Machine Learning
Lucid Dreaming for World Models: Learning to Doubt Imagination and Decide by Trust
Ziqi Wen, Ting Xu, Lianyu Wang et al. (+5 more)
World models enable
agents to learn and plan in imagination, but predictions beyond their experience can become unreliable and mislead decisions. Existing uncertainty estimates derived from predictions can remain overconfident on unfamiliar state-action pairs. We propose the Lucid World Model (LucidWM), which learns doubt from experience and propagates trust through imagination. By integrating Subjective Logic into categorical latent transitions, LucidWM distinguishes predicted outcomes from their evidential support and assigns each transition a degree of doubt. The complement of this doubt defines transition-level trust, which accumulates multiplicatively along imagined trajectories to reweight returns for policy learning and guide action selection. Uncertainty estimation requires no additional parameters or forward passes. Evaluated on four base world models against seventeen uncertainty readouts, LucidWM detects environmental changes and signals uncertainty during action-corrupted rollouts. In a controlled
navigation case study, acting on trust reduces the number of steps required to reach the goal from 362 to 190. Fifteen demonstration
videos show how LucidWM doubts its dreams and acts on that doubt.
Videos are available at https://lucidwm.github.io.
Read the full abstract ▾
All authors Ziqi Wen, Ting Xu, Lianyu Wang, Xian Lin, Yanda Meng, Huazhu Fu, Meng Wang, Ching-Yu Cheng
Computational Engineering
Explainable and Trustworthy AI for Anti-Money Laundering: A Graph-based Hybrid Framework for Real-World Financial Crime Detection
Ali Shahbazi, Faraz Sasani, Arshia Hossein zadeh et al. (+2 more)
Purpose: Money laundering threatens financial systems, while rule-based monitoring suffers from high false-positive rates and limited ability to capture relational transaction patterns. This study proposes an explainable graph-based framework for anti-money laundering (AML) detection that jointly addresses predictive performance, explanation faithfulness, and uncertainty calibration. Methods: Using the IBM Transactions for Anti-Money Laundering (HI-Small)
benchmark, comprising 5,078,345 transactions among 518,573 accounts with 0.10% illicit transactions, a directed attributed graph with 4,487,133 edges was constructed using temporal, leakage-safe partitioning. Three GATv2 architectures, BASE, BASE-Large, and IMPROVED, incorporating bidirectional message passing, edge updates, and port-aware features, were compared with XGBoost and Random Forest using identical features. The best model was evaluated using GNNExplainer against documented typologies and Mondrian conformal prediction for uncertainty calibration. Results: IMPROVED achieved the strongest performance (AUPRC = 64.85%, Best F1 = 68.42%), exceeding BASE-Large by 30.65 AUPRC points and XGBoost (AUPRC = 38.46%) by 26.4 points. The proposed mechanisms contributed more than capacity scaling alone. GNNExplainer recovered documented typologies with higher fidelity than
attention-weight and random baselines (mean Jaccard overlap: 21.4% vs. 1.0%, p < 0.001). Mondrian conformal prediction achieved coverage close to the 90% nominal target with an average prediction-set size near one. Conclusion: Explicit transaction
topology modeling substantially improves AML detection, while faithful explanations and calibrated uncertainty support interpretable, human-in-the-loop compliance decision-making.
Read the full abstract ▾
All authors Ali Shahbazi, Faraz Sasani, Arshia Hossein zadeh, Sheyda safaeimoradi, Hossein Najafzadeh
Statistical Methodology
CovEsts: Nonparametric Autocovariance Estimation and Analysis
Adam Bilchouris and Andriy Olenko
The paper introduces the R package CovEsts, which implements several nonparametric
estimators for the autocovariance function. First, it presents the theoretical foundations of the implemented
estimators, their properties, assumptions, and potential limitations. Next, it outlines the structure of the package and its key functions, including several methods for estimating autocovariance functions, constructing corresponding bootstrap confidence regions, and correcting the provided
estimators. The package also includes diagnostic tools, such as several metrics for comparing
estimators, and additional functions for broader use. The article illustrates a high degree of flexibility of the package in the selection of function parameters and the tuning of the
estimators. Applications of selected
estimators and package functions are illustrated using simulated data, yearly sunspot counts, and US unemployment increments data.
Read the full abstract ▾
Artificial Intelligence
When Tools Silently Lie: Evaluating and Mitigating Blind Compliance in Tool-Augmented Data Agents
Zifu Tao and Changqing Yin
Tool-augmented data
agents rely on tool outputs for analytical decisions. Yet successful execution can return plausible but incorrect evidence, requiring
agents to decide whether to trust or verify it. Understanding this failure requires examining both the evidence obtained through checking and the answer ultimately adopted. We introduce ToxicBench to measure checking and adoption under numerical, label, schema, and
retrieval errors, pairing clean and poisoned observations over fixed source data. In the 118-task GPT evaluation across three adapters, poisoning lowers task success by 26 to 39 percentage points. Ordinary retries help under one-shot poisoning, whereas repeated poisoning reveals wrong-answer adoption after checking. Controls on three public tables isolate how supplied evidence affects recovery. After freezing the scorer, we compare its judgments with human annotations on 200 trajectories, finding 96% task-success agreement. Human judgments support retry gains over Base and confirm adoption after checking on audited tasks. We release trajectories, versioned scoring, and reference and delivery audits. These findings highlight evidence availability and answer selection as complementary dimensions of
agent reliability.
