive evidence acquisition, uncertainty-aware intervention planning, and robustness under model discrepancy through controlled environments and independent comparisons. NeuronSifter demonstrates how an AI system can actively identify missing evidence, refine biological hypotheses, and optimize intervention decisions under limited experimental budgets. Code is available at https://anonymous.4open.science/r/NeuronSifter.
Read the full abstract ▾
All authors Haowei Xu, Wanyi Fu, Hongbin Han, Zhaoheng Xie
Computer Vision
DiMoP: Diffusion-Driven Motion Representation Learning With Frame-Level Pseudo-Classification for Skeleton-Based Action Recognition
Shanaka Ramesh Gunasekara, Wanqing Li, Nikalal Kaldera et al. (+2 more)
Robust skeleton-based action recognition requires representations that capture a wide spectrum of motions, from subtle to moderate and strong ones. Existing methods often focus on strong motions. This paper introduces DiMoP, a masking- and
diffusion-driven motion representation learning method with frame-level pseudo-classification to explicitly learn the distribution of joint motions rather than regressing deterministic coordinates, as existing methods often do. By diffusing masked joints with progressive noise and denoising them conditioned on visible joints, DiMoP learns through controllable noising and denoising processes, enabling uniform learning of weak, moderate, and strong dynamics. To enable the masking-based generative
diffusion learning with a discriminative capability, a pseudo-frame classifier is proposed that enforces the learning towards sequence-consistent and temporally coherent pseudo-labels without manual annotations. Together, these strategies provide a principled mechanism for joint generative and discriminative motion modeling. DiMoP achieves state-of-the-art performance across NTU RGB+D 60/120, and PKUMMD, including a 1.1 percentage point gain over prior works on NTU RGB+D 120 with the cross-subject protocol.
Read the full abstract ▾
All authors Shanaka Ramesh Gunasekara, Wanqing Li, Nikalal Kaldera, Philip Ogunbona, Jack Yang
Artificial Intelligence
Just Initialize: A Training-Free Initialization Component for Large-Scale Routing Optimization
Jiale Zhao, Sirui Mao, Zimu Chen et al. (+5 more)
Large-scale routing problems are difficult to solve efficiently as their search spaces grow rapidly with problem size. Existing approaches primarily improve the optimization procedure itself, often at increasing computational cost. We instead shift the focus to a useful initialization that can be refined into a high-quality solution with limited downstream refinement. We propose Just Initialize, a training-free and solver-agnostic initialization component for large-scale routing optimization. Just Initialize compresses a large routing instance into a compact surrogate space, optimizes its global routing structure, and recovers the resulting solution as an optimization-friendly starting point in the original space. Extensive experiments on Traveling Salesman Problems (TSPs), Capacitated Vehicle Routing Problems (CVRPs), Vehicle Routing Problems with Time Windows (VRPTWs), and Prize-Collecting Traveling Salesman Problems (PCTSPs) demonstrate that Just Initialize achieves high-quality solutions comparable to or better than state-of-the-art methods while substantially reducing computational cost across instances ranging from 1K to 100K nodes, including an average speedup of approximately 70$\times$, sub-second runtimes on 10K-node instances, and runtimes within tens of seconds on 100K-node instances.
Read the full abstract ▾
All authors Jiale Zhao, Sirui Mao, Zimu Chen, Wentao Yang, Zihan Wang, Xuefeng Huang, Junji Cheng, Liyuanjun Lai
Machine Learning
Riccati State Space Models: Non-iterative Parallelization for Nonlinear Sequence Modeling
Mónika Farsang, Ramin Hasani, Daniela Rus et al. (+1 more)
State space models (SSMs) achieve efficient sequence processing because their affine state updates are closed under composition and can therefore be evaluated with an associative parallel scan. Nonlinear recurrent models can provide richer, state-dependent dynamics, but generally lose this compositional structure: parallel evaluation then requires iterative methods that repeatedly linearize and scan the recurrence. We ask, what state-dependent nonlinear dynamics can be designed to remain exactly composable? We answer by introducing RiccatiSSM, a nonlinear SSM, in which each state dimension follows an input-conditioned Riccati differential equation. Its quadratic state dependence makes the local Jacobian explicitly state-dependent, while its exact per-step flow under piecewise-constant inputs is a Möbius transformation. Since Möbius maps are closed under composition and compose through $2\times 2$
matrix multiplication, the complete nonlinear state trajectory can be evaluated exactly with a single associative parallel scan, without iterative linearization. We further derive a constrained parameterization that ensures bounded, contractive dynamics, and avoids poles in the fractional-linear state update. Across long-sequence classification, regression, and
forecasting tasks, RiccatiSSM achieves competitive predictive performance while reducing runtime by $22{-}33\%$ compared to the nonlinear LrcSSM under matched architectures. These results demonstrate that state-dependent nonlinear dynamics can retain exact composability and be evaluated efficiently within a single parallel scan.
