Artificial Intelligence
MAADBench: The Refreshable Paradigm for Anomaly Detection in Multi-Agent Systems
Lei Ma, Dennis Hofmann, Haowen Xu et al. (+4 more)
Recent studies report that LLM-based multi-
agent systems (MAS) fail at rates of 41%-87%, yet to our knowledge, no
benchmark to date supports systematic anomaly detection (AD) for them. Building MAS AD
benchmarks is hard because they must remain fresh as LLM systems evolve: tasks may leak into training data and thus be memorized by LLMs, traces and anomaly patterns expire as backbones evolve, and labels must be provided reliably for each refresh. To address these challenges, we present MAADBench (MA: multi-
agent; AD: anomaly detection), the first refreshable MAS AD
benchmark designed for diverse, evolving LLM backbones underlying the
agents. MAADBench combines (1) sampled-and-coupled generative tasks over an approximately 10^37-task space to mitigate task leakage, (2) refreshable trace generation under configurable LLM backbones, and (3) automated provision of cost-free, deterministic step-level labels for fine-grained AD evaluation. Beyond offering the paradigm itself, we run MAADBench with five state-of-the-art LLM backbones and release the MAADBench-Full
dataset with 5,200 step-labeled traces. Benchmarking 25 AD methods on the MAADBench
dataset reveals substantial limitations in current approaches: they rely heavily on supervision, struggle with subtle MAS-specific anomalies, and lack robustness across LLM backbones. These gaps point to a rich research agenda for MAS-specific anomaly detection, with MAADBench providing a systematic and refreshable testbed for method development and evaluation. We open-source MAADBench-Full at https://huggingface.co/
datasets/hww123/MAADBench-full.
Read the full abstract ▾
All authors Lei Ma, Dennis Hofmann, Haowen Xu, Joshua DeOliveira, Peter VanNostrand, Lei Cao, Elke Rundensteiner
Quantum Gases
Generalized Effective Spin-Chain formalism for multicomponent anyons in one-dimensional optical lattices
Sagarika Basak, Xi-Wen Guan and Han Pu
We develop a generalized effective
spin-chain (GESC) formalism for strongly interacting multicomponent anyons in a one-dimensional (1D) optical
lattice. By mapping particle motion onto spinless
fermions and
spin states onto an ordered chain, this framework provides a
spin-charge-separated perspective on the physical effects of fractional exchange statistics. In the strong-interaction regime, the leading-order charge
Hamiltonian becomes independent of the statistical phase; instead, virtual tunneling through doubly occupied states directly imprints this phase on the
spin-exchange coefficients. The GESC formalism captures ground-state properties of the full Anyon-Hubbard model reported in S.~Basak, X.-W.~Guan, and H.~Pu, unpublished manuscript (2026), and reveals that their statistical dependence is predominantly encoded in the explicit anyonic structure of the observables rather than the underlying
spin ground state. This formalism uncovers that, out of
equilibrium, successive
spin exchanges accumulate direction-dependent phases; the resulting interference drives crossover from dispersive to localized
impurity transport and generates inversion-asymmetric propagation at intermediate statistics. Suppressed expansion persists for identical and distinguishable
impurities, with differences in propagation governed by the interplay between
impurity identity and interaction anisotropy. Ultimately, GESC offers a versatile, computationally efficient theoretical framework that connects macroscopic dynamics to microscopic virtual exchange processes, conferring a unique
spin-charge-separated vantage for resolving how fractional exchange statistics survives strong interactions and manifests distinctly across static and dynamical regimes.
Read the full abstract ▾
Quantum Gases
Multicomponent anyons in one-dimensional optical lattices
Sagarika Basak, Xi-Wen Guan and Han Pu
We investigate the ground-state and dynamical properties of multicomponent interacting anyons confined in a one-dimensional (1D) optical
lattice. Adopting the Anyon-Hubbard model, we explore
spin and charge correlations as functions of interaction strength and the anyonic statistical parameter. Our results demonstrate that multicomponent effects combined with fractional exchange statistics substantially reshape both charge and
spin correlations. For fractional exchange statistics, the canonical symmetric shell structure with prominent singularities in the spinor fermionic momentum distribution undergoes notable structural reconstruction, exhibiting emergent asymmetry, new spectral peaks, and broadened singular features. Unlike the Tonks-Girardeau bosonic gas, pseudobosons, representing a limiting case of anyons, cease to display a dominant zero-momentum peak. Instead, a quasi-fermionic shell structure emerges with finite-momentum peaks, whose locations are modulated by
lattice site occupancy and interaction strength. The structure factor of
spin correlations uncovers
antiferromagnetic ordering in spinor anyon systems, with correlation magnitudes tunable by anyonic exchange statistics. Furthermore, quench dynamics analysis reveals the statistical phase parameter as an effective tuning knob. It enables the switching of dipole oscillations between underdamped and overdamped relaxation regimes and governs overall cloud expansion dynamics. Our findings pave the way for exploring statistics-driven ground-state and dynamical phase transitions in spinor anyon systems.
