Materials Science
Magneto-optical Kerr spectroscopy of exciton Rydberg states in a magnetic van der Waals heterostructure
Astha Khandelwal, Benran Zhang, Madhusmita Jena et al. (+8 more)
Excitonic states in two-dimensional
semiconductors are sensitive to time-reversal
symmetry breaking, yet how interfacial exchange acts on higher-lying
exciton Rydberg states remains largely unexplored. Here we show that wavelength-resolved magneto-optical Kerr spectroscopy of a proximity-coupled MoSe$_2$/Fe$_3$GaTe$_2$ van der Waals
heterostructure resolves
magnetic symmetry breaking across the A- and B-
exciton manifolds. Combining reflection, photoluminescence, and photoluminescence excitation spectroscopy with first-principles many-body calculations including screening by the metallic Fe$_3$GaTe$_2$ layer, we assign the strongly enhanced Kerr resonances to the ground and 2s states of both
manifolds. Reversing the
magnetic field inverts the polarity of every resonance. A field-odd/field-even decomposition confirms a
magnetic origin. The opposite Kerr polarities of the A- and B-
exciton manifolds can be explained by proximity-induced exchange coupling to the opposite valence-band
spins of the two
exciton series. Our results establish resonant Kerr spectroscopy as a sensitive probe of magnetically induced
symmetry breaking across the excitonic
manifold.
Read the full abstract ▾
All authors Astha Khandelwal, Benran Zhang, Madhusmita Jena, Zhenchao Wen, Kenji Watanabe, Takashi Taniguchi, Saroj P. Dash, Zhenglu Li, Ryo Kitaura, Bhagwati Prasad, Daichi Kozawa
Algebraic Geometry
Connectivity of the slice filtration
Dipankar Maity
Using methods similar to Morel's stable connectivity
theorem, we prove that, over an arbitrary field, the Tate truncation functors $f_{0/n}$ preserve motivic connectivity of $ S^1$-spectra. An analogous result for complexes yields a Hurewicz
theorem for the $L^{p,n}$- and $L_{bir}^n$-localizations over perfect fields. Over such fields, we establish a stronger connectivity property for categories of correspondences, showing that $f_n^\mathcal{C}$ preserves connectivity of motivic spectra with $\mathcal{C}$-transfers, whenever $\mathcal{C}$ satisfies cancellation. We then use the motivic reconstruction
theorem to deduce that $f_n$, and consequently $s_n$ and $f_{0/n}$, also preserve connectivity for effective (and thereby, $\mathbb{P}^1$-) motivic spectra. Along the way, we also establish the slice analog of the motivic (effective) reconstruction
theorem, as well as the slice analog of motivic $S^1$- and $\mathbb{P}^1$-recognition
theorems.
Read the full abstract ▾
Computation and Language
Papers Without Code: Availability of GitHub Repositories Linked in *CL Publications
Selina Meyer and Michael Roth
Source code and data published at computational linguistics (*CL) venues are increasingly being shared via GitHub. While this generally is a favourable development for the accessibility and potential reusability of research artifacts in natural
language processing (NLP), the long-term availability of such repositories has not been evaluated. In this squib, we discuss the availability of repositories linked in papers published in the Computational Linguistics (CL) journal as well as at ACL and its co-located events over the past ten years. Contrary to our expectations, we find that GitHub repositories linked in more recent ACL publications are unavailable at similar rates as in older publications, in parts due to an increase in empty and placeholder repositories. Similar trends hold for other *CL venues, but not for platforms other than GitHub.
Read the full abstract ▾
Mathematical Physics
Classical BV Pushforward
Xin Geng, Andrey Losev and Vyacheslav Lysov
We define the classical BV pushforward and establish its two main properties. First, we prove that it maps solutions of the classical master equation (CME) on the total space to solutions of the CME on the smaller space. Second, we prove that, for isolated critical points, it maps BV canonically equivalent solutions to BV canonically equivalent solutions. We illustrate the construction with examples and discuss its applications.
Read the full abstract ▾
Quantum Physics
Near-Optimal Bounds on the Density of Low-Energy States of $k$-Local Hamiltonians and Faster Quantum Algorithms
Sevag Gharibian, François Le Gall, Ranitha Mataraarachchi et al. (+1 more)
Low-energy estimation and state preparation for general $k$-local
Hamiltonians are fundamental challenges in
quantum complexity theory. Buhrman et al.~ [BGLGST, PRL 2025] recently broke the natural Grover bound $O^\ast(2^{n/2})$ for both problems, with the improvement depending on the relative accuracy $\varepsilon$ and the locality $k$. In this work, we present faster exponential
quantum algorithms for these problems, where the binary
entropy function governs the runtime exponent. For sufficiently small $\varepsilon/k$, our
algorithms improve the exponent by a factor of $\log(k/\varepsilon)$ over [BGLGST, PRL 2025]. Our main technical result is an
entropy-governed lower bound on the dimension of the
Hamiltonian's low-energy subspace, obtained by depolarizing its ground state. For fixed $k$, this bound is optimal up to constant factors in the exponent. The same framework yields tighter bounds for Heisenberg, $XY$, and Ising models on arbitrary interaction graphs.