Read the full abstract ▾
All authors Mónika Farsang, Ramin Hasani, Daniela Rus, Radu Grosu
Machine Learning
SOLO: Pretraining Billion-Parameter Language Models with Shared-Output Local Learning
Bojian Yin, Shurong Wang, Yuqi Pan et al. (+1 more)
Large
language models are trained with backpropagation, whose global gradient coordinates all layers but forces each to hold its activations and wait for the gradient to pass back through every deeper layer. Conventional local learning removes this update locking by training each module to predict the target through its own readout, but has not scaled to billion-parameter pretraining. We identify these private readouts as a key weakness, since they leave each module without information from deeper modules. We propose Shared-Output LOcal learning (SOLO), which replaces them with a shared, read-only copy of the final module's readout, the only one trained on the output of the whole network. Taken from the previous step, the copy transmits information from the final module without passing gradients between modules or reintroducing update locking. SOLO approaches backpropagation on
Transformers of 340M to 2B parameters pretrained on 15B
tokens, staying within one point in average zero-shot accuracy with a perplexity gap that narrows with scale. Readout ablations attribute SOLO's improvement over private readouts to sharing. Without update locking, each of p pipeline stages holds activations for O(1) micro-batches instead of O(p). The freed
memory permits larger micro-batches, which reach up to 1.44x the best measured throughput of pipeline backpropagation on the same partition. To our knowledge, SOLO is the first local learning method to show such
memory and throughput gains in billion-parameter
language-model pretraining. Local learning thus becomes a practical alternative to backpropagation for large-scale pretraining.
Read the full abstract ▾
All authors Bojian Yin, Shurong Wang, Yuqi Pan, Guoqi Li
Robotics
Revision, Not Restart: Revisable Visual Plans for Closed-Loop World-Action Models
Pengyiang Liu, Junbo Niu, Wenhao Zheng et al. (+5 more)
World-action models use predicted visual futures to condition
robot actions, yet execution feedback can invalidate parts of a prediction while leaving its task structure useful. We propose Revisable Temporal Planning (RTP), which maintains the visual future as a persistent action condition and revises it after feedback. Its central mechanism is a learned revision bridge: it resumes an intermediate state saved during visual generation and adapts its continuation to current observations. Visual and action supervision connect this revision to subsequent control. Time-aware history supplies observed evidence, and an adaptive policy selects retention, bridge revision, or fresh replanning from new noise before decoding the next action. On RoboMME and RMBench, RTP achieves task-averaged success rates of 48.6% and 84.8%, respectively. Matched comparisons support learned continuation; estimated checkpoint-source and action-prefix effects are positive but less precisely resolved. These results connect feedback-driven visual-plan revision to closed-loop task performance. Project Page: https://PLACEHOLDER.github.io/RTP/
Read the full abstract ▾
All authors Pengyiang Liu, Junbo Niu, Wenhao Zheng, Xinchen Chen, Canyu Li, Zhongyue Shi, Jiahao Xie, Si Liu
Computer Vision · Cross-listed from q-bio.QM
An integrated geometric quantification and shape analysis framework for axillary lymph node metastasis in breast cancer patients
Zixi Yi, Limeng Qu and Gary P. T. Choi
Quantitative characterization of lymph node morphology is important for assessing axillary lymph node metastasis in breast cancer. However, surfaces reconstructed from computed
tomography (CT)
segmentation may contain
geometric and topological defects that compromise subsequent analysis, while conventional shape descriptors predominantly characterize global morphology. To address these issues, we developed an integrated framework combining
topology-aware surface processing with multi-resolution spherical harmonic (SH) analysis of CT-derived axillary lymph nodes. The processing pipeline produced
topology-valid genus-0 surfaces with improved mesh quality, which were then represented at multiple SH degrees and characterized using 20 predefined
geometric feature families.
Geometric fidelity increased with SH degree, whereas predictive performance peaked at intermediate resolutions. Preferred SH degree also differed across feature families. A family-specific mixed-resolution model achieved an AUC of 0.918, compared with 0.884 for the conventional PyRadiomics Shape14 baseline, corresponding to an improvement of 0.0344. Controlled perturbation experiments showed that higher SH degrees transmitted more fine-scale
geometric variation and yielded lower stability of
curvature-based predictions. Representative
geometric descriptors provided interpretable characterization of metastasis-associated surface morphology. Independent validation further supported the framework's transportability: label-free replication in a multicenter lymph node cohort reproduced the family-specific resolution effects, while a labeled LIDC-IDRI lung-nodule experiment reproduced the resolution-dependent relationship between SH degree and predictive performance. Altogether, the framework provides a
topology-valid basis for quantitative characterization of lymph node morphology and metastasis-associated imaging phenotypes.