Read the full abstract ▾
Robotics
Learning to Explore Hidden Kinematics for Articulated Object Manipulation
Ruiyao Liu, Boshu Lei, Zhuoyang Pan et al. (+1 more)
The kinematics of an articulated object is often ambiguous from
vision alone. Interaction resolves the ambiguity, and active perception methods exploit this by searching for the single action that most sharpens a belief over the kinematic parameters at each step. Such greedy search cannot be extended over a horizon without forward models of the contact and inertial dynamics, which are themselves unknown. We instead amortize action selection into training. We maintain a belief distribution over joint type and parameters, initialized from a generative prior and updated by
Bayesian filtering on the observed part motion. To condition the policy on this belief, we render it as a per-point articulation flow field, the motion that the current posterior predicts for every point on the object. Carrying the inductive bias of articulated motion, this representation generalizes better than a latent encoding of the belief or flow tracked from observation. We train the policy with reinforcement learning, rewarding the
entropy that each interaction removes from the posterior, so that informative exploration becomes learned behavior rather than a search at every step. Our method outperforms previous approaches across door and drawer manipulation on the PartManip
benchmark, and reaches 61.7% success on ArticuRiddle, a new
dataset of objects whose appearance implies the wrong articulation, against 44.4% for the best previous method. Project Website: https://hiddenkinematics.github.io/
Read the full abstract ▾
All authors Ruiyao Liu, Boshu Lei, Zhuoyang Pan, Kostas Daniilidis
Machine Learning
SERA: Scale-Equalized Rollout Allocation for Maximum Likelihood Reinforcement Learning
Zihao Chen, Fanxiang Xiong, Hongran Ren et al. (+6 more)
Maximum Likelihood Reinforcement Learning (MaxRL) targets
prompt-wise log-success and has shown strong performance on
reasoning tasks. Under finite rollout budgets, however, the
estimator used by MaxRL attenuates each
prompt's likelihood gradient by a factor that depends on its success probability and rollout count. Under uniform rollout allocation, the common rollout count fails to compensate for success-dependent attenuation, leaving low-success
prompts more strongly attenuated and distorting their relative contributions to the expected aggregate gradient. We introduce SERA (Scale-Equalized Rollout Allocation), which redistributes a fixed rollout budget to approximately equalize these finite-rollout scaling factors. Building on our theoretical analysis of how finite rollouts distort
prompt-wise likelihood gradients, we formulate the allocation as a fixed-budget max–min problem, derive a waterline solution to its continuous relaxation, and introduce a multiplicity correction to remove the additional
prompt weighting induced by heterogeneous rollout counts. Experiments show stronger
alignment with exact likelihood gradients in a controlled ImageNet setting and improved multi-sample solution coverage over MaxRL on maze
navigation and mathematical
reasoning under matched training rollout budgets.
Read the full abstract ▾
All authors Zihao Chen, Fanxiang Xiong, Hongran Ren, Xuefeng Bai, Zhongxiang Dai, Kehai Chen, Zhiguo Zhang, Zhiyong Wang, Yu Cheng
Combinatorics
Degree conditions for $k$-strong orientations of digraphs
Jørgen Bang-Jensen and Shuo Wei
Jackson and Thomassen conjectured that every $2k$-strong digraph contains a spanning $k$-strong oriented subdigraph. We prove sharp degree conditions for the existence of such a subdigraph. For every fixed positive integer $k$ and all sufficiently large $n$, every $n$-vertex digraph $D$ with $\delta^0(D)\ge\lfloor(n+k-1)/2\rfloor$ admits a $k$-strong orientation. This threshold is best possible even for the weaker conclusion that $D$ itself is $k$-strong. We also prove a sharp Woodall-type analogue for every fixed positive integer $k$ and all sufficiently large $n$: if $d_D^+(x)+d_D^-(y)\ge n+2k-2$ for every missing arc $xy$, then $D$ admits a $k$-strong orientation, and this bound is again best possible. As a further consequence, we determine the sharp minimum total degree threshold. Finally, the semi-degree result also remains valid when $k\le\alpha n$ for every fixed $0<\alpha<0.094882\ldots$.
Read the full abstract ▾
Computation and Language
Grounded Revision vs. Prior Injection: Probing Retrieval-Augmented Patent Claim Amendment
Josepha Michiko Leo, Hyun-seok Min, Yehoon Jang et al. (+3 more)
Retrieval-augmented generation is widely used in professional writing, yet whether
retrieval grounds revision or merely injects templates is rarely tested where "correct" has a definable meaning. Patent claim amendment supplies that signal: the examiner names the attacked limitation and cites prior art, providing per-case ground truth. We release three artifacts: (i) a corpus of 7,385 USPTO prosecution cases with XML-aligned pre/post claims, rejection, and cited prior art; (ii) a seven-probe battery comparing random and structural-match
retrieval as two policies under a fixed
prompt scaffold; (iii) a deterministic five-channel metric (C1-C3 and C5 in main, C4 supplementary) requiring no LLM evaluation. Across 9,600 pre-registered calls on four frontier LLMs (Claude Sonnet 4, Claude Haiku 4.5, GPT-5.4, GPT-4o-mini), no tested model exhibits detectable classical prior-injection behavior;
retrieval effects are small and direction-inconsistent between random and structural
retrieval, and the null is unchanged under a dense (semantic) retriever, across
retrieval depths k in {1,3,5,10}, and under a paraphrase-sensitive grounding metric. Revision locality reveals a model-specific difference that the template channel misses. The four-
cell taxonomy, which we treat as exploratory, leaves the prior-injector
cell unoccupied.