Read the full abstract ▾
All authors Sevag Gharibian, François Le Gall, Ranitha Mataraarachchi, Suguru Tamaki
Systems and Control
Frequency Measurement Practices for Inertia Assessment in Inverter Based Resources Dominated Power Systems
Christos M. Nikolakakos, Hassan Haes Alhelou and Nikos Hatziargyriou
Transmission system operators and protective relays compute the rate of change of frequency (RoCoF) from frequency averaged over tens to hundreds of milliseconds, as standards and grid codes specify. Any response delivered inside that window lowers the measured RoCoF, so the inferred inertia depends on the window and the responses in service. This paper defines the windowed inertia estimate and derives an exact identity the system inertia divided by one minus the fraction of the disturbance energy delivered inside the window and a closed-form prediction that applies the same
estimator to the trajectory of a linear response model. Eleven IEEE 9-bus RMS runs synchronous, grid-following and grid-forming configurations, disturbance sizes and parameter sweeps are reproduced by that model, and the identity closes on the measured response powers. Fast frequency response inflates the estimate most: at 500 ms a grid-following system with 20 % less synchronous inertia reports the same value as the all-synchronous system, the grid-forming system a higher one. On a reduced Greek transmission system the 500ms estimate exceeds the inertia by 17-26%, and a scenario with 10% less synchronous inertia reports more than the all-synchronous scenario's inertia. Operators can thus compute what their measurement reports.
Read the full abstract ▾
Geophysics
A Review of Deep-learning-based Seismic Data Denoising and Its Promising Paradigm Shift to Foundation Models
Xintong Dong, Zhengyi Yuan, Changxin Wei et al. (+3 more)
Denoising is a long-standing and widely-concerned topic in seismic data processing, for it can significantly increase the signal-to-noise ratio of seismic data. Numerous deep learning (DL) methods have shown promising denoising performance, but most of them are task-specific and focus on a certain type of seismic background noise. The real condition that seismic
datasets are often contaminated by various types of noises motivates us to explore a well-generalized and versatile DL model for seismic data denoising. Recently, in the fields of computer
vision and natural
language processing, foundation models (FMs) pre-trained on vast
datasets demonstrate outstanding adaptability and generality across diverse downstream tasks. This paradigm offers a promising path to address the challenges faced by task-specific DL denoising models, such as poor generalization, retraining from scratch for different noise, and the lack of labeled data. We first provide a brief review of traditional seismic denoising methods, followed by a comprehensive review of existing DL-based denoising methods categorized by noise type. Furthermore, we conduct a case study on a dedicated seismic denoising foundation model termed SeisDeFM. This is the first study in geophysical research to develop and validate a seismic denoising foundation model on pre-stack gathers with diverse noise conditions. Experimental results demonstrate that, compared with task-specific DL baselines, SeisDeFM achieves superior denoising performance and cross-noise generalization by the advantages of sufficient pre-training and appropriate downstream adaptation, and it effectively preserves weak reflection events while suppressing complex noise in pre-stack seismic data.
Read the full abstract ▾
All authors Xintong Dong, Zhengyi Yuan, Changxin Wei, Wenshuo Yu, Shiqi Dong, Jun Lin
High Energy Physics (Theory)
Cylinder Defect Casimir Energy and Weyl Anomalies
Yang Zhou
We distinguish two Casimir observables associated with a conformal surface defect. A circular cylinder embedded in flat space has a logarithmic contribution to its Casimir energy, $-d_1\log(R/\epsilon)/(24R)$, fixed by the displacement anomaly. A great cylinder obtained by radial quantization of the ambient theory instead has Casimir energy $-b/(12R)$ in the conformal defect
vacuum with a fixed local anomaly convention. Free-field and
holographic calculations provide checks of the proposed relations in both
embeddings.
Read the full abstract ▾
Computer Vision
Preference-Guided Adaptation for Open-Vocabulary Semantic Segmentation via Prompt Disagreement
Hyun-Kurl Jang, Jihun Kim and Kuk-Jin Yoon
Open-vocabulary semantic
segmentation (OVSS) enables pixel-level prediction over arbitrary text-specified vocabularies and has shown strong generalization on common
benchmarks. However, OVSS performance often degrades in specialized domains such as medical imaging, remote sensing, and industrial inspection, where dense pixel-level masks for adaptation are costly to obtain and require domain-specific expertise. We propose a preference-guided adaptation framework that replaces dense mask supervision with binary preferences. We observe that different
prompt templates produce systematically different
segmentations for the same image, a phenomenon we call
prompt disagreement, and we repurpose it as a built-in source of preference supervision. Building on this, we mine localized preference queries from regions of high cross-template uncertainty, and adapt the OVSS model with Region-Localized Preference Optimization (RLPO) together with consistency regularization that stabilizes updates outside the queried region. Across extensive experiments on the MESS
benchmark, the proposed method achieves consistent gains across diverse OVSS backbones without any pixel-level annotation, and remains effective under noisy preferences. Our code is available at https://github.com/blue-531/pref-ovss.