Read the full abstract ▾
Artificial Intelligence
Building Transformation Layers for Riemannian Neural Networks
Ziheng Chen
Recently, deep neural networks on
manifold-valued representations have garnered significant
attention across various machine learning applications. One recent focus is the generalization of Euclidean fully connected (FC) and convolutional layers to non-Euclidean geometries. However, previous approaches typically focus on a few selected
manifolds and rely on specific properties of the target
manifold. In contrast, this work proposes a framework for constructing FC and convolutional layers over computationally tractable Riemannian spaces. This framework incorporates several previous FC layers across different geometries as special cases and is instantiated on ten representative
manifolds, including three
hyperbolic models, five geometries of the symmetric positive definite (SPD)
manifold, and two Grassmannian perspectives. Experiments on different
manifolds demonstrate the effectiveness and applicability of our approach. Code can be found at https://github.com/GitZH-Chen/RieTrans.
Read the full abstract ▾
Cryptography and Security
LLM-Assisted Automatic Security Proofs for Cryptographic Protocols: How Far Are We?
Tianjian Liu, Shicheng Feng, Jin'ao Shang et al. (+5 more)
Large
language models (LLMs) have shown strong potential for assisting software and security analysis tasks, yet their effectiveness in cryptographic symbolic protocol verification remains insufficiently understood. In this paper, we conduct the first systematic evaluation of the capability of state-of-the-art LLMs in cryptographic symbolic protocol verification. To quantify this capability, we propose CRoST (Coverage Rate of Solve Tree), a proof-based metric derived from the verifier's proof skeleton that measures the similarity between generated lemmas and reference lemmas. We then establish the rationale of CRoST through both theoretical analysis and empirical validation. The evaluation results show that state-of-the-art models achieve 38.82% coverage on average, with 14.4% of generated lemmas exceeding 80% coverage, indicating that LLMs can already generate useful lemmas to a certain extent. However, they still exhibit non-trivial failure modes on complex multi-phase protocols, show diminishing returns under naive scaling, and incur substantial verification overhead. These findings clarify the practical potential and limitations of LLMs for protocol verification and motivate future work on complex real-world protocols.
Read the full abstract ▾
All authors Tianjian Liu, Shicheng Feng, Jin'ao Shang, Xiaoting Lyu, Bin Wang, Zonghua Zhang, Lei Xue, Wei Wang
Machine Learning
ReSPO: Reshaped Sequence Policy Optimization for Gradient Starvation in Off-Policy Learning
Yihang Chen, Yuanhao Ban and Cho-Jui Hsieh
Reinforcement learning from verifiable rewards (RLVR) frequently reuses rollouts across multiple policy updates, increasing the mismatch between the current policy and the data-generating policy. We identify a sign-dependent gradient starvation problem in clipped policy optimization: clipping suppresses under-generated positive responses at the low-importance-weight tail while permitting severely over-generated negative responses to dominate the high-weight tail. To address this, we propose ReSPO (Reshaped Sequence Policy Optimization), which replaces clipping with a smooth, two-branch sequence-level kernel derived from an $\alpha$-divergence variational objective and an exponential variance-control tilt. The positive branch preserves a nonzero gradient weight for under-generated positive responses, while the negative branch suppresses heavily over-generated negative responses. We demonstrate that ReSPO effectively learns from long positive
reasoning trajectories during early training, even when accumulated policy drift relegates them to the low-importance-weight tail. On dense and MoE Qwen3 models, ReSPO accelerates early optimization, improves final training scores, and achieves higher held-out
benchmark performance under a rollout reuse, validating our approach on importance-weight tail control in off-policy learning.
Read the full abstract ▾
Robotics
Self-Evolving Coding Agents: From Digital Programs to Physical-World Intelligence
Hongcheng Gao, Jingjing Zhou, Zelin Zheng et al. (+10 more)
Vision-
language-action (VLA) and world-action (WAM) models map observations and instructions directly to
robot actions. This directness ties a policy to training: minor layout or viewpoint changes cause failure, and instructions generalize poorly. The root cause lies in representation: task requirements, conditions, progress, and failure recovery are implicitly encoded in action sequences, making them difficult to inspect or revise. Digital coding
agents offer a precedent: LLMs call tools, verify results, and revise from feedback as executable code. The same working pattern of explicit state, manageable execution, and revisable procedures underlies generalization and long-horizon execution in the physical world, letting physical experience return as reusable programs,
memory, or evidence. We propose Physical Coding, representing task state and execution as code. Code as World records objects, relations, constraints, and progress; Code as Policy organizes planning, verification, recovery, and execution. We build HexaAnything, which calls perception, planning, and control tools, including VLA/WAM policies, and makes in-the-loop decisions from external feedback. Verified traces become data and
memory, enabling
evolution from tools and Harness to model weights, architectures, and ultimately hardware and task design. On RoboCasa365, HexaAnything improves Composite-Unseen and overall success over XR-1 VLA, and its Harness-trained HexaModel beats the base on every split, indicating code traces internalize physical execution. On PhyBench and a dual-arm AgileX
robot, the
agent autonomously completes physics experiments and most tabletop tasks, often faster than published results. We observe data, model, and tool self-
evolution; future work targets weight internalization, autonomous redesign of architectures,
languages, representations, and tasks, and deployment in manufacturing and science.