Read the full abstract ▾
All authors Josepha Michiko Leo, Hyun-seok Min, Yehoon Jang, Irvan Zidny, Jin-Woo Chung, Sungchul Choi
Image and Video Processing
MedForge-RSI: Medical Deepfake Detection via Recursive Self-Improvement
Zhihui Chen and Mengling Feng
Text-guided image editors can generate high-fidelity medical deepfakes, challenging the reliability of clinical imagery. Although
reasoning-based detectors perform strongly in distribution, they degrade substantially under deployment shift. MedForge-Reasoner, an 8B
vision-
language model trained with supervised fine-tuning and reinforcement learning, achieves 99.2% accuracy on its target distribution, yet misclassifies 40% of authentic scans, reaches only 77% accuracy on unseen generators, and falls to 59% under transmission distortion. Adapting such models through conventional retraining is costly, requiring large-scale supervision and expert-designed guidelines. We introduce MedForge-RSI, a recursive self-improvement framework that enables a deployed detector to adapt while keeping its model weights frozen. Over 20 rounds, the detector analyzes verified errors, accumulates reusable experience, and autonomously develops image-analysis tools, while an independent acceptance test retains only validated improvements. Across 49 registered configurations, MedForge-RSI increases average accuracy over four test sets from 75.0% to 84.4%. On a held-out 4,000-image evaluation, it improves clean accuracy from 76.5% to 87.9% and transmission-distorted accuracy from 59.0% to 70.5%, with the largest gains in authentic-image recall. Controlled analysis across all 49 trajectories identifies which self-improvement mechanisms replicate across seeds, shows acceptance testing to be the largest individual contributor, and reveals a taxonomy of failed adaptations. We release the complete trajectories, including all rejected and rolled-back changes.
Read the full abstract ▾
High Energy Physics (Theory)
Classical Shadows of Higher-Form BRST Anomalies
Ruizhi Shen and Fu-Wen Shu
We formulate a precise sense in which a BRST anomaly may possess a classical phase space shadow. It is the non-equivariance class of a
Hamiltonian symmetry action on covariant phase space. To organize this statement, we introduce the Weil covariant phase space bicomplex, whose basic subcomplex controls equivariant
Hamiltonian lifts. We then compare this purely classical class with local BRST descent. For a
Hamiltonian admissible mixed transgression, the ghost-number-two descendant $a^{2}_{d-1}$ defines a deghostification map to the charge
algebra, which is generated by the failure of the Cartan representative to be equivariant. The induced
cohomology class is independent of Bardeen type changes of descent representative. Thus the classical anomaly is a shadow of the BRST anomaly in the precise sense that both are realizations of the same Weil transgression class, while remaining objects of different theories. As an explicit example, we analyze a five dimensional inflow transgression and show that the classical charge cocycle and the BRST descendant arise as two realizations of the same Weil transgression class.
Read the full abstract ▾
Signal Processing
Single-Voxel Wireless NeRF for Spatial Spectrum Prediction
Raj Bakhunchhe, Arun Paidimarri, Sara Garcia Sanchez et al. (+3 more)
Wireless channel measurements across multiple spatial directions are crucial for AI-driven applications, such as RF digital twins and integrated communication and sensing. However, collecting channel data across large scenes is labor-intensive. Wireless NeRFs address this challenge by learning propagation behavior from sparse measurements and synthesizing channel spatial spectrum magnitude at unseen locations. However, existing wireless NeRFs inherit dense volumetric sampling from
vision NeRFs, which requires substantial computation. This paper asks whether such dense sampling is necessary for predicting magnitudes of the wireless spatial spectrum. We empirically show that wireless NeRFs are over-parameterized for this task and introduce SV-INGP, a sparse volumetric sampling variant of Instant Neural Graphics Primitives (INGP). Across real-world and simulated
datasets, SV-INGP matches the median Structural Similarity Index Measure (SSIM) of the NeRF2 baseline while reducing training time by 184x. These results generalize across LoS and NLoS scenes, sub-6 and millimeter-wave frequencies, and
antenna array tapering configurations, suggesting a simpler and more efficient design path for RF digital twins
Read the full abstract ▾
All authors Raj Bakhunchhe, Arun Paidimarri, Sara Garcia Sanchez, Asaf Tzadok, Ali Tajer, Ish Kumar Jain
Machine Learning
Interactive-Policy Distillation with Bidirectional Propose-and-Verify
Shutong Wu, Xiwen Chen, Brendan Rappazzo et al. (+4 more)
On-policy
distillation (OPD) trains a student model on its self-generated trajectories with dense
token-level teacher feedback. However, naive OPD may suffer from teacher unanchoring, where the student's
reasoning trajectory drifts far from the teacher, causing the teacher to be queried on states it would hardly visit and thus provide unreliable supervision. We propose Interactive-Policy
Distillation (IPD), which applies adaptive teacher intervention to the student rollout. Under a bidirectional propose-and-verify state machine, the student and teacher alternately exchange their roles as proposer and verifier, and collaboratively generate mixed-source trajectories. Then different supervisions are applied according to the source of each
token. This bidirectional propose-and-verify mechanism and the source-split loss make IPD not only a more performant
distillation method, but also a unified bridge between on-policy and off-policy paradigms. To make the interleaved dual-model rollouts more efficient, we also design a dedicated fused inference engine that co-hosts both models in one serving instance with separate KV caches and instantiates the state machine model to distribute, collect, and process requests. On math
reasoning tasks and across multiple teacher-student model pairs, student models trained with IPD not only outperform those trained with OPD, but also demonstrate higher data efficiency. Specifically, when distilling Qwen3-30B-A3B into Qwen3-1.7B-Base, IPD brings a +3.28 mean@8 and a +3.28 best@8
benchmark-averaged accuracy improvement compared with OPD. Besides, IPD only consumes about 1/4 of the training examples and steps to outperform OPD trained on the whole training
dataset for one epoch. We also investigate the impact of different loss variants and takeover / handback configurations, and demonstrate the robustness of IPD on different training data.