Read the full abstract ▾
Computer Vision
RRG-SLAM: Real-time Reflection-aware Gaussian SLAM for Indoor Scenes
Yong Liu, Keyang Ye, Zhexi Peng et al. (+3 more)
We introduce the first real-time reflection-aware Gaussian SLAM system for indoor scenes. The system features a reflection-aware TSDF-Gaussian hybrid representation that explicitly separates diffuse scene appearance from reflection components. The base scene is modeled by a TSDF volume and a set of base Gaussians capturing geometry and diffuse appearance, while planar reflections are represented by reflection Gaussian groups associated with detected reflective planes. The rendering is performed in three passes: TSDF raycasting first yields surface color, depth, plane IDs and reflection masks; base Gaussians are then rendered order-independently with depth culling and combined with the TSDF output to form the base image; finally, under the guidance of the plane ID map, reflection Gaussians from different reflection groups are rasterized only into their corresponding planar regions to generate the reflection image, which is subsequently composited with the base image via the reflection mask to produce the final output. For online reconstruction, our system first estimates the camera pose through reflection-aware tracking to suppress interference of reflection-dominated regions. It then identifies reflective planes using
geometric, semantic, and temporal cues, and fuses the observations into the augmented TSDF volume with reflection-aware attributes. Afterwards the base and reflection Gaussians are initialized, optimized, and pruned online to maintain both reconstruction quality and efficiency. Experiments on a variety of
datasets show that our method outperforms existing SLAM systems in reconstruction quality, tracking robustness, and novel-view rendering for indoor environments with reflections, while preserving real-time performance.
Read the full abstract ▾
All authors Yong Liu, Keyang Ye, Zhexi Peng, Ruixian Mei, Kun Zhou, Tianjia Shao
Artificial Intelligence
PairPref: When Should Memory Guide the Answer? A Benchmark for Contextual Preference Use
Mingfei Lu, Mengjia Wu and Yi Zhang
Memory-augmented assistants use retrieved preferences to guide their responses. A small change in the situation can change whether a preference is appropriate while barely affecting its
retrieval similarity.
Memory benchmarks typically test whether systems store and retrieve preferences, with less
attention to when those preferences should apply. We introduce PairPref, a
benchmark of contextual preference use. Each pair changes only the situation, keeping the preference, request, and four candidate replies fixed. The preference remains valid in both situations. In the selection track, models must choose the reply that applies the preference only where appropriate. In the free-generation track, they must decide when to apply it without seeing candidate replies. Both tracks use the same 1,227 pairs across 45 preferences and eight situation categories. We evaluate eight models, most of which achieve selection scores ($\Delta$) of 51 to 65 points. In free generation, however, both responses are appropriate for their respective situations in only 3.6% to 18.3% of pairs. Models continue to apply the preference in both situations even with fewer retrieved
memories, alternative presentation formats, and a stricter
prompt. These results show that models still struggle to judge when user preferences apply and respond accordingly.
Read the full abstract ▾
Computer Vision
SAGE: Subspace Alignment for Classifier-Free Guidance in Mixture-of-Experts Diffusion Models
Boyu Zhang, Yangming Cheng, Ning Zhang et al. (+5 more)
Diffusion Transformers with Mixture-of-Experts (MoE) routing are a leading recipe for scaling generative models. Classifier-Free Guidance (CFG) is essential for generation quality, yet excessively high guidance scales trigger collapse. We identify a previously unreported failure mode in their combination: the two CFG branches route independently, so their realized activations occupy different subspaces. The unconditional write then leaves the conditional subspace, and CFG amplifies that residual linearly in the guidance scale. We propose SAGE, a training-time regularizer that aligns unconditional MoE activations to the conditional subspace without restricting routing diversity, at zero inference cost. Toy experiments show that SAGE dramatically suppresses extreme drift by 9.2x. When scaled to a 1B-parameter text-to-image model, SAGE significantly improves generation quality, delivering a 9.3% boost in peak DPG-Bench performance. Extensive experiments demonstrate that SAGE consistently outperforms the baseline.
Read the full abstract ▾
All authors Boyu Zhang, Yangming Cheng, Ning Zhang, Pengfei Liu, Weijie Li, Yifan Gao, Hangyu Li, Litong Gong
Audio and Speech Processing
Measurement-Based Bitrate-Energy-Quality Analysis of Neural Audio Codec Decoders on Laptop and Phone Platforms
Seunghyeon Shin and Seokjin Lee
Neural audio codecs can achieve similar objective quality at lower bitrates than conventional codecs, but their decoder-side computational cost may offset this bitrate advantage on battery-powered client devices. This paper presents a measurement-based rate–energy–quality analysis of four neural audio codecs–EnCodec, DAC, HILCodec, and SNAC–and two conventional baselines, AAC-LC and Opus, on laptop and phone platforms.