Read the full abstract ▾
All authors Hongcheng Gao, Jingjing Zhou, Zelin Zheng, Shijia Ge, Jay Zhu, Yazhe Wang, Jianshu Zeng, Xuan Shangguan, Di Wu, Lingyu He, Zhiqi Jia, Sihang Wu, Xiao He
Robotics
Memory in the Sky: Low-Altitude Question Answering with Multi-Agent Memory Aggregation
Chengyang Li, Yujie Wan, Shuai Wang et al. (+6 more)
This paper studies low-altitude question answering (LAQA), in which distributed unmanned aerial vehicle (
UAV)
memories are aggregated at a ground server to answer questions about observations over a long horizon. Unlike conventional resource allocation based on sensing, communication, control, or computation metrics, LAQA requires an explicit measure of
memory value. We propose a generative
adversarial exam (GAE) that uses forward simulation to evaluate
memory retrieval and exam scores to quantify
memory quality. This enables the downstream QA value of candidate
memories to be measured and optimized without accessing the internal mechanisms of the black-box captioning,
retrieval, and
reasoning pipeline. Building on this metric, we develop a
memory-centric (MemCen) framework that jointly selects
UAVs and allocates transmit power to maximize
memory quality under communication constraints. In the noise-limited regime, we derive a QoM-aware capped water-filling law that explicitly connects task utility with physical-layer power allocation. We further develop penalty successive optimization (PSO) and learning to memorize (L2M) solvers. MemCen achieves QA accuracies of 92.4% and 84.0% in CARLA Town04 and Town05 under static and dynamic communication conditions, respectively. In real-world experiments, MemCen achieves 88.5% QA accuracy on the panoramic multi-
agent system (PMAS)
benchmark. Finally,
UAV-to-
robot-dog demonstrations further validate the practical utility of the acquired
memories for environmental understanding and
navigation.
Read the full abstract ▾
All authors Chengyang Li, Yujie Wan, Shuai Wang, Kejiang Ye, Weijie Yuan, Boyu Zhou, Yik-Chung Wu, Chengzhong Xu, Huseyin Arslan
Information Retrieval
Can Generative Retrievers Learn Semantic IDs Without Forgetting How to Speak?
Junchen Fu, Kleomenis Katevas, Vandana Rajan et al. (+2 more)
Generative
retrieval (GR) enables end-to-end
retrieval by generating document semantic identifiers (SIDs). However,
retrieval-only fine-tuning can over-specialize pretrained
language models to SID prediction, substantially distorting their natural-
language distribution and limiting their suitability for interactive systems that must both retrieve documents and generate natural-
language responses. We introduce SpeakGR, a dual-objective framework that learns SIDs while preserving
language generation. It combines supervised SID learning with speak-preserving regularization: an on-policy
distillation objective that aligns the current model with a frozen copy of the original model on student-generated prefixes using forward KL over the original text vocabulary. We further propose Adaptive SpeakGR, which dynamically adjusts the preservation strength based on observed
language drift. Compared with SFT-only, SpeakGR reduces WikiText-2 forward KL by 81.3-93.8% on MS MARCO and 81.2-85.2% on Natural Questions (NQ) while retaining effective
retrieval across three different LLMs. Adaptive SpeakGR further improves
retrieval over SpeakGR in most settings while maintaining substantially lower
language drift than SFT-only.
Read the full abstract ▾
All authors Junchen Fu, Kleomenis Katevas, Vandana Rajan, Sofía Celi, Hamed Haddadi
Machine Learning · Cross-listed from stat.ML
Multi-Task Learning of Conditional Mean Operators: applications to dynamical systems and uncertainty quantification
Sami Chemlal, Thibaut Germain, Rémi Flamary et al. (+2 more)
Estimating conditional statistics and learning representations of a population of conditional distributions are central problems in many data-driven applications, including uncertainty quantification and dynamical systems analysis. Conditional mean operators (CMOs), a class of linear operators between function spaces, resolve these objectives by providing access to a broad class of conditional statistics. However, existing methods typically estimate each CMO independently or constrain it to prespecified function spaces, thereby preventing the exploitation of shared structure across related distributions. In this work, we posit that related CMOs share finite-dimensional input and output function spaces, and are specialized for each task with a linear operator mapping these spaces. Based on this hypothesis, we introduce MTL-CMO, a multi-task framework that jointly learns shared function spaces and task-specific operators across multiple
datasets. We further introduce T-CMO, a transfer learning method that reuses the shared spaces to estimate, in closed form, the operator of a new conditional distribution. We establish statistical guarantees quantifying the benefits of jointly learning the shared function spaces. Our experiments demonstrate that learning shared function spaces improves uncertainty quantification across a broad range of conditional distributions and, when applied to Langevin and
plasma dynamics, yields compact representations of complex dynamics that retain physically meaningful information and enable parameter identification.