Read the full abstract ▾
High Energy Physics (Phenomenology)
Impact of Open Heavy-Flavour Production on nCTEQ Nuclear PDFs
Jan Wissmann, Tomáš Ježo, Michael Klasen et al. (+2 more)
We present nCTEQ26OHF, a global analysis of
nuclear parton distribution functions (nPDFs) within the nCTEQ framework. In this analysis, constrasting the main nCTEQ26 release, we replace the
Crystal-Ball function, an effective
matrix element that is used in nCTEQ26 for open heavy-
flavor (OHF) and quarkonium production, by an SACOT GM-VFNS calculation for the OHF predictions. The resulting nCTEQ26OHF PDFs are compatible with the nCTEQ26 PDFs within uncertainties, implying that the OHF data alone provide ample constraint.
Read the full abstract ▾
All authors Jan Wissmann, Tomáš Ježo, Michael Klasen, Ingo Schienbein, Hubert Spiesberger
Multiagent Systems
FlowMAS: Learning Multi-Agent Workflow Topology via Information-guided Generative Flow Network
Haitao Wang, Chenjing Liang, Haipeng Zhang et al. (+6 more)
Automated multi-
agent systems offer clear advantages over manually designed ones in scalability and adaptability, but existing workflow
topology methods still face important limitations. Search-based methods are often computationally expensive, textual-gradient-based methods rely on coarse-grained feedback, and existing generation-based methods are not well suited to discrete workflow
topologies with complex dependencies. To address these limitations, we propose FlowMAS, a multi-
agent workflow
topology method based on Generative Flow Networks (GFlowNets). FlowMAS models workflow generation as reward-guided flow over the
topology space and introduces three components: a GFlowNet-based
topology generation backbone, a curiosity-driven module for structure-aware exploration, and an information-guided optimization module for evaluating intermediate
topologies. Concretely, the curiosity-driven module encourages exploration of structurally novel workflows, while the information-guided module measures both the information contribution and the communication efficiency of different operators to favor more informative and effective collaboration patterns. Experiments on six
benchmark datasets with three LLM backbones show that FlowMAS consistently outperforms multiple baselines.
Read the full abstract ▾
All authors Haitao Wang, Chenjing Liang, Haipeng Zhang, Jiawei Hu, Sicheng Wang, Songzhu Mei, Chenglu Wen, Siqi Shen, Cheng Wang
Robotics
CoRe-VLA: Preserving Cross-View Coordination in VLAs under Camera Shifts
Tianhang Pan, Xuanhao Wang, Yiwen Pang et al. (+4 more)
VLAs combine pretrained
vision-
language representations with action generation to enable
language-guided control across diverse tasks, becoming a mainstream paradigm in
embodied intelligence. However, multiple studies have reported VLA's substantial declines in task success under camera shifts, revealing a key vulnerability that limits reliable deployment. To address this vulnerability, existing methods collect paired observations of the same scene from different viewpoints to fine-tune the VLA or train visual adaptation modules. Unfortunately, they require additional data collection and VLA training costs. In this paper, we first identify cross-view coordination breakdown under external camera shifts: the
robot may rely too heavily on wrist-view cues and consequently execute subtasks in the wrong order when losing global view. Motivated by this, we propose CoRe-VLA, a plug-and-play framework requiring neither additional multi-view data collection nor VLA fine-tuning, which can incorporate with exsiting VLAs. It reconstructs a scene point cloud and renders the observation from the VLA's training viewpoint to restore cross-view coordination. In CoRe-VLA, Render-to-Camera (R2C) Restoration reduces rendering-induced visual degradation, while Execution-Trajectory-Conditioned
Alignment (ETCA) reduces
robot idle time and mitigates motion conflicts during asynchronous execution. Experiments on 5 real-
robot tasks, LIBERO-100 and LIBERO-Plus demonstrate CoRe-VLA substantially improves task success across mainstream VLAs under camera shifts. For example, CoRe-VLA raises PI0.5's success rate from 13.3% to 83.3% at a 1.6m camera shift in real-
robot environment.
Read the full abstract ▾
All authors Tianhang Pan, Xuanhao Wang, Yiwen Pang, Bo Zhou, Jun Yang, Min-Ling Zhang, Shimin Di
Quantum Physics
A quantum thermal machine surpassing the classical thermodynamic limit on precision
Simon Sundelin, Ludvig Nordqvist, Khalak Mahadeviya et al. (+4 more)
Precision in nonequilibrium processes comes at a thermodynamic cost: suppressing fluctuations generally requires increased dissipation. Thermodynamic uncertainty relations (TURs) make this trade-off quantitative by linking current fluctuations to
entropy production in classical
stochastic dynamics. In the decade since its discovery, the canonical steady-state TUR and its finite-time generalizations have become a cornerstone of non-
equilibrium thermodynamics, constraining the performance of molecular machines and allowing heat dissipation to be inferred from observable fluctuations. Whether the canonical TUR can be violated in a controlled
quantum device remains an outstanding experimental question, in part because doing so requires resolving extremely small steady-state currents as well as their fluctuations. Here we experimentally show that steady-state
quantum transport can surpass the precision permitted by the canonical TUR. We observe this violation in a superconducting
quantum thermal machine coupled to a microwave waveguide acting as a cold bath and to a classical noise source providing an effective infinite-temperature bath. We observe a TUR ratio $Q = 1.71 \pm 0.17$, in violation of the classical bound $Q \geq 2$. Our results demonstrate a fundamental distinction between classical and
quantum thermodynamics, paving the way for
quantum thermal devices that achieve enhanced precision at reduced energy cost.