Read the full abstract ▾
All authors Shutong Wu, Xiwen Chen, Brendan Rappazzo, Daiheng Zhang, Anderson Schneider, Yuriy Nevmyvaka, Jiawei Zhang
Computer Vision
SCCM: Spherically Consistent Coarse Matching for ERP Dense Feature Correspondence
Gyeonggwan Lee, Eunsoo Im, Seunghwan Hong et al. (+1 more)
Dense feature matching between 360$^\circ$ panoramas underpins omnidirectional pose estimation, 3D reconstruction, and SLAM. Such panoramas are stored in the equirectangular projection (ERP), which unrolls the viewing sphere onto a flat chart and thereby introduces three distinct distortions – a longitudinal seam (
topology), latitude-dependent stretch (metric), and non-uniform pixel area (area) – that the coarse stage of perspective-trained dense matchers does not model, so these matchers degrade systematically on ERP. We show that correcting the three distortions at the coarse-stage interfaces where they arise – pairwise distortions in
attention, per-pixel distortion in covisibility gating – improves PCK@1$^\circ$ from 0.230 to 0.275 on Matterport3D under a fixed coarse scaffold, with the refiner architecture unchanged – our central result. Concretely, SCCM (Spherically Consistent Coarse Matching) augments a chart-naive cross-
attention/dual-softmax coarse matcher with two sphere-derived priors: Spherical Positional
Attention (SPA) pairs a yaw-periodic RoPE (
topology) with a tangent-plane bias (metric), and Area-Aware Covisibility (AAC) applies a pre-sigmoid log-area correction (area). The chart-naive scaffold serves as a controlled reference, separating the scaffold-replacement effect from the spherical-prior effect. Instantiated in the RoMa V1 framework with the same frozen encoder, refiner architecture, and loss, SCCM also outperforms the ERP-native EDM (0.163) and an ERP-retrained RoMa V1 (0.198) under a unified ERP dense matching protocol, while perspective-trained matchers largely fail on ERP. It further transfers zero-shot to Stanford2D3D and, when trained on outdoor Holo360D, leads there as well.
Read the full abstract ▾
All authors Gyeonggwan Lee, Eunsoo Im, Seunghwan Hong, Junghun Suh
Computation and Language
DraftTrace: A Multi-View Analytics Environment for AI-Integrated Writing
Divyansh Chandarana, Sandipan De and Vivek Gupta
Generative AI has changed how students produce writing assignments. The final artifact is no longer sufficient to understand the process through which it was produced. We introduce DraftTrace, a writing environment that jointly captures three complementary views of writing: the final product, the writing process and interactions with an integrated AI-assistant. DraftTrace reconstructs how a document develops over time and organizes these signals into submission, longitudinal, and class-level analytics for instructors. We deployed DraftTrace in a graduate NLP course with 81 students and compared their sessions with LLM-generated responses entered by automated tools and with copy-typed responses. While product measures distinguish differences in text formulation, process measures distinguish differences in how text is entered. Considering both views together helps characterize cases such as copy-typing. Interaction traces show that students use the assistant differently across stages of writing: to clarify the question at an early stage and to verify answers at a later stage. A preliminary instructor survey highlights the importance of multi-view writing analytics and their interpretability.
Read the full abstract ▾
Robotics
Planning Oriented 3D Scene Completion via Coupled TUDF Occupancy Representation Learning from Partial Observations
Tianyou Yu, Pengfei Zhao and Chao Xu
Partial observability remains a fundamental challenge in robotic
navigation, where limited sensor coverage and occlusions leave large portions of the environment unobserved. Existing scene completion methods primarily focus on improving incomplete mapping or reconstructing partially observed 3D structures, but rarely investigate how scene completion can be designed to benefit downstream tasks such as path planning. In this work, we propose a path-planning-oriented 3D scene completion framework that moves beyond pure occupancy modeling toward a coupled
geometric formulation. Specifically, given partial
LiDAR observations as input, the proposed framework jointly predicts completed Truncated Unsigned Distance Field (TUDF)-based continuous
geometric representations and voxel-wise occupancy maps. This coupled representation allows the network to better reason about obstacle boundaries and free-space geometry. To fully exploit the synergy between the two representations, we introduce a bidirectionally coupled learning scheme, where TUDF features provide dense
geometric guidance to improve occupancy reconstruction, while occupancy features in turn offer complementary structural constraints that refine distance-field estimation. Consequently, the proposed network directly predicts complete occupancy and TUDF representations, allowing seamless integration of TUDF into trajectory planning without post-processing. Extensive experiments on unseen environments demonstrate that the proposed method consistently improves both
geometric reconstruction quality and downstream planning performance.