Speech and music are evaluated separately using the original-reference ViSQOL protocol. For the main comparison, operating points are matched by selecting the measured point nearest to the midpoint of the common overlap in treatment-level mean ViSQOL. A secondary analysis includes only explicitly evaluated bitrate settings. Decoder energy is measured as idle-subtracted energy per second of audio (J/s), and results are summarized as the median of three repeated runs for execution-valid runtime and device paths. Pairwise break-even transmission-energy thresholds are then derived analytically from the measured decoder energy and bitrate. At the matched-quality operating points, the evaluated neural codecs achieved similar ViSQOL scores at lower bitrates but generally required more decoder-side energy than the applicable conventional codecs. EnCodec produced the lowest neural break-even thresholds in both matched-quality cohorts on both platforms. By contrast, several DAC and SNAC comparisons on the Phone XNNPACK CPU path exceeded 200 mJ/kbit, and their full-band Phone configurations required more than 1 s to decode 1 s of audio. These results show that lower bitrate alone does not guarantee an energy benefit: deployment efficiency also depends on decoder complexity, the effective runtime and device mapping, execution validity, accelerator availability, and the transmission-energy coefficient.
Read the full abstract ▾
Earth and Planetary Astrophysics
Resolved Structure and Orbital Motion of a Localized 1 au-scale Dust Accumulation in the Protoplanetary Disk around TW Hya
Takashi Tsukagoshi, Takayuki Muto, Hideko Nomura et al. (+6 more)
We present the results of high-resolution ($\sim1$~au) Atacama Large Millimeter/submillimeter Array (ALMA) observations of the TW~Hya protoplanetary disk in the Band 6
dust continuum, as well as the \ce{^13CO} and \ce{C^18O} J=2–1 emission lines. The primary focus of this study is to investigate the kinematics, internal morphology, and local gas environment of the prominent
dust blob at a radius of 52~au. By comparing our 2021 data with archival observations from 2017, we detect the proper motion of the blob. The measured azimuthal velocity of 3.3$\pm$0.9~km~s$^{-1}$ is fully consistent with local Keplerian rotation. Combined with the lack of significant radial migration over the four-year baseline, this confirms that the structure is robustly co-moving with the disk system. Crucially, our high-resolution continuum map resolves the blob into a distinct double-peaked morphology separated by $\sim$1.7~au azimuthally. We robustly validate this double-peaked substructure by reproducing it in the independent 2017
dataset using a sparse-modeling image reconstruction technique. We discuss potential physical origins for this double-peaked morphology, including an inclined circumplanetary disk with an inner
dust cavity, the roots of
planet-induced spiral arms, or alternative hydrodynamic scenarios that do not involve an actively accreting
planet such as the U-turn trajectory of secondary
dust or a short-lived hydrodynamic gas
vortex. We detect no compact gas emission counterparts associated with the continuum blob. Since these CO lines likely trace optically thick upper atmospheric layers, the absence of localized vertical gas perturbations suggests that if an embedded
planet is responsible for the
dust structure, its mass must be exceptionally low.
Read the full abstract ▾
All authors Takashi Tsukagoshi, Takayuki Muto, Hideko Nomura, Ryohei Kawabe, Satoshi Okuzumi, Shigeru Ida, Jun Hashimoto, Taichi Uyama, Motohide Tamura
Geometric Topology
Exceptional pretzel knots are not algebraically slice
Suman Saurabh
We consider the three-strand pretzel
knot $P_a=P(a,-a-2,-(a+1)^2/2)$ and prove that, when $a\equiv1\pmod4$, its Alexander
polynomial fails the Fox–Milnor condition. The same remains true after adding any number of opposite odd parameter pairs $q,-q$ and reordering the parameters. Combined with work of Kim–Lee–Song, this gives the same conclusion for every odd $a\ge3$. In particular, no
knot in Lecuona's exceptional family is algebraically or topologically slice. Together with earlier work of Lecuona, Miller, and Kim–Lee–Song, this completes the slice–ribbon and topological-sliceness classifications for three-strand pretzel
knots and removes the nonexceptional hypothesis from the Lecuona–Wand classification of
prime fibered ribbon pretzel
knots.