Read the full abstract ▾
All authors Simon Sundelin, Ludvig Nordqvist, Khalak Mahadeviya, Vyom Kulkarni, Mohammed Ali Aamir, Mark T. Mitchison, Simone Gasparinetti
Machine Learning
Multimodal Detection of Higher-Order Behavioral Constructs: Self-Compassion in Structured Reflective Interaction
Siddhant Jain and Dimitra Tsovaltzi
Many of the qualities that matter most in how people learn and grow, how someone regulates their
emotions, reflects on a setback, or stays aware of others during a difficult conversation, are not directly observable. They have to be inferred from how someone speaks, moves, and sounds over time, and they resist the kind of clean labeling that most machine learning pipelines are built around. We study this challenge through a case that is well grounded in psychological theory but rarely modeled computationally: self-compassion, the tendency to respond to one's own setbacks with patience rather than harsh self-criticism. We examine how it appears during structured reflective interviews in a technology-mediated training setting, where people naturally talk through socio-emotionally demanding situations. Since no existing
dataset captures this kind of construct in this kind of setting, we collected and annotated 51 reflective dialog sessions using an independent, temporally overlapping annotation scheme grounded in established theory. We consolidate the underlying six-component psychological model into a three-class supervision space, balancing self-kindness and mindfulness against self-critical or overwhelmed states, and build a reproducible window-based pipeline that aligns
video, audio, and text on a shared timeline. Unimodal models trained on each modality separately are compared against a simple probability-level fusion strategy, which yields modest but consistent gains over the best single modality. We close by discussing where each modality succeeds or struggles, what this suggests about how this kind of construct is actually expressed in reflective
speech, and what would be needed to model it, and constructs like it, more effectively.
Read the full abstract ▾
Computer Vision
Improved Distributional Diffusion Models
Tommaso Martorella, Alexandre Galashov, Felix Krause et al. (+4 more)
Distributional
Diffusion Models (DDMs) replace the standard mean-prediction denoiser with a distributional denoiser trained via a scoring rule objective, learning a
stochastic approximation to $p(x_1 \mid x_t)$ rather than its conditional mean. However, scaling DDMs to modern image-generation settings faces two obstacles: (i) multi-particle training incurs overhead that scales with the number of particles, (ii) DDMs use globally fixed scoring rule hyperparameters, forcing a single trade-off across sampling budgets. We mitigate these limitations by deferring particle expansion to late
transformer layers, and the hyperparameter trade-off by introducing time-dependent scoring rule schedules informed by the dynamical regimes of~\citet{Biroli2024}. Combined with a DiT-based latent setup, these changes make DDM training practical on class-conditional ImageNet-$256^2$, achieving 4.48 FID at 4 steps and 2.38 at 50 steps with DiT-XL/2, from a single model trained from scratch in one stage, without a teacher, self-
distillation or JVPs. The result is a
stochastic few-step generator whose FID does not degrade as the sampling budget grows from 4 to 50 NFE, and the same recipe transfers to text-to-image generation. Code and pre-trained models available at https://github.com/CompVis/iDDM.
Read the full abstract ▾
All authors Tommaso Martorella, Alexandre Galashov, Felix Krause, Stefan Andreas Baumann, Valentin De Bortoli, Arthur Gretton, Björn Ommer
High Energy Physics (Experiment)
Measurements of electroweak penguin $B$ decays with missing energy at Belle and Belle II
Lorenz Gärtner
Electroweak penguin $B$ decays with
neutrinos or $\tau$
leptons in the final state are sensitive probes of physics beyond the Standard Model. Using the full Belle and Run 1 Belle II
datasets, the first search for $B^0 \to K^0_S\tau^+\tau^-$ sets the upper limit on the branching fraction to $\mathcal{B} < 8.3 \times 10^{-4}$, the search for $B^+ \to K^+\tau^+\tau^-$ sets $\mathcal{B} < 5.6 \times 10^{-4}$, and their isospin combination gives $\mathcal{B}(B \to K\tau^+\tau^-) < 5.4 \times 10^{-4}$, all at 90% confidence level (CL$_s$). The first search for inclusive $B \to X_s\nu\bar\nu$ decays at Belle II sets $\mathcal{B} < 3.3 \times 10^{-4}$. Finally, the Belle II $B^+ \to K^+\nu\bar\nu$ result is reinterpreted in the Weak Effective Theory using its published likelihood. The data prefer an enhanced vector contribution and mildly favour a nonzero tensor contribution.