Read the full abstract ▾
High Energy Physics (Theory)
Emergent Symmetries in Ensemble Averages
Masahito Yamazaki
It is widely believed that there are no exact global
symmetries in
quantum gravity. We review recent works showing that global
symmetries can nevertheless emerge after ensemble averages of theories, as encountered in holography. As a precision case study, we discuss the ensemble average of two-dimensional Narain-type conformal field theories associated with an even quadratic form of general signature, which is computed by the Siegel-Weil formula and interpreted as a sum over geometries in a three-dimensional Abelian Chern-Simons theory. Global
symmetries of the bulk anyons can emerge after the average, as T-dualities relating different theories in the ensemble are "folded" into
symmetries of a single theory (duality origami). We discuss the relation to the Swampland program, including the emergence of global
symmetries at infinite distance. Finally, we recast strong-to-weak spontaneous
symmetry breaking (SWSSB) of mixed states as an emergence of
symmetries in ensemble averages; in holography, wormholes connecting the two sides of a thermofield-double construction provide its bulk realization. This motivates a new Swampland
conjecture.
Read the full abstract ▾
Quantum Physics
Exponential Advantage of Quantum over Classical References in Leakage Detection
Zheng An
Leakage from a known encoding subspace can be detected by projection. However, the corresponding projector is unavailable when the encoding subspace is not classically specified. Here we show how independent
quantum references prepared by the same encoder enable leakage detection without a classical description of the encoding. A coherent measurement on the references and a single message leaves every state within the two-dimensional encoding subspace, including its
entanglement with a remote system, exactly unchanged. We derive the exact detection law for $M$ ideal references and prove optimality among tests with zero false alarm for every encoding. For orthogonal leakage, the miss probability is asymptotic to $4/M$, independent of the ambient dimension $d$. Measuring all references first, even collectively, gives zero detection at every finite budget under the same zero-false-alarm requirement. For a fixed detection target between zero and one, the optimal measurement-first cost is $\Theta(d/\epsilon)$ at sufficiently small tolerance $\epsilon$ on normal false alarm and conditional disturbance; a dimension-independent coherent budget suffices. For one logical
qubit encoded in three physical
qubits, seven coherent references achieve at least $50\%$ orthogonal-leakage detection, whereas any measurement-first receiver needs at least $588$ under the same $1\%$ normal tolerances.
Quantum references thus support leakage checks without classical reconstruction, with a sample advantage exponential in the number of physical
qubits.
Read the full abstract ▾
Robotics
Reactive Real-Time Flow Policies via Asynchronous Distribution Alignment
Moritz Zoellner, Reece O'Mahoney, Ioannis Havoutis et al. (+1 more)
Generalist
robot policies such as
vision-
language-action models (VLAs) have achieved remarkable generalization, but their inference delays can conflict with the demands of real-time control. Asynchronous execution avoids pauses between action chunks by predicting the next sequence of actions while the
robot carries out the previous one. In this paper, we study whether asynchronous execution produces the same action distribution as the original VLA. We find that, for non-Markovian demonstrations, asynchronous execution can produce a fundamentally different action distribution, which can limit the policy's reactivity. In our method, we seek to restore this reactivity by aligning the asynchronously produced action distribution with that of the original VLA through two complementary mechanisms. First, Recursive Flow-Field
Distillation trains the asynchronous policy using the VLA's action-generation flow. We characterize the learned distribution theoretically and show experimentally that our asynchronous policy can generate nearly the full range of actions the original VLA would produce, while existing asynchronous methods recover only a fraction of that range. Second, Propose-Resolve prepares multiple action sequences asynchronously and uses the latest observation to select among them based on a lightweight approximation of their likelihood under the VLA's action distribution. Our resulting method matches the original VLA's success on LIBERO and retains about 80% of its success on RoboMimic, about 30 percentage points more than existing asynchronous methods.
Read the full abstract ▾
All authors Moritz Zoellner, Reece O'Mahoney, Ioannis Havoutis, Rohan Paleja
Systems and Control
Invariance is Compositional for Continuous-time Systems: From Sleekness to Lebesgue Density
Othman Cherkaoui Dekkaki, Sadek Belamfedel Alaoui, Olayo Reynaud et al. (+3 more)
This work establishes the first bidirectional compositional invariance result: robust forward invariance of a global Cartesian product set is shown to be equivalent to robust forward invariance of each local subsystem under coupling inputs from its neighbors. To facilitate this, we introduce the notion of tangential Lebesgue-density, a new condition that is weaker than classical sleekness but sufficient to ensure that the product of individual tangent cones equals the tangent cone of the product set. This equivalence reduces the curse of dimensionality by allowing the verification of a high-dimensional global system to be decomposed into a series of local sub-checks. The framework's scalability is demonstrated through a DC microgrid numerical example, confirming that the verification complexity grows only linearly with the number of subsystems.
Read the full abstract ▾
All authors Othman Cherkaoui Dekkaki, Sadek Belamfedel Alaoui, Olayo Reynaud, Mohamed Maghenem, Alessio Iovine, Adnane Saoud
Materials Science
Frustrated Gd3+ Double Perovskites as High-Performance Magnetocaloric Materials for Sub-100 mK Adiabatic Demagnetization Refrigeration
Tim Treu, Marvin Klinger, Christian Heil et al. (+6 more)
Achieving temperatures below 100 mK is essential for advancing
quantum technologies and exploring fundamental
quantum phenomena. While paramagnetic salts have traditionally enabled adiabatic demagnetization refrigeration (ADR), their limitations have driven the search for more effective alternatives. In this work, we present Gd3+-based double perovskites, Ba2GdSbO6 and Sr2GdSbO6, as high-performance magnetocaloric materials. Starting ADR from 2 K and 5 T, these compounds reach 67 mK and 68 mK in small finite
magnetic fields, and 83 mK and 78 mK in zero field, respectively, which are the lowest reported ADR temperatures for Gd3+ magnets (S = 7/2) under these conditions. The frustrated geometry and the complex interplay of exchange and dipolar interactions suppress their
antiferromagnetic ordering temperatures to 100 and 166 mK for Ba2GdSbO6 and 190 mK for Sr2GdSbO6, while maintaining an outstanding
entropy density of 189 mJ K-1 cm-3 and 201 mJ K-1 cm-3, respectively. Notably, Ba2GdSbO6 sustains sub-100 mK cooling even at finite fields of 0.5 T, making it particularly promising for practical ADR applications.