Read the full abstract ▾
Statistical Methodology
Population-Level Decision Curve Analysis May Mislead the Evaluation of Prediction Model Usefulness under Subgroup Utility Heterogeneity
Junfeng Wang, Kim Zhipei Wang, Ben Van Calster et al. (+3 more)
Background Decision curve analysis (DCA) evaluates prediction model usefulness using net benefit (NB). Population-level NB is often interpreted as a proxy for population-level expected utility, but this assumes comparability of utility across subgroups. We aimed to characterize when subgroup utility heterogeneity may invalidate population-level DCA conclusions. Methods We compared prediction-driven treatment with default strategies in populations containing subgroups with different utility values. We derived an inconsistency region: combinations of subgroup-specific {\Delta}NB values for which population-level NB and utility favor different strategies. We also developed a practical robustness framework. Results Opposite signs of subgroup-specific {\Delta}NB provide a warning signal for possible inconsistency. The inconsistency region is larger when subgroup sizes are more similar and subgroup-specific \(a-c\) values are more different, where \(a-c\) is the incremental utility of a true positive relative to a false negative. When \(a-c\) differs across subgroups, population-level NB combines quantities on different implicit utility scales and may conflict with population utility. A real-world case study illustrates the problem. Conclusions Using population-level NB as a proxy for population utility implicitly assumes homogeneous \(a-c\) across subgroups. Subgroup DCA and our framework can identify and assess when utility heterogeneity may invalidate population-level conclusions.
Read the full abstract ▾
All authors Junfeng Wang, Kim Zhipei Wang, Ben Van Calster, Nan van Geloven, Ewout Steyerberg, Laure Wynants
Computer Vision
Evidence Before Accuracy: A MRI-PET Fusion Network for Alzheimer Disease Classification with Causal Regional Validation
Saeid Firouzi Daghigh and Saeed Ayat
Deep learning models for Alzheimer disease (AD) classification routinely report near-perfect discrimination, yet few are shown to rest on AD-relevant neurobiology rather than on
dataset artifacts, subject-level leakage, or non-
brain image content. We present a fusion network combining T1 MRI and FDG PET across axial, coronal, and sagittal planes, trained on ADNI consists of 554 paired subjects. The fusion model reaches AUC 0.962, accuracy 0.909, and F1 0.891, competitive with recent 3D CNN and
multimodal transformer systems at substantially lower cost. We first quantify how much modality, plane and slice geometry matter. A validation-only search over slice centres and neighbour spacings moves AUC by 0.180 for MRI and 0.078 for PET, selecting narrow spacing for MRI and wide spacing for PET, with the chosen coronal centres falling on the hippocampal body and on the posterior cingulate respectively. The contribution, however, is the evidence layer built around that number. Shortcut controls collapse the model to AUC 0.622 (silhouette), 0.608 (exterior), and 0.500 (blank), and a label-permutation null yields 0.456. Forward region-of-interest (ROI) ablation shows that masking medial temporal cortex in MRI and the posterior default-mode network (DMN) in PET produces the largest shift in the AD logit, while area-matched controls remain indistinguishable from that null. Reverse ROI ablation shows that the medial temporal lobe alone retains 89.2% of above-chance discrimination in MRI and the posterior DMN alone retains 79.0% in PET. A quantitative comparison of attribution methods shows occlusion sensitivity reaching 3.5-5.0* enrichment inside a priori AD regions against 0.10-0.43* in controls. Ablation and attribution independently establish a biologically correct double dissociation: hippocampal evidence is carried by MRI, posterior cingulate evidence by PET.
Read the full abstract ▾
Artificial Intelligence
EOPSA: Efficient On-Policy Self-Distilled Safety Alignment
Qirui Liu, Yichen Sun, Yan Wang et al. (+5 more)
On-Policy Self-
Distillation (OPSD) has emerged as a promising paradigm for safety
alignment, delivering dense,
token-level supervision by distilling from a teacher conditioned on refusal-oriented privileged
prompts. However, we reveal that this paradigm suffers from critical inefficiencies that degrade both training efficiency and general
reasoning capabilities. Specifically, we diagnose two fundamental bottlenecks: (1) supervisory collapse over extended rollouts, where the teacher's corrective efficacy degrades precipitously as the student's generation prefix lengthens, injecting noisy gradients into late-stage
tokens; and (2) gradient dilution from stylistic shifts, where the
distillation objective is dominated by safety-irrelevant stylistic discrepancies induced by privileged prompting, washing out genuine safety signals and impairing base
reasoning. To resolve these issues, we propose Efficient On-Policy Self-Distilled Safety
Alignment (EOPSA), which concentrates computational and gradient budgets exclusively on reliably supervised, safety-critical
tokens. EOPSA incorporates two coordinated mechanisms: (i) Adaptive Rollout Scheduling, which dynamically bounds the generation horizon guided by a novel Teacher Rescue Rate (TRR) metric to operate strictly within reliable supervision regimes; and (ii) Selective
Distillation, which filters out safety-neutral
tokens to restrict gradient updates exclusively to safety-pivotal transitions. Extensive evaluations across
reasoning models up to 32B parameters demonstrate that EOPSA slashes rollout computation by $\sim$50% and backpropagates through merely $\sim$2% of
tokens, consistently outperforming full-
token distillation baselines in both safety compliance and
reasoning retention.