Read the full abstract ▾
Artificial Intelligence
From Judgment Quality to Downstream Utility: Rethinking LLM-as-a-Judge for Open-Ended Tasks
Zheng Zhang, Lufei Li, Xinyue Tan et al. (+4 more)
LLM-as-a-Judge is increasingly used to evaluate policy responses on open-ended tasks that lack ground-truth answers. Existing work often directly converts the resulting judgments into reward signals for policy training, paying limited
attention to intrinsic judgment quality and largely restricting the use of Judges to training-time supervision. We systematically investigate judgment quality and downstream utility by examining both how judgments are elicited and how they are used. For judgment elicitation, we vary the Judge protocol along three dimensions: verdict granularity, critique usage, and evaluation batching. For judgment usage, beyond policy training, we extend Judge to test-time inference through Best-of-N selection, Judge-guided revision, and beam search. We find that, (i) Surprisingly, judgment quality and downstream utility do not always align. (ii) Judge protocol design substantially affects both intrinsic judgment quality and downstream utility. (iii) Judge guidance effectively converts test-time compute into performance gains, with benefits varying across inference strategies. Our results call for a multifaceted evaluation of LLM Judges on open-ended tasks, encompassing intrinsic judgment quality, and downstream utility.
Read the full abstract ▾
All authors Zheng Zhang, Lufei Li, Xinyue Tan, Yuanhao Zeng, Ziwei Shan, Yexin Li, Kan Ren
Quantum Physics
Margenau-Hill distribution as a Necessary and Sufficient Signature of Measurement Incompatibility
Partha Patra and A. K. Pan
Measurement incompatibility and the negativity of quasi-probability distributions both demonstrate the signature of nonclassicality. However, their relations largely remained qualitative, and no explicit operational connection has been established. We provide an operational link between measurement incompatibility and the Margenau-Hill(MH) quasi-probability distribution associated with two dichotomic observables. { We first derive the joint measurability (measurement compatibility) condition for any pair of dichotomic observables in arbitrary finite dimension $d$.} We then rigorously prove that for any pair of unsharp dichotomic measurements in dimension $d$, the positivity of the MH distribution is equivalent to joint measurability i.e., the MH distribution is positive if and only if the measurements are jointly measurable. We further introduce the MH-like quasi-probability distribution for $n$ dichotomic unsharp observables in arbitrary finite dimension and derive a sufficient condition of joint-measurability when the observables are mutually anticommuting. Finally, we propose an interferometric setup, inspired by
quantum-switch architectures, that directly reconstructs MH quasi-probability which in turn provides a test of incompatibility.
Read the full abstract ▾
Software Engineering
LoLBench: Evaluating Coding Agents with Long-Horizon Proposals on Large Software Systems
Yun Peng, Zihan Wu, Zeyang Zhuang et al. (+6 more)
Modern coding
agents can deliver increasingly large repository-level changes, and recent
benchmarks reflect this by emphasizing long-horizon tasks with large reference implementations. Many
benchmarks evaluate coding
agents' implementation capability to produce correct code edits from detailed specifications. However, practical modular development tasks also require the perception capability of grounding user intent and high-level design to derive a specification. We introduce LoLBench to evaluate both capabilities through the entire proposal-to-implementation process on large software systems. It is a
multilingual benchmark of 100 tasks across 29 software systems in five domains. Each task provides a human-written enhancement proposal with user intent and high-level design. On average, proposals contain about 5,000 words, software systems contain 2.4 million source lines of code (LoC), and implementation pull requests (PRs) change approximately 5,500 LoC. Across 28
agents we evaluated, the best
agent resolves only 14% of tasks and achieves a 52.7% Fail-to-Pass (F2P) pass rate. Failure analysis identifies incomplete code localization as a major bottleneck, while providing reference-derived file trees alongside API specifications improves resolved rates by 16–22 percentage points (2.4–17$\times$), reaching at most 34%. These results show that both perception and implementation remain central challenges for coding
agents in practical modular development on large software systems. LoLBench is available at https://huggingface.co/
datasets/lolbench26/LoLBench.
Read the full abstract ▾
All authors Yun Peng, Zihan Wu, Zeyang Zhuang, Xin Zhou, Rui Shu, Xu Han, Chun Yong Chong, Yuan Wang, Jiakun Liu
Quantum Gases
Second-order coherence properties of ultrafast polariton dynamics in plasmonic lattices
Evgeny A. Mamonov, Lukas Freter, Sioneh Eyvazi et al. (+2 more)
The first-order coherence properties of strongly coupled systems supporting polariton lasing and Bose-Einstein condensation (BEC) have been thoroughly studied and show consistent results. In contrast, second-order coherence properties have been found to vary for different systems, and the second-order coherence function can deviate from the value of one, typical for atomic BEC, revealing super-Poissonian fluctuations even above the BEC/polariton lasing threshold. This calls for a deeper understanding of the emission statistics and dynamics in light-based condensates, especially in the non-
equilibrium regime. Here we demonstrate a coherent state of polariton emission from a plasmonic
lattice measured as the second-order coherence function value separately for $\mathrm{\Gamma}$- and non-$\mathrm{\Gamma}$-point radiation (ground state and high-energy tail). We observe a high degree of coherence with the value $g^{(2)}(\tau=0)-1<10^{-4}$. We also demonstrate the ultrafast ($>$1 THz) nature of the dynamics. The process can be interpreted as a BEC with ultrafast, sub-50-fs thermalization, but we also suggest an alternative explanation as superradiance in a system where the emitters have vibrational degrees of freedom.