Read the full abstract ▾
All authors Tim Treu, Marvin Klinger, Christian Heil, Hans-Albrecht Krug von Nidda, Vinicius E. S. Frehse, Mamoun Hemmida, Anton Jesche, Alexander A. Tsirlin, Philipp Gegenwart
Robotics
Asymmetric Scout-Worker Reconnaissance for Route Validation in Unknown Environments
Kashif Khurshid Noori, Jaskrit Singh, Athanasios Ch. Kapoutsis et al. (+2 more)
This paper studies asymmetric scout-worker reconnaissance in unknown environments, where a small, agile autonomous scout explores routes for a larger worker
robot that must visit an ordered sequence of goal locations. Because the scout has a smaller footprint and greater mobility, a scout-traversable route may be infeasible for the worker; worker feasibility must therefore be inferred from scout observations. This setting is not explicitly addressed by existing exploration and replanning methods, which typically assume a single traversability model and seek optimal paths for the same
robot performing the exploration. We introduce a symbiotic scout-based framework that exploits the scout's superior mobility to explore only the portions of the unknown environment needed to identify worker-feasible path segments connecting the ordered goals. Evaluations in simulated and real-world settings demonstrate that the proposed approach validates feasible routes, repairs blocked segments with validated worker-feasible detours, and substantially reduces scout travel compared to baseline exploration and planning methods. A real-world indoor deployment further demonstrates the scout navigating narrow corridors to identify a worker-feasible route.
Read the full abstract ▾
All authors Kashif Khurshid Noori, Jaskrit Singh, Athanasios Ch. Kapoutsis, Jing Xiao, Constantinos Chamzas
Plasma Physics
Diffusion prior for KSTAR equilibrium reconstruction under sensor dropout
Hyungkeun Nam and Jaemin Seo
We study
equilibrium reconstruction for the Korea Superconducting Tokamak Advanced Research (KSTAR) device under
magnetic sensor dropout, with a
diffusion model as the prior.
Magnetic measurements leave most components of the toroidal current density $J_\phi$ undetermined, and sensor loss leaves more of them to the prior. The
diffusion model learns only $J_\phi$, conditioned on the coil currents, the
plasma current and the product of major radius and toroidal field, which do not depend on the dropout. The physics enters through a linear forward operator built from sensor response
matrices of the LIUQE code. Because the observations are linear, a variant of decoupled annealing posterior sampling fits them with a closed-form linear correction. On measured signals of 37 KSTAR shots, we switch off a random fraction (0.0 to 0.9, ten settings) of the 124
magnetic channels in use and compare with LIUQE under the same masks, taking the full-sensor LIUQE reconstruction as the label. Both methods fit the remaining channels to a similar level. The median relative distance of $J_\phi$ from the label stays at 3.22–4.07% for the
diffusion reconstruction in all settings, whereas that of LIUQE reaches 11.89% at dropout 0.9. At dropout 0.5–0.9, the
diffusion reconstruction is closer to the label on 27–37 of the 37 shots. At low dropout, LIUQE is closer on most shots. For the derived flux, the
diffusion reconstruction is closer on 21–36 shots at dropout 0.5–0.9, and its error at high dropout comes from the vessel current estimate, not from the prior.
Read the full abstract ▾
Computation and Language
When Updating Stops Being Learning: Rethinking LLM Self-Evolution via learnable information gain
Chenxu Wang, Chaozhuo Li, Xinze Shi et al. (+6 more)
Self-
evolution lets large
language models (LLMs) improve iteratively using their own generated data, but often suffers from self-
evolution degeneration: performance improves, plateaus, then declines. Existing methods address this issue at the component level, targeting either the Questioner or the Solver, and overlook that self-
evolution is a tightly coupled system. We propose a holistic framework based on learnable information gain, which measures how much novel, parameterizable information a round provides relative to the previous round. Theoretically, this gain equals the Kullback-Leibler divergence between the two rounds' data distributions plus their
entropy change. Practically, it is estimated by fitting a small
language model to the previous round and scoring new data via negative log-likelihood. Based on this diagnostic, we propose ATRI (Adaptive Training Regulation via Information-gain), which reweights samples within a round and halts training across rounds when information gain remains low. Experiments on popular
datasets demonstrate the superiority of our proposal.