Read the full abstract ▾
All authors Sami Chemlal, Thibaut Germain, Rémi Flamary, Vladimir R. Kostic, Karim Lounici
Machine Learning
LLMs are General Asynchronous Agents
George Yakushev, Denis Mazur, Vladimir Bartenev et al. (+4 more)
Modern LLMs are increasingly capable as autonomous
agents, but they follow sequential interaction cycles: read, think, reply or call tools, repeat. Many real-world use cases are not sequential: voice assistants,
embodied agents, and monitoring systems receive new inputs while they think or perform another task. Modern LLMs address this with specialized architectures for voice interaction and
video streams, VLAs for
robot control, asynchronous tool calling for API usage, and others. In this work, we generalize from different asynchronous tasks to general asynchronous
agents that can adapt to different types of concurrency. To achieve this, we develop an asynchronous LLM framework that lets users (or the
agents themselves) define inference coroutines with overlapping
memory states. We showcase that Qwen 3.x models are capable of asynchronous operation for streaming
video understanding, videogames, and monitoring, without task-specific training.
Read the full abstract ▾
All authors George Yakushev, Denis Mazur, Vladimir Bartenev, Vyacheslav Zhdanovskiy, Timofey Byzov, Vladimir Kaurkin, Vadim Pastushenko
Machine Learning
Frontier Learning: Training LLM Reasoners at the Edge of Capability
Robin Faro, Shyam Sundhar Ramesh, Ilija Bogunovic et al. (+1 more)
Reinforcement Learning-based post-training of Large
Language Models (LLM) has been successfully applied to improve their
reasoning capabilities. Existing pipelines primarily finetune LLMs on a fixed pool of problems specified prior to training using the GRPO loss. This is fundamentally limiting, as learning signal arises only when policy rollouts mix successes and failures, causing the useful portion of any fixed pool to quickly become stale as the model improves. To address this, we propose frontier learning, an open-ended post-training approach in which procedural generators are used online to continually produce informative training problems. It treats the generator's task-specific parameters as a search space and uses a regret signal to prioritize and explore frontier difficulty levels in order to focus training at the edge of the model's evolving
reasoning capabilities. Across several
reasoning tasks and model families, our approach consistently achieves higher relative gains over fixed-pool baselines, demonstrating that effective post-training requires not only selecting useful problems, but continually generating them at the edge of capability.
Read the full abstract ▾
All authors Robin Faro, Shyam Sundhar Ramesh, Ilija Bogunovic, Aurelien Lucchi
Programming Languages
Semantic Prefix Oracles for LLM Decoding: Contracts and Differential Validation
Paul Kronlund-Drouault
Constrained decoding can enforce regular or context-free output formats, but many program-generation failures are semantic: scope, typing, and declaration effects depend on context. We present semantic grammar specifications, a declarative formalism that attaches such constraints to a context-free surface and executes them during Earley descent. Our implementation enforces safe
pruning: it rejects only prefixes whose semantic contradictions cannot be repaired by any continuation. A separate, grammar-dependent, dead-end freedom property guarantees the existence of a realizable witness for each remaining branch. We give simple sufficient conditions based on surface productivity, type coverage, and left-to-right constraint flow. Our finite-lambda, core ML, and C-like fragments satisfy them, while the STLC instance used in our experiments does not: plain STLC can violate type coverage, and we show how restricting its type universe recovers it. A tokenizer-lifting lemma carries character-level witnesses to
token sequences under an explicit vocabulary-coverage hypothesis. We validate the implementation differentially against production compilers (ocamlc, cc). Across every prefix of 65 compiler-valid programs we observe zero false prunes. The semantic oracle localizes 25/30 invalid programs mid-stream, against 0/30 for a syntax-only oracle, and agrees on 42/42 recursion probes. A twelve-model generation study, including a matched semantic-versus-syntactic ablation for nine models, finds nonnegative observed semantic-minus-syntactic point estimates for every model-
language pair, with maxima of $+15.2$ points on STLC task correctness and $+14.3$ points on ML validity.