Read the full abstract ▾
All authors Evgeny A. Mamonov, Lukas Freter, Sioneh Eyvazi, Elliot W. Lloyd, Päivi Törmä
Computer Vision
MSTypography: Multi-character Semantic Typography via Balancing Word Legibility and Object Recognizability
Xinye Yang, Xinding Zhu, Kai Fang et al. (+4 more)
Semantic typography is a design technique where the visual representation of a word conveys its semantic meaning, while maintaining its legibility. Existing digital typography methods mainly focus on single-character scenarios. They suffer from a lack of legibility constraints and insufficient local deformation when extended to multi-character words, as the intricate structures among multiple characters are hardly preserved during the typography process. In this paper, we propose a global-to-local typography framework for multi-character scenarios. It performs mask-driven silhouette approximation at the global level, while semantic-guided refinement at the local level, with a culling step in between to improve efficiency. To preserve word legibility, we designed structural losses (including explicit collision constraints and implicit Jacobian singular value constraints) and an OCR constraint for character-level readability. To enhance the object recognizability, we leverage semantic guidance with
diffusion priors, which drives the character glyph toward the target concept while preserving its structural integrity. To the best of our knowledge, this is the first multi-character semantic typography method that effectively balances word legibility and object recognizability. Evaluations on five representative
languages (English, Chinese, Japanese, Korean, Arabic) demonstrate superiority over SOTA methods. Codes will be open-sourced.
Read the full abstract ▾
All authors Xinye Yang, Xinding Zhu, Kai Fang, Xinyi Ren, Mengjian Li, Bin Cao, Jiazhou Chen
Numerical Analysis
BandPC: Learning Residual-Band Preconditioner Combinations for Flexible Conjugate Gradient Solvers
D. M. Li, Shang-Tian Yang and Xin Qiu
The conjugate gradient (CG) method is a classical iterative solver for sparse symmetric positive definite (SPD) linear systems, but its convergence strongly depends on the spectral properties of the system
matrix. Preconditioning can improve these properties; however, designing preconditioners that generalize across diverse and irregular sparsity patterns remains challenging. We propose BandPC (Band Preconditioner Combinations, where "Band" denotes residual-norm bands), a data-driven framework that leverages graph neural networks (GNNs) to predict multi-stage preconditioning strategies for the flexible conjugate gradient (FCG) method. BandPC partitions the FCG iteration into three residual-norm-based bands and defines a structured search space of 125 candidate combinations over five classical preconditioners. By exploiting FCG's ability to switch preconditioners across iterations, BandPC learns to map
matrix structure directly to a promising preconditioner sequence. To improve training and generalization, we introduce a soft-labeling mechanism that retains near-optimal combinations and normalizes their scores into a label distribution, together with a hierarchical feature representation that captures node-level attributes, edge-level algebraic coupling strengths, and global
matrix statistics. Experiments on a hybrid
benchmark of synthetic SPD problems and diverse
matrices from the SuiteSparse Collection show that BandPC achieves the best iteration count and solution time on 41.3% and 33.1% of test
matrices, respectively, even when compared with each
matrix's individually best traditional preconditioner. These results demonstrate that learned residual-band preconditioner scheduling can effectively accelerate flexible CG solvers.
Read the full abstract ▾
Computer Vision
TaoFlowForge: Progressive Native Mesh Generation via Cascaded Flow Matching
Xianze Fang, Qiyuan Feng, Dongfang Sun et al. (+6 more)
3D content generation technology has significantly advanced the work of designers, as well as the 3D printing and gaming industries. However, it remains difficult to produce lightweight, editable, and topologically clean artistic content that is directly production-ready. To achieve this, we present TaoFlowForge, an artistic mesh foundation model that generates production-ready meshes. Specifically, TaoFlowForge decomposes the mesh generation process into vertices generation and their connectivity prediction, i.e., edges. We formulate vertices generation as a two-stage coarse-to-fine process and incorporate several effective loss functions to further enhance its performance. In the connectivity prediction stage, we propose a simple yet effective method for estimating the connectivity affinity between vertices and additionally predict per-vertex normals, which determines the correct orientation of faces. Besides, we construct a large-scale
dataset combining hand-crafted 3D assets with public high-quality
topology datasets. Based on this, a carefully designed data curation pipeline is employed to filter the raw
dataset, retaining only high-quality
topology data for model training. Our model is trained on the combined
dataset and tested on both out-of-distribution hand-crafted set of 3D assets and public
datasets. Under image-conditioned generation, TaoFlowForge outperforms autoregressive methods and achieves state-of-the-art results among open-source mesh
topology generators. We will release all the code and weights together with a portion of our test
dataset.
Read the full abstract ▾
All authors Xianze Fang, Qiyuan Feng, Dongfang Sun, Yan Zhang, Xiuchao Wu, Jingnan Gao, Jiangjing Lyu, Chengfei Lyu, Gang Yu
Analysis of PDEs
Sequences of generalized moments and their applications to some moment partial differential equations
Sławomir Michalik and Hiroshi Yamazawa
We introduce sequences of generalized moments that are a generalization of sequences preserving summability and sequences of moments from Balser's theory of moment summability. We construct extensions of Laplace and Borel integral operators for these new sequences. We find integral representations of moment derivatives for generalized moments. We show the characterization of summable solutions of moment-PDEs with generalized moments. We describe singular directions for summable solutions of moment-PDEs with generalized moments and we express jumps across these singular directions in terms of hyperfunctions and in terms of kernel functions associated with sequences preserving summability, which are connected with given sequences of generalized moments.