Read the full abstract ▾
All authors Qirui Liu, Yichen Sun, Yan Wang, Yu Mi, Wei Cao, Yue Shen, Zhixuan Chu, Kui Ren
Cryptography and Security
SEAD: A State-Based Perspective on Attack and Defense in Tool-Using Agents
Xinjie Shen, Junran Wang, Rongzhe Wei et al. (+1 more)
Language-model
agents increasingly use tools to act on external systems. Earlier actions can alter files, permissions, database records, or other state, making a later routine-looking action harmful. Yet the visible interaction may not reveal the underlying state needed to assess that action. We formulate attack and defense as partially observed state control in SEAD, deriving their design requirements from this shared execution process. Because attackers supply instructions while the target chooses concrete actions, DART decomposes harmful goals into locally plausible steps and uses feedback from actual tool execution to guide trajectory search. The defender must decide before execution with incomplete state evidence. SAGE can therefore investigate relevant state through read-only queries before allowing or blocking each action, including those proposed after a block. We construct an environment-verifiable
dataset integrating controlled initial states, replayable tool environments, and task-specific executable checks. Across four target models, DART improves semantic attack success by 18.8–35.9 percentage points over the competing baseline, with consistent gains under executable verification. On recorded trajectories, SAGE preserves 95.79% of benign trajectories while intercepting 92.73% of harmful paths by the harm-enabling boundary. In online attack-defense evaluation, it reduces DART's executable attack success from 48.0% to 4.0%. SAGE remains effective across four attack methods and generalizes to out-of-domain environments. Our code and data is available at https://github.com/EverywhereSafety/SEAD.
Read the full abstract ▾
All authors Xinjie Shen, Junran Wang, Rongzhe Wei, Pan Li
General Topology
Baire 1 functions and the strong Choquet property
Ľubica Holá
Let X be a Tychonoff space and B_1(X) be the space of real-valued Baire 1 functions. Using the classical Lebesgue's characterization of Baire 1 functions and the strong Choquet game we prove that if X is locally compact or a topological sum of sigma-compact spaces, the space B_1(X) equipped with the
topology \tau_UC of uniform convergence on compacta, is a strong Choquet space. For the Sorgenfrey line S, the space (B_1(S); \tau_UC) is of the first Baire category in itself.
Read the full abstract ▾
Classical Analysis and ODEs
On the Nazarov–Shcheglova Conjecture for Sharp Sobolev Inequalities: The Case (n,p)=(3,2)
Mohamed Jleli and Bessem Samet
For integers $n>k\geq0$ and $1\leq p,q\leq\infty$, let $\lambda_3(n,k,p,q)$ denote the optimal constant in the one-dimensional Sobolev inequality \[ \|u^{(k)}\|_{L^q(0,1)} \leq \lambda_3(n,k,p,q) \|u^{(n)}\|_{L^p(0,1)}, \qquad u\in\mathring W_p^n(0,1). \] Nazarov and Shcheglova \cite{NazarovShcheglova} conjectured that \[ \lambda_3(n,1,p,1) = 2\lambda_3(n,0,p,\infty), \qquad n\geq2,\quad 1\leq p\leq\infty, \] and that the corresponding extremal functions coincide and are symmetric about the midpoint of the interval. We prove this
conjecture for $(n,p)=(3,2)$ and, in particular, obtain \[ \lambda_3(3,1,2,1) = \frac{1}{32\sqrt5}. \] The proof is based on a reduction to an operator norm problem with an additional moment constraint. After rescaling to $(-1,1)$, this constraint leads to orthogonality with respect to quadratic
polynomials. We extend the inverse of the second derivative through the corresponding orthogonal projection and analyze the adjoint operator by an odd–even decomposition. The odd component is controlled by a positive Gram kernel, while the even component is treated by a weak-$L^2$ estimate. The equality cases yield the characterization of all extremal functions.
Read the full abstract ▾
Game Theory
Nash Equilibria in Auctions with Pacing Strategies: Complexity and Inefficiency
Aris Filos-Ratsikas, Charalampos Kokkalis and Mohamad Latifian
We introduce and study Auctions with Pacing Strategies (APS) games, a full-information model in which utility-maximizing bidders compete across many simultaneous first-price auctions, each choosing a single pacing multiplier that uniformly scales their values into bids. We settle three central questions. First, we show that there are instances that admit no approximate pure Nash
equilibria. Then, we prove that the problem of deciding whether an APS game admits an (approximate)
equilibrium is NP-complete in general, but can be solved in
polynomial time if either the number of bidders or the number of items is fixed. Finally, when an
equilibrium does exist, we characterize its inefficiency exactly, showing that both the Price of Anarchy and the Price of Stability equal $\frac{e}{e-1}$.