Read the full abstract ▾
Systems and Control
Scalable Small-Signal Stability Assessment of Power Systems Based on Frequency-Domain Quadratic Constraints
Xiaoying Liu, Linbin Huang, Hangyu Chen et al. (+1 more)
Small-signal stability analysis of power electronics-dominated power systems is becoming increasingly challenging due to the large-scale integration of heterogeneous grid-following (GFL) and grid-forming (GFM) converters. Classical centralized methods, such as
eigenvalue analysis and the generalized Nyquist criterion, provide direct stability assessment tools but exhibit limited scalability as the system size increases. This motivates scalable analysis methods for stability assessment and device-level diagnosis, and calls for less conservative stability conditions. To this end, this paper proposes a scalable small-signal stability assessment method with quantitative stability indices for multi-converter systems based on frequency-domain quadratic constraints (FQCs). It is shown that several existing stability conditions, including
geometric conditions based on the Davis-Wielandt (DW) shell, numerical range, $x$-$z$ graph, scaled relative graph (SRG), mixed gain-phase condition, and passivity condition, are special cases of the FQC-based stability condition. This FQC-based formulation also clarifies the requirements for decentralized verification of these existing stability conditions. To further reduce conservatism, a full-multiplier FQC-based stability condition is developed, based on which an optimization problem is formulated for scalable stability certification and device-level diagnosis using quantitative stability indices without relying on graphical inspection. Combined with the mixed gain-phase condition, this optimization problem forms a hierarchical screening-diagnosis procedure that improves efficiency. Case studies demonstrate that the proposed method is less conservative than existing criteria and can effectively identify problematic devices associated with potential instability risks.
Read the full abstract ▾
All authors Xiaoying Liu, Linbin Huang, Hangyu Chen, Huanhai Xin
Cryptography and Security
Indistinguishability of Sum of Permutations: A Fourier Analytic Route to Classical and Quantum Security
Ritam Bhaumik, Chun Guo, Xiaoning Guo et al. (+1 more)
We study classical and
quantum indistinguishability of sums of independent random permutations and related transformations from permutations to functions. Let $G$ be a finite abelian group of order $N$, and let $\pi^k_+(x)=\pi_1(x)+\cdots+\pi_k(x)$ for $k\geq2$ independent uniform random permutations of $G$. We give a unified
Fourier analytic treatment in which the construction is represented by its probability density and a distinguisher by its acceptance function, with the classical and
quantum query models imposing different restrictions on the
Fourier support of the latter. Classically, we obtain the bound $O_k(q/N^{k-1/2})$ for every $q<N$, and refine it below the birthday threshold to $O_k(q^2/N^k)$. In the
quantum model, a simulation argument gives $O_k(N^{-(k-3/2)})$ for $q\leq(N-1)/2$, while
Fourier interpolation gives concrete finite bounds up to $q\leq4N/15$ and the query-dependent bounds $O\left(\min\left\{N^{-1/2},q^3/N^2 + 1/N\right\}\right)$ and $O_k\left(\min\left\{q^3/N^k,N^{-(k-3/2)}\right\}\right)$, for $k=2$ and $k \geq 3$, respectively, throughout $1\leq q\leq(N-1)/2$. For $q = 1$, the first bound sharpens to $O(N^{-2})$. Over $G=\mathbb F_2^n$, a one-query
Fourier attack matches the order of our one-query bound, while an $N/2$-query parity attack with advantage $1/2$ shows that our bounds reach the constant-advantage query threshold. We further study two variants of sum of permutations over binary vector spaces. First, we allow arbitrary surjective linear postprocessing, which includes truncation, and obtain classical and
quantum bounds that retain the output-size dependence. Second, we analyse Dinur's variable-output single-permutation construction, $\mathsf{LXoP}$, for every fixed output width, and derive its classical and
quantum security bounds; for one- and two-block outputs, we give concrete
quantum security bounds.
Read the full abstract ▾
All authors Ritam Bhaumik, Chun Guo, Xiaoning Guo, Ashwin Jha
Computer Vision
Automated Species Identification in Camera Trap Images for Wildlife Conservation
Nowshin Amin, Nafisa Tabassum Oyshi, Tahmid Abrar Zidan et al. (+2 more)
Wildlife conservation involves protecting, preserving, and managing wildlife species and their habitats. With today's rapid pace of human development,
climate change, and other unsustainable practices, the need for wildlife conservation has heightened. Despite significant progress in species identification using deep-learning models, significant challenges still remain in effectively detecting small animals in low-contrast trap images due to limited feature extraction capabilities. This thesis presents a novel end-to-end framework integrating a self-
attention mechanism to address these limitations. The proposed architecture involves a Swin-BiFPN backbone integrated in a Faster RCNN detection network, coupled with a visual semantic extraction module driven by the LLaVA v1.5 (13B)
multimodal large
language model. The detection framework, capable of extracting crucial features in challenging trap images, demonstrates consistently high results and robust generalization capabilities. Furthermore, the visual semantic extraction module provides zero-shot detection capability, as well as providing valuable insights and emergent cues of the animal's behavior, further supporting the conservation effort. The MLLM evaluation was conducted using both traditional NLP metrics (precision, recall, F1, and SBERT similarity) and subjective scoring by LLM-based judges (GPT-4.1 and GROK 3.0), across five MLLMs, demonstrating the model's strong performance in visual description generation. The proposed framework improves detection accuracy across low-contrast trap images and small animals while also demonstrating zero-shot detection capability leveraging the MLLM.