Read the full abstract ▾
Numerical Analysis
Mixed-Precision Computing for Scientific Discovery: Formats, Co-Design, and Responsible Approximation
Emmanuel Agullo, Hartwig Anzt, Daniel Bauer et al. (+37 more)
Reduced and mixed precision have moved from a niche optimization to a central design axis in scientific computing and engineering, driven by energy constraints, heterogeneous accelerators, and the convergence of simulation and machine learning. This paper organizes the landscape around seven coupled themes—number formats, floating-point emulation, emerging architectures, hardware/software co-design, relation to other approximations, software design, and precision as a multilevel resource —and, for each theme, synthesizes the state of the art, future directions, and open questions. We emphasize energy per trusted solution as the core objective, and we frame \say{recklessly responsible} computing as a pragmatic doctrine: exploit low precision aggressively, but with systematic detection, escalation, and certification pathways.
Read the full abstract ▾
All authors Emmanuel Agullo, Hartwig Anzt, Daniel Bauer, David Bindel, Alfredo Buttari, Alexandru Calotoiu, Erin Claire Carson, Pasqua D'Ambra, Ieva Daužickaitė, James W. Demmel, Jack Dongarra, Iain Duff, Massimiliano Fasi, Dominik Göddeke, Stef Graillat, Laslo Hunhold, Roman Iakymchuk, Fabienne Jézéquel, Nils Kohl, Harald Köstler, Jakub Kružík, Julien Langou, Xiaoye Sherry Li, Hatem Ltaief, Piotr Luszczek, Yuxin Ma, Theo Mary, Mantas Mikaitis, Hiroyuki Ootomo, Daniel Osei-Kuffuor, Enrique S. Quintana-Ortí, Ulrich Rüde, Jennifer Scott, John Shalf, Linda Stals, Rasmus Tamstorf, Stefan Turek, Petr Vacek, Bastien Vieublé, Rio Yokota
Quantum Physics
Arbitrary high-dimensional entanglement purification with the superposition of paths
Zi-Han Zheng, Wen-Qiang Liu, Chen-Ming Bai et al. (+1 more)
High-dimensional
quantum entanglement offers significant advantages over its two-dimensional counterparts in various
quantum information processing tasks. We propose a scalable method for implementing an arbitrary high-dimensional
entanglement purification protocol (HDEPP) using path superposition. By selecting appropriate single-qudit gates, qudit-flip errors and phase-flip errors can be purified. Counterintuitively, the HDEPP becomes more robust for higher-dimensional systems and exhibits enhanced purification fidelity as the dimension increases. Our HDEPP not only requires fewer operational resources but also in principle can achieve unity fidelity through multiple rounds of iteration. Furthermore, we develop feasible optical schemes to purify the frequency degree of freedom of
photons suffering from qudit-flip and phase-flip errors using hyperentanglement. This work paves the way toward high-dimensional
quantum communication and
quantum networks.
Read the full abstract ▾
All authors Zi-Han Zheng, Wen-Qiang Liu, Chen-Ming Bai, Hai-Rui Wei
Computer Vision
Multi-Granularity Language-Guided Imitation Learning via Instruction Decomposition
Yi-Pei Chiu and Wei-Ta Chu
Using
language instructions as conditions to guide
robot policy learning has recently become an important research domain. However, existing
language-guided policy learning methods typically use an overall task description to guide the entire demonstration trajectory. For manipulation tasks involving multiple execution stages, these methods assign the same
language description to different subtasks, making it difficult to distinguish the behaviors required at different stages. In this work, we propose a multi-granularity
language guidance method based on instruction decomposition. The proposed method decomposes an overall task description into more fine-grained, concrete subtask-level
language instructions, thereby enhancing learning efficiency and improving performance. We evaluate the proposed method in the setting of multi-task imitation learning and validate its effectiveness.
Read the full abstract ▾
Quantum Physics
SQUARE: Structured Quantum Representation Adapters as Compact Quadratic Feature Maps for Frozen Language Models
Emily Jimin Roh, Hyojun Ahn, Hoyeong Lee et al. (+4 more)
Frozen
language models (LMs) are increasingly used as fixed feature extractors for downstream reranking, scoring, and preference modeling, raising a practical question: how should a compact module represent interactions among features in a fixed low-dimensional bottleneck? Common linear and low-rank adapters remain linear at the adaptation module itself, whereas explicit second-order alternatives introduce pairwise interactions through direct parameterization or predefined factorizations. We propose SQUARE, a Structured
QUAntum REpresentation adapter that amplitude-encodes the bottleneck vector, applies a parameterized
quantum circuit, and measures the resulting state. We show that each basis-probability feature is exactly a normalized quadratic form in the bottleneck coordinates, while the additional Pauli-$Z$ readouts are signed linear combinations of these probabilities. The measured map can therefore parameterize interactions over $O(d^2)$ coordinate pairs through a small set of shared
circuit parameters, where $d$ is the bottleneck dimension. It provides a structured parameterization within, rather than beyond, the classical normalized-quadratic feature class. In a disjoint same-pipeline evaluation over eight GLUE-derived controlled interaction tasks and five shared seeds, SQUARE achieves an average test accuracy of $0.7565$, compared with $0.7355$ for an affine normalized-quadratic predictor, $0.7271$ for the evaluated parameter-matched Givens mixing model, $0.6817$ for an MLP, and $0.6155$ for a frozen-
circuit control. Under reduced supervision, it also shows consistent gains over the strongest evaluated classical comparator, with the same qualitative pattern across multiple frozen LM backbones. All
circuit experiments use simulation, while the learned feature map can be evaluated exactly in batched PyTorch without
quantum hardware.