Read the full abstract ▾
Cryptography and Security
How to Tame a Multi-Headed Hydra? Adaptive Multi-Category Safety Steering for Large Language Models
Chenxi Wang, Ruiyang Huang, Li Huang et al. (+1 more)
As large
language models (LLMs) become increasingly widespread, preventing unsafe responses to harmful
prompts is essential for their safe deployment. Activation steering offers an approach to improving LLM safety by modifying internal activations during inference without updating model parameters. However, a single
prompt can involve multiple harm categories, and steering toward safety in one category may leave harmful content from another unaddressed. Despite advances in adaptive steering, existing methods do not explicitly coordinate steering direction and strength when multiple harm categories co-occur within a single
prompt. To address this problem, we propose CAM-Steer, a Category-Adaptive Multi-category Safety Steering framework. Specifically, it estimates the risk associated with each harm category by comparing the current hidden state with safe and unsafe prototypes. The estimated risks are then used to combine the safety directions for different harm categories into a single steering direction and to determine the strength of the intervention. Finally, it rotates the hidden state along the composed steering direction, with the rotation angle determined by the estimated risks, while preserving the hidden-state norm. Experiments across three LLM backbones and seven harm categories show that CAM-Steer outperforms the evaluated baselines in average defense success rate, including when categories co-occur. Further analyses support its component designs and informative risk scores, with negligible inference overhead.
Read the full abstract ▾
All authors Chenxi Wang, Ruiyang Huang, Li Huang, Yifan Wu
Computer Vision
XFlow: A Workflow Model for Instruction-Guided Lesion Segmentation in Chest X-rays
Geon Choi, Hangyul Yoon, Hyunki Park et al. (+2 more)
Existing text-guided
segmentation models in the medical domain cover only a narrow set of anatomical structures and lesions in chest X-rays (CXRs), and most of them assume that the queried target is always present in the image. Instruction-guided lesion
segmentation (ILS) was introduced to overcome these limitations by segmenting diverse lesion types from simple user instructions while also recognizing when the queried lesion is absent, and ROSALIA was proposed as the first model for this task. However, the masks produced by ROSALIA remain of limited quality, often carrying scattered noise. Moreover, ROSALIA predicts the mask in a single shot, which differs fundamentally from how radiologists perceive and delineate lesions in practice. A radiologist first surveys the entire thorax, then localizes the approximate region of abnormality, and only then refines the lesion contour. Motivated by this coarse-to-fine, multi-level perception process, we present XFlow, a workflow model for ILS that combines box-based localization with multi-turn point refinement. XFlow detects the lungs, decides whether the queried finding is present in each of them, and
prompts a fine-tuned SAM with the lesion box for an initial mask. It then corrects that mask through point
prompts until its boundary follows the lesion, leaving every intermediate decision visible. Our experiments show that XFlow achieves the best
segmentation quality on both internal and external evaluation. Notably, it surpasses ROSALIA in
segmentation quality even when the two are trained on the same lesion annotations. Code and model weights will be made publicly available.
Read the full abstract ▾
All authors Geon Choi, Hangyul Yoon, Hyunki Park, Sang Hoon Seo, Edward Choi
Robotics
MonoEgo: Monocular Metric Egocentric Demonstration Capture with Passive Wrist Constellations and Sparse Workstation Anchors
Jie Xu, Kangjin Yu, Ziyi Jin et al. (+2 more)
Image-aligned metric demonstrations often require dedicated tracking hardware and synchronization across devices. We present MonoEgo, a capture system that replaces active wrist instrumentation with offline
monocular reconstruction. One 90-FPS global-shutter camera observes calibrated passive wrist constellations, sparse workstation anchors, and the scene on a shared image clock. MonoTag SLAM combines marker corners with ORB geometry and uses visual evidence to reject ambiguous planar-marker poses. Its metric Atlas supports interval scale re-anchoring, verified map merging, and retrospective localization of earlier frames supported by the final map. Camera and wrist-constellation outputs retain validity and map provenance, and unsupported motion is left missing. Experiments show metric tracking beyond continuous anchor visibility, reconnection of supported map components, and recovery of some missing camera poses. Comparisons against a multisensor camera reference and separate stationary-constellation tests characterize trajectory agreement and precision while revealing incomplete coverage and residual
geometric uncertainty. The results indicate that passive fixtures and offline reconstruction can reduce capture-side requirements. Dynamic accuracy, deployment, and downstream policy benefits require further study.
Read the full abstract ▾
All authors Jie Xu, Kangjin Yu, Ziyi Jin, Beichen Wang, Zhongpu Xia
Machine Learning
HALO: Enhancing Time Series Generation via Hyperspherical Latents and Masked AutoregRessive Modeling
Chunyi Hou, Xiangfei Qiu, Hanyin Cheng et al. (+2 more)
Most existing time series generators rely on a two-stage modeling paradigm: the first stage learns discrete latent representations of time series; the second stage performs autoregressive modeling on these discrete latents through next
token prediction. However, this paradigm suffers from two stage-specific limitations: the first stage can lead to information loss when discretizing continuous time series, while the second stage is prone to error accumulation during autoregressive generation. To address these limitations, our core idea is to perform generative modeling in a continuous latent space with a more efficient autoregressive framework. We propose
HALO, which enhances time series generation via Hyperspherical Latents and Masked Autoregressive modeling to achieve this goal by tackling two key bottlenecks: (1) variance and scale heterogeneity of continuous latent representations; (2) the difficulty of balancing generation efficiency with temporal correlation modeling.