Read the full abstract ▾
All authors Nowshin Amin, Nafisa Tabassum Oyshi, Tahmid Abrar Zidan, Miftaun Noor, Md. Abrar Rahman Shafin
Numerical Analysis · Cross-listed from math.AP
Convex counterexamples to the Schiffer and Pompeiu conjectures in dimensions two to eighteen
Jizhou Guo
We construct the first convex planar counterexample to the Schiffer and Pompeiu
conjectures: a bounded strictly convex non-disc domain with real-analytic boundary carrying a nonconstant solution of $\Delta u+u=0$ with $u=1$ and $\partial_\nu u=0$ on the boundary. Together with the higher-dimensional constructions, this gives convex non-ball counterexamples in every dimension from two to eighteen, and in dimensions twenty and twenty-one, with real-analytic boundaries diffeomorphic to spheres and indicator
Fourier transforms vanishing on the unit sphere. In dimensions $4$, $6$, $8$, $10$ and $14$, planar reductions through compact Lie
algebras of rank two use Harish-Chandra's radial part formula and Kostant's convexity
theorem. In dimensions $3$, $5$ and $7$, an axisymmetric formulation on the unit ball gives a quartic equation in weighted coefficient spaces. The exact two-sided inverse of the residual-corrected linearisation combines a Dirichlet Helmholtz inverse, a holomorphic boundary real-part problem and the material-derivative identity; it gives the planar Schiffer domain and, by one dimension-parametrised argument, the domains in dimensions $5$, $9$, $11$ to $13$, $15$ to $18$, $20$ and $21$. We also construct the first convex planar counterexample to Berenstein's
conjecture: a strictly convex non-disc domain with real-analytic boundary carrying a sign-changing solution of $\Delta u+u=0$ with $u=0$ and $\partial_\nu u=1$ on the boundary. Planar existence for this problem is due to Colbrook, Sadeghi and Stepaniants. All existence proofs in this paper are computer-assisted.
Read the full abstract ▾
Machine Learning · Cross-listed from math.OC
Convex Optimization Is Free When Accuracy Is Expensive
Arthur Paing and Arthur Jacot
This paper studies convex optimization when the gradient cannot be evaluated exactly, but only approximated by a hierarchy of
algorithms whose compute grows like $\delta^{-\gamma}$ in the accuracy $\delta$. When $\gamma>2$, falling into the Harder-Than-Monte-Carlo (HTMC) regime, the price of accuracy outruns the variance reduction that Monte Carlo would buy and we show that minimizing a loss function costs no more, up to a factor depending only on $\gamma$, than a single evaluation of its gradient at the accuracy the problem demands. A randomized multilevel oracle replaces the deterministic approximation of accuracy $\delta$ by an unbiased
estimator of it, whose variance $\sigma^2$ becomes a second, independently priced dial: the cost of one call drops from $\delta^{-\gamma}$ to $\delta^{2-\gamma}\sigma^{-2}$. Plain inexact gradient descent driven by that oracle reaches loss $\varepsilon$ at expected compute $\Theta(\varepsilon^{-\gamma})$ in the convex case, against $\Theta(\varepsilon^{-(\gamma+1)})$ for the same method run at a fixed accuracy: randomization buys a full power of $\varepsilon$. Under $\mu$-strong convexity the exponent halves, to $\varepsilon^{-\gamma/2}$, because the iterates settle at a noise floor and the bias budget relaxes accordingly. Both bounds are independent of the step size, and hence of the smoothness constant, and we show that the cost is a functional of the underlying gradient flow rather than of any discretization of it.
Read the full abstract ▾
Computers and Society · Cross-listed from physics.soc-ph
A citizen science behavioural dataset of an urban collective-risk dilemma on air pollution mitigation under resource inequality
Marc Sadurní, Martín F. Díaz, Julián Vicens et al. (+4 more)
This article describes a behavioural
dataset from a collective-risk dilemma public experiment carried out in Barcelona, Spain, within the citizen science xAire air quality project. Participants played a ten-round game in groups of six, each deciding every round how many monetary units to contribute from a private endowment toward a shared target; groups that fell short forfeited all remaining capital. Three endowment treatments held the group total fixed: an equal condition and two unequal conditions in which low-endowment participants formed either the minority or the majority of the group. Data were collected at two public outdoor venues using a tablet-based interface under lab-in-the-field protocols. Participants were recruited on-site from a demographically heterogeneous pool of individuals passing through the venues, mostly local residents. The
dataset comprises 462 participants across 77 six-player games, 4\,620 contribution decisions in total. Each participant record links the full ten-round contribution trajectory to initial endowment, wealth treatment, venue, district of residence and its measured nitrogen-dioxide concentration, and socio-demographic and attitudinal survey items covering gender, age, education, economic and working status, air-quality perception,
fairness norms, environmental-justice beliefs, residential mobility, and community participation. The release provides raw and cleaned participant tables, a variable codebook, the survey instrument, the district-pollution linkage, and the preprocessing code.