Read the full abstract ▾
All authors Emily Jimin Roh, Hyojun Ahn, Hoyeong Lee, Soohyun Park, Sung Whan Yoon, Vaneet Aggarwal, Joongheon Kim
Functional Analysis
Stability of isometries of quantum states
Sam Looi
We prove dimension-independent Hyers–Ulam stability for surjective approximate isometries of the positive trace-class cone in both the trace and Bures metrics. Here $\varepsilon\geq0$ is a uniform upper bound on the additive error in preserving distances. Such an approximate isometry is uniformly approximated by a Wigner
symmetry (a
unitary or antiunitary conjugation), with error at most $9\varepsilon/2$ in the trace metric and $2\sqrt2\varepsilon$ in the Bures metric, even if it does not fix zero. For maps fixing zero, the trace bound improves to $3\varepsilon$ even under the hypothesis of $\delta$-surjectivity, independently of $\delta$. In infinite dimensions, we construct spectrum-preserving homeomorphisms of the full state space, including mixed states, whose distortions tend to zero in both metrics but which stay a fixed uniform distance from every Wigner
symmetry. For state spaces, stability holds in each fixed finite dimension with a dimension-dependent modulus tending to zero as $\varepsilon\to0$, but no such modulus can be chosen independently of dimension.
Read the full abstract ▾
Artificial Intelligence
Train Ahead, Distill Back: Bootstrapping On-Policy Self-Distillation for Large Language Models
Zheng Zhang, Xinyue Tan, Lufei Li et al. (+3 more)
On-policy self-
distillation (OPSD) improves large
language models by letting a self-teacher with privileged information provide dense
token-level supervision on the model's own trajectories. Yet existing methods typically construct the self-teacher from the current, initial, or slowly averaged policy state, leaving the quality of supervision constrained by the teacher's ability to exploit privileged information. We ask whether the model's own optimization progress can instead be recycled into a stronger self-teacher. In this paper, we introduce Bootstrapped On-Policy Self-
Distillation (B-OPSD), which temporarily trains the policy ahead to obtain a future teacher, restores the student to the original policy state, and then uses the future teacher to supervise the restarted student. The future teacher improves supervision in two complementary ways, it can generate more reliable privileged trajectories and, conditioned on them, provide more informative
token-level targets along the restarted student's on-policy trajectories. Experiments on mathematical
reasoning with Qwen3-4B and Qwen3-8B show consistent improvements over standard OPSD in both settings, including gains from 27.50 to 41.30 and from 48.80 to 64.44 in the rollout-privileged setting. Our findings point to a broader principle for self-improving models that future learning progress can be distilled backward, preserving acquired knowledge while bootstrapping beyond the optimization state that produced it.
Read the full abstract ▾
All authors Zheng Zhang, Xinyue Tan, Lufei Li, Xinyi Zhang, Yexin Li, Kan Ren
Robotics
ReF-HIL: Shaping the Critic around Human Action Neighborhoods for Efficient Human-in-the-Loop Reinforcement Learning
Shaoyin Luo, Song Wang, Shibo Xia et al. (+5 more)
Human-in-the-loop reinforcement learning (HIL-RL) offers a promising route to efficient training of robotic manipulation policies by combining autonomous learning with human demonstrations and online corrections. However, insufficient use of successful human experience in value learning prolongs costly real-world training, while persistent imitation penalties can limit value-driven policy improvement. To address these limitations, we propose ReF-HIL, an efficient HIL-RL framework that uses human guidance to accelerate the learning process. Human-Reference-Guided Value Shaping learns an independent value reference from successful human experience to guide online value learning, while incorporating local corrective feedback. A Human Action Fence defines a learned human-action neighborhood, allowing value-driven optimization for better performance without imitation penalties inside while constraining policy and value updates outside. Experiments on five diverse and challenging real-world manipulation tasks demonstrate improved overall learning efficiency and higher success rates compared with the evaluated baselines. Specifically, ReF-HIL reaches 90% autonomous success in only 18-63 minutes of active training and achieves final success rates of 91.7-100%. These results highlight the potential of human-guided reinforcement learning to acquire reliable manipulation skills efficiently in the real world. Project website: https://anonymous.4open.science/w/ReF-HIL-7762/
Read the full abstract ▾
All authors Shaoyin Luo, Song Wang, Shibo Xia, Tianle Zhang, Zhaowei Liang, Guanghui Shen, Bin Wang, Dan Wu
Lattice Field Theory
Progress on $Sp(4)$ lattice theories with Grid: Möbius domain wall fermions and continuum extrapolations
Gianmarco Simonetti, Ed Bennett, Peter A. Boyle et al. (+10 more)
We present preliminary results obtained in the first extensive study of the Möbius domain wall
fermion (MDWF)
lattice formulation of the four-dimensional Sp(4) gauge theory coupled to two dynamical
fermions transforming in the fundamental representation. This theory is important for extensions of the Standard Model, as it provides the microscopic origin for both composite
Higgs and dark matter models based on the SU(4)/Sp(4) coset. We scan the space of bare
lattice parameters, including those entering the MDWF, and monitor spurious effects, such as the residual mass. We identify optimised regions of parameter space in which
lattice artefacts affecting the global
symmetries of the theory are suppressed. We then perform preliminary measurements of the lightest pseudoscalar and vector
meson masses, together with the pseudoscalar decay constant. We demonstrate the potential of the MDWF formulation to enable continuum-limit extrapolations with reduced
lattice discretisation effects.
Read the full abstract ▾
All authors Gianmarco Simonetti, Ed Bennett, Peter A. Boyle, Luigi Del Debbio, Niccolò Forzano, Ryan C. Hill, Deog Ki Hong, Jong-Wan Lee, C.-J. David Lin, Biagio Lucini, Maurizio Piai, Davide Vadacchino, Alexis Verney-Provatas
Strongly Correlated Electrons
Evolution of Magnetism in Ce$_4$Ge$_7$ under Magnetic Field and Pressure
Kaixin Ye, Qihe Yu, Yongjian Li et al. (+8 more)
We report the
magnetic and transport properties of single-crystalline Ce$_4$Ge$_7$, which crystallizes in a non-centrosymmetric orthorhombic structure (space group C222$_1$). It undergoes an
antiferromagnetic transition at $T_N$ = 7.3 K. When a
magnetic field is applied along the $b$-axis, a metamagnetic transition occurs at 1.3 T (at 2 K). Correspondingly, this transition gives rise to an anomalous
Hall effect, which is dominated by the intrinsic Karplus-Luttinger mechanism. Under hydrostatic pressure, $T_N$ of Ce$_4$Ge$_7$ initially increases slightly and is then gradually suppressed. For pressures above $P_c$ = 10.2 GPa, no
magnetic order is observed. The divergence of the $A$ coefficient, the maximum of the residual resistivity, and the non-
Fermi liquid behavior around $P_c$ indicate the possible existence of an
antiferromagnetic quantum critical point in Ce$_4$Ge$_7$.
Read the full abstract ▾
All authors Kaixin Ye, Qihe Yu, Yongjian Li, Yanan Zhang, Ye Chen, Rui Li, Lin Jiao, M. Smidman, Yongjun Zhang, Yu Liu, Huiqiu Yuan
Artificial Intelligence
SkillCome: Group Contrast Skill Optimization with Dual Memory
Haolin Li, Feng Hong, Ang Li et al. (+6 more)
Skill
evolution improves the capabilities of large
language models by analyzing trajectories generated under a given skill and modifying the skill accordingly. Existing approaches typically generate a single trajectory per question. However, this provides insufficient optimization signals since it requires inferring effective skill edits from a solitary path. It is difficult to pinpoint which actions caused the failure in a failed trajectory, or to determine which actions in a successful one should be incorporated into the skill. Furthermore, they rely on a local batch of trajectories for analysis, making the optimization direction susceptible to noisy evidence. To address these, we propose SkillCome, a Skill-
evolution method based on group Contrast optimization with dual
memory. For each question, SkillCome generates trajectories and performs group contrast analysis to precisely identify key behavioral divergences between successful and failed trajectories, offering reliable optimization signals. The dual
memory system further accumulates evidence from historical steps to track patterns shared across different groups, leading to more generalized optimization directions. Together, SkillCome builds a systematic optimization process that transforms experience from observed successful trajectories into reusable skills. Extensive experiments on six
benchmarks spanning question answering,
reasoning, and agentic tasks demonstrate the effectiveness of our method. SkillCome consistently outperforms baselines across five models of varying families and scales, with gains up to +5.69 points.
Read the full abstract ▾
All authors Haolin Li, Feng Hong, Ang Li, Chilin Fu, Weichang Wu, Ya Zhang, Yanfeng Wang, Xiaolu Zhang, Jiangchao Yao
Computation and Language
LLM unbranding: Erasing Commercial Identity while Preserving Generic Utility
Kajetan Ożóg, Alicja Wojciechowska, Dawid Malarz et al. (+3 more)
Establishing unbranding as a critical practice to prevent visual logos from acquiring negative connotations is standard in image generation. Large
Language Models (LLMs) now face a parallel and emerging challenge. These models frequently generate brand descriptions within diverse contexts. This frequency introduces significant risks, such as trademark dilution, false attribution, and brand defamation. In response, we formally define the novel task of LLM Unbranding. We specifically address the complex challenge of managing trade dress within textual outputs. This involves neutralizing characteristic
language, slogans, and stylistic markers that define brand identity. Crucially, these elements are less evident than explicit visual logos. To
benchmark this task, we introduce a comprehensive evaluation
dataset incorporating prominent brands from multiple commercial domains. We rigorously evaluate existing state-of-the-art machine unlearning models using this
benchmark. This evaluation identifies their limitations in selective textual unbranding. Finally, we propose MUTE, a novel inference-time method that effectively neutralizes textual trade dress while preserving the LLM's general capabilities and utility. By leveraging an iterative refinement loop, MUTE systematically optimizes system instructions to safely eliminate brand leakage without requiring fragile parameter updates. Code and
dataset: The evaluation
dataset and code for LLM Unbranding are available at https://github.com/KajetanOzog/LLM_unbranding. The implementation of MUTE is available at https://github.com/KajetanOzog/MUTE.
Read the full abstract ▾
All authors Kajetan Ożóg, Alicja Wojciechowska, Dawid Malarz, Paweł Batorski, Artur Kasymov, Przemysław Spurek