Read the full abstract ▾
All authors Chenxu Wang, Chaozhuo Li, Xinze Shi, Songyang Liu, Kyrie You Wu, Ziluowen Luo, Shun Zhang, Chenxi Li, Litian Zhang
Computation and Language
Retrieval Sensitivity to Identity Signals in Queries
Andrew Tang, Nicholas Deas, Kathleen McKeown et al. (+1 more)
Dense retrievers decide which documents reach users and the
language models that use them, yet they are typically evaluated with neutral queries. We ask whether the identity signals that real users express in their queries—political ideology and dialect—bias what a retriever returns. We design evaluations in two domains, political news and consumer-health questions, each pairing a controlled synthetic set that varies only the identity signal with naturalistic queries. Across five dense retrievers and a sparse baseline, every retriever (i) retrieves articles that align with the query's own political lean and (ii) performs worse for questions written in African American
Language (AAL) than in White Mainstream English (WME). Two analyses tie these gaps to queries' identity signals beyond surface vocabulary: partialling out an aggregate lexical-asymmetry score leaves the synthetic gaps largely intact, and linear probes recover lean and dialect from the retrievers' query
embeddings beyond
token-level features. Left unaddressed, such
retrieval biases risk contributing to polarization and reinforcing the health disparities already faced by AAL speakers. Code is available at https://github.com/Andrewtcr/bias-ret.
Read the full abstract ▾
All authors Andrew Tang, Nicholas Deas, Kathleen McKeown, Vishal Misra
Accelerator Physics
Calculation of the temporal structure for a pulsed slow positron beam based on superconducting accelerator
H. Q. Zhang, P. Kuang, Y. H. Hu et al. (+4 more)
The development of superconducting accelerator (SCA) provides a novel approach to generating pulsed slow positron beams with high time resolution and high intensity. SCAs can produce electron beams with repetition frequencies on the order of MHz and pulse widths of less than 100 ps. A pulsed positron beam based on SCA can, to a certain extent, preserve the excellent temporal structure of the primary electron beam. However, thermalization,
diffusion, and surface re-emission of positrons in the moderator will inevitably lead to time broadening of positron pulses. In this paper, a calculation model coupling Geant4 Monte Carlo simulations with positron
diffusion theory is established to evaluate the effect of the positron moderation process on time broadening. After being moderated by a tungsten foil, the initial time broadening of the pulsed slow positron beam is approximately 320 ps. By combining the time broadening calculation of pulsed beam during its transport, it is expected that the pulsed slow positron beam generated by the SCA, can be directly applied to the measurement of positron annihilation lifetime.
Read the full abstract ▾
All authors H. Q. Zhang, P. Kuang, Y. H. Hu, F. Y. Liu, X. Z. Cao, B. Y. Wang, X. P. Li
Machine Learning (Statistics)
Hierarchical Utility Calibration for Structured Multiclass Decisions
Futoshi Futami, Jerry Huang and Ichiro Takeuchi
In multiclass probabilistic prediction, Utility Calibration (UC), which focuses
auditing on specified utilities, has recently received
attention as a way to guarantee downstream decisions while controlling computational and sample requirements. At the same time, some multiclass problems have meaningful label hierarchies that play important roles in medicine and image classification, yet how UC evaluates utility within a hierarchy remains insufficiently understood. We show that the difference between realized utility and predicted mean utility admits an exact decomposition into a sum of contributions from the internal nodes of the label tree. This decomposition shows that positive and negative contributions from different nodes can cancel, and that even when UC is small, the utility errors remaining in parts of the hierarchy need not be small. To address this problem, we propose Hierarchical Utility Calibration (HUC), which evaluates each node contribution before summation while retaining the same target utility, subgroup, and predicted-utility interval. We further provide finite-sample evaluation over all predicted-utility intervals and propose HUC-Boost, which updates only violated internal nodes, with theoretical guarantees for both.
Read the full abstract ▾
Computer Vision
Foresight at the Event Boundary: Evaluating Physical Prediction in Video World Models
Estela Monserrat Arriaga Santana, Julian Rosas Scull, Ehécatl Sacamch'en Núñez Rico et al. (+1 more)
Video world models are largely regarded as predictive models of the physical world and are therefore expected to anticipate the consequences of observed events. However, evaluation has mainly focused on reference similarity, physical-law consistency, or judgment plausibility, estimating anticipation only indirectly. We address this directly: when a release or impact has just occurred but its consequence is withheld, can a world model anticipate what should happen next? We introduce an event-anchored evaluation based on 62 controlled real-world free-fall recordings and 124 clips spanning three object types, with fine-grained release and impact annotations and ground-truth trajectories. The protocol separates consequence production, temporal placement, and physical realization. Across six contemporary
video generation and world models, Runway and Veo produce release and subsequent impact events at rates above 93% but often initiate them substantially late, whereas Cosmos-Predict-2.5 and MAGI-1 frequently preserve the pre-event state and produce little or no measurable consequence. Among measurable falls, plausible timing does not necessarily imply physically consistent motion. We further conduct a 15-participant, 20-condition human study in which participants describe the expected consequence from a single event-anchored frame and draw its trajectory. Human predictions favor the recorded future in aggregate while revealing genuine ambiguity among plausible continuations. Overall, physical foresight emerges as a sequence of distinct challenges: initiating a consequence, anchoring it in time, and realizing its motion.
Read the full abstract ▾
All authors Estela Monserrat Arriaga Santana, Julian Rosas Scull, Ehécatl Sacamch'en Núñez Rico, Hugo Jair Escalante
Robotics
Trajectory-Level Mode Guidance for Controllable Diffusion-Based Multi-Robot Motion Planning
Tianyou Yu, Shengze Cai and Chao Xu
Motion planning often admits multiple feasible solutions, making
multimodal generation valuable, particularly for flexible multi-
robot coordination.