Read the full abstract ▾
All authors Marc Sadurní, Martín F. Díaz, Julián Vicens, Anna Cigarini, Isabelle Bonhoure, Miquel Montero, Josep Perelló
Computer Vision
When Should the Count Change? Learning State Maintenance for Causal Video Counting
Pengyiang Liu, Dongyue Lyu, Junbo Niu et al. (+3 more)
Continuous
video counting requires distinguishing new observations from new objects or completed events. We introduce StaMina (State Maintenance), which learns to maintain counting state through state-conditioned updates. Recurrent visual context supports recognition; learned transitions maintain visibility, persistent identities, and completed-event records. A differentiable recurrence trains event transitions over legal paths constrained by count endpoints; visibility and association objectives train the object branch. A multi-source pipeline organizes 39.8K spatial queries and complementary event annotations into counting trajectories. On SVCBench, we evaluate counting adaptation with partial
video overlap and held-out groups of linked annotations. Under prefix replay (Full) and persistent streaming (Stream), 4B and 8B models reach 41.9/36.4 and 44.9/38.2 Gaussian Precision Accuracy, respectively. The 8B model gains 10.9/3.2 points over Counting-SFT on the same queries. Matched-graph comparisons isolate phase conditioning and trajectory supervision, assessing training objectives alongside hard decisions. Online
video benchmarks and count-conditioned decisions assess online understanding and task eligibility. Project Page: https://PLACEHOLDER.github.io/StaMina/
Read the full abstract ▾
All authors Pengyiang Liu, Dongyue Lyu, Junbo Niu, Zhongyue Shi, Jiahao Xie, Si Liu
Robotics
Adaptive Safety Filtering for Frozen ACC Policies via Conformal Residual Calibration
Zhiruo Zhou, Rigaudiere Z. Li, Chen Xiwen et al. (+3 more)
Frozen adaptive cruise control (ACC) policies can violate constraints when deployment dynamics differ from their training conditions. We propose residual-aware conformal action filtering (RACF), which calibrates residuals of a fixed nominal predictor and converts their
quantile into an operating margin for finite-model action projection. Completed transitions update margins and candidate selection without retraining the policy. In a registered comparison over 2,400 controller-trial units, Adaptive RACF achieves 94.3% episode safety, improving by 19.9 percentage points over the evaluated nominal CBF-QP baseline while reducing projection frequency from 8.11% to 6.63%. A controlled study isolates a 4.54-point improvement from residual-margin injection. In a separate matched-hardware evaluation, Adaptive reduces mean amortized rollout time by 21.2% relative to Robust CBF-QP, with 161/180 versus 170/180 safe episodes. We characterize conditions linking one-step residual coverage to constraint satisfaction and quantify the observed safety-computation trade-offs.
Read the full abstract ▾
All authors Zhiruo Zhou, Rigaudiere Z. Li, Chen Xiwen, Yucheng Chen, Xiaojun Zhu, Houde Liu
Cryptography and Security
AI-Based Vulnerability Assessment Capability and Cyber Attack Graph Analysis
Joni Herttuainen, Kirsi Hellsten, Vesa Kuikka et al. (+4 more)
Cyber threats targeting mission-critical infrastructure are becoming more sophisticated while the barrier to launching attacks continues to fall. Traditional point solutions like antivirus and firewalls are reactive and fail to address the combinatorial complexity of modern attack surfaces. This paper presents an investigation combining two complementary methodologies: Lockheed Martin's
Vortex/Crow framework, which applies multi-
agent reinforcement learning (MARL) over industry-standard cyber knowledge graph to identify and prioritize attack vectors and TTPs (tactics, techniques, and procedures); and Aalto's probabilistic attack graph model that combines network
topology and its vulnerabilities to compute system-level risk metrics. The 2015 Ukraine Power Grid cyberattack serves as a well-documented validation scenario. Applied independently to the same operational technology (OT) network
topology, both methodologies converge on the same attack vectors and exploit sequences as those documented in the incident record, thus providing mutual cross-validation. Attack graph analyses using node-level elimination experiments identify industrial control systems (ICS) as the most critical enablers of attack propagation, representing high-priority targets for defensive hardening. Comparison of CVSS (v2.0) and IronMiner vulnerability scoring yields in general consistent results, with IronMiner providing more actionable differentiation at network periphery nodes. The layered methodology of baseline assessment and node-level elimination proves to be scalable to large enterprise networks, thus offering defenders a structured, AI-enabled path to prioritize mitigation under realistic time and resource constraints.
Read the full abstract ▾
All authors Joni Herttuainen, Kirsi Hellsten, Vesa Kuikka, Ambrose Kam, Arlanda Johnson, David Welsh, Kimmo K. Kaski