HALO first introduces a hyperspherical VAE that constrains continuous latents to a fixed-radius hyperspherical shell, effectively stabilizing the numerical fluctuations of continuous latent representations. Secondly, we develop a masked autoregressive model that balances parallel decoding and temporal correlation learning, substantially reducing the number of inference steps required for generation and improving generation stability. Our extensive experiments demonstrate that
HALO achieves state-of-the-art generation performance while offering significantly improved inference efficiency over existing advanced baselines.
Read the full abstract ▾
All authors Chunyi Hou, Xiangfei Qiu, Hanyin Cheng, Yutong Li, Bin Yang
Artificial Intelligence
Can AI Make Money in Crypto? Measuring the Gap from Backtests to Real Markets
Xingtong Yu, Jiarun Zhou, Guanlin Ding et al. (+12 more)
AI-based trading methods have rapidly evolved from machine learning and reinforcement learning to large
language models (LLMs) and trading
agents, yet their performance is still predominantly assessed through historical backtesting. Such evaluations provide limited evidence of whether a method can generalize to unseen future
markets or whether its backtested performance can be sustained in realistic trading frictions (e.g., latency, slippage, liquidity constraints, and
market impact). We present a unified
benchmark that evaluates representative machine learning, reinforcement learning, LLM-based, and
agent-based trading methods in cryptocurrency
markets through three progressively more realistic stages: historical backtesting, prospective exchange-based paper trading, and real-money live trading. These stages jointly increase temporal realism by moving from historical to unseen future
markets, and execution realism by moving from offline simulation toward live trading. This protocol enables us to quantify the backtest-to-realization gap, identify when performance begins to deteriorate, and compare how this gap differs across major classes of AI trading methods. We further provide a unified open-source system supporting all three evaluation stages, together with a public platform that continuously updates
benchmark results. Code is available at https://github.com/Starlien95/Awesome-TradingAI.
Read the full abstract ▾
All authors Xingtong Yu, Jiarun Zhou, Guanlin Ding, Wenkang Wei, Jiarui Liu, Chang Zhou, Fangzhou Ge, Chenyi Xu, Xikun Zhang, Renqiang Luo, Jie Zhang, Hong Cheng, Xinming Zhang, Hui Zhang, Yuan Fang
Machine Learning
Low-Confidence Remasking Traps Flexibility: Realizing Arbitrary-Order Potential for Diverse Rollouts in Diffusion LLMs
Moongyu Jeon, Dongjae Jeon, Bumjun Kim et al. (+2 more)
Masked
diffusion language models support arbitrary-order generation, suggesting a natural way to produce diverse outputs. However, recent work argues that this flexibility reduces diversity by delaying high-uncertainty
tokens that can lead to different generation paths. We trace this diversity loss not to arbitrary-order generation itself, but largely to low-confidence remasking (LCR), a widely used decoding rule. At each step, LCR samples a
token at every masked position but commits only the sampled
token with the highest probability, filtering out the rest. We show that this mechanism can exponentially suppress lower-probability
tokens as more positions compete, and observe the same suppression in LLaDA. In contrast, top-probability position selection (TPP), which has often been conflated with LCR under the shared label confidence-based decoding, avoids this diversity loss. TPP first selects the position whose most likely
token has the highest probability, then samples directly from that position's distribution. Replacing LCR with TPP restores diversity and yields Pass@$k$ comparable to left-to-right decoding, suggesting that the reported diversity loss stems largely from LCR's filtering rather than from generating high-confidence positions first. To further exploit order flexibility, we introduce
Entropy-Guided Initialization (EGI), which samples the first
token at the highest-
entropy position and then follows TPP. This simple modification further improves rollout diversity and solution coverage beyond left-to-right decoding, with gains extending to downstream policy optimization, highlighting the potential of arbitrary-order generation for diverse rollouts.
Read the full abstract ▾
All authors Moongyu Jeon, Dongjae Jeon, Bumjun Kim, Mingyu Kim, Albert No
Cosmology and Nongalactic Astrophysics
Field-level inference of growth rate from DESI BGS galaxy and ZTF supernova simulations
Corentin Ravoux, Bastien Carreres, Julian Bautista et al. (+60 more)
Galaxy peculiar velocities are a powerful probe of the growth rate of structure $f$ at low
redshifts ($z < 0.2$), measured through the combination $f\sigma_8$. These constraints are improved by combining peculiar velocities with
redshift space