Diffusion models naturally learn such trajectory distributions, yet incorporating coarse and partial trajectory priors without restricting generation remains challenging. Such priors indicate a desirable region of the solution space rather than a single solution, motivating conditioned generation that preserves multimodality. In this paper, we guide trajectory generation in the clean trajectory space and progressively incorporate trajectory priors with a timestep-dependent guidance strength. At each reverse
diffusion step, the reconstructed clean trajectory provides a unified space for integrating planning costs and partial trajectory priors. Planning costs are incorporated through gradient-based refinement, while the partial prior is progressively injected at the corresponding noise levels with decreasing guidance strength. This guides generation toward the prior in early stages while gradually releasing the constraint to preserve the inherent multimodality of the
diffusion model. The framework naturally extends to multi-
robot planning by incorporating inter-
robot collision costs. Experiments on single- and multi-
robot planning tasks demonstrate controllable trajectory synthesis, diverse feasible solutions, and safe multi-
agent coordination.
Read the full abstract ▾
Machine Learning
Triadic Linear Attention: Three-Dimensional Recurrent States for Long-Context Sequence Modeling
Oliver Sieberling, Bharat Runwal, David Jin et al. (+3 more)
Recurrent neural networks (RNNs) compress the historical context into a
memory state of fixed size, thus allowing for constant-time inference. The
memory state size is a crucial factor in their performance, as exemplified by the strong performance and resurgence of linear
attention, which extends the vector-valued hidden states of ordinary RNNs to
matrix-valued hidden states. Crucially, linear
attention does so in a parameter-efficient way, in particular by using an outer product of the key and value vectors to write to the
matrix-valued hidden state. We generalize this construction and propose triadic linear
attention, which writes the triadic outer product of a key, a second key, and a value, into a third-order (i.e., 3D) tensor state, and reads from it by contracting both key axes with two queries. An $E$-dimensional second key thus yields an $E$-fold increase in state size while adding only two projections. Triadic linear
attention is compatible with data-dependent forgetting, the delta rule, and chunkwise-parallel training. Applied to Gated DeltaNet and scalar-gated linear
attention, triadic linear
attention substantially improves long-context
language modeling and recall, outperforming alternatives that enlarge the state.
Read the full abstract ▾
All authors Oliver Sieberling, Bharat Runwal, David Jin, Ryan Chin, Rameswar Panda, Yoon Kim
Plasma Physics
Experimental demonstration of broadband-laser suppression of cross-beam energy transfer
Guoxiao Xu, Xiaoran Li, Ning Kang et al. (+22 more)
In direct-drive inertial confinement fusion (ICF), cross-beam energy transfer (CBET) redirects a significant fraction of the incident
laser energy out of the
plasma. We report the experimental demonstration that broadband
lasers reduce CBET-enhanced reflected-light return, performed at the low-coherence Kunwu
laser facility with two crossed beams of 0.6% bandwidth and up to 550J at $\sim$2.6$\times$10$^{14}$ W/cm$^{2}$. A coupled ray-tracing model shows that the strong CBET amplification of narrowband reflected light is much weaker under broadband illumination. In symmetric incidence condition of two orthogonal beams, the total fractional scattered energy decreases from 7.57% to 4.64%; in asymmetric incidence condition, which separates stimulated Brillouin scattering (SBS) from specular reflection, the SBS fraction decreases from 3.75% to 0.46% and the specular-reflection fraction decreases from approximately 2.2% to 1.3%. These results establish broadband
lasers as an effective, experimentally validated approach to reducing energy escape through CBET-enhanced reflection in direct-drive ICF.
Read the full abstract ▾
All authors Guoxiao Xu, Xiaoran Li, Ning Kang, Huiya Liu, Liang Hao, Zhiyuan Li, Guowei Yang, Honghai An, Xichen Zhou, Jian Wang, Lin Yi, Jun Xiong, Zhiyong Xie, Junjian Ye, Yuchun Tu, Zhiheng Fang, Guo Jia, Wei Wang, Lan Xia, Wei Feng, Lailin Ji, Xiaohui Zhao, Anle Lei, Lifeng Wang, Yanqi Gao
Machine Learning
SCOPE: Observation-Conditioned Full-Target Prediction for Sparse PDE Inference
Ruichen Xu, Siyao Wang, Fang Wan et al. (+8 more)
Recovering complete physical fields from sparse observations is challenging because the measurements may not uniquely determine the underlying state.
Diffusion-based PDE solvers address this problem through iterative sampling whereas neural operators provide deterministic one-pass predictions. We propose SCOPE (Sparse-Context Observability-aware Predictive
Embeddings) to recover complete PDE fields from sparse observations by coupling full-field latent prediction with physical reconstruction. A shared decoder reconstructs fields from both predicted and complete-view representations so that representation learning is guided by both physical recovery and latent matching. We derive a quadratic risk decomposition at fixed teacher-decoder pairs showing why optimal latent prediction need not yield optimal field reconstruction. We also establish sufficient conditions for decoder improvements on complete inputs to transfer to recovery from partial observations. Experiments across five PDE settings show that SCOPE outperforms mask-aware neural operators on all ten forward and inverse tasks and achieves lower errors than those reported for
diffusion-based solvers including DiffusionPDE and FunDPS. Decoder-only adaptation further improves recovery without retraining the backbone while retaining deterministic single-pass inference.
Read the full abstract ▾
All authors Ruichen Xu, Siyao Wang, Fang Wan, Jiacheng Qiu, Wenhan Gao, Jiaxing Zhang, Linsey Pang, Ravid Shwartz-Ziv, Prakhar Mehrotra, Yann LeCun, Yuefan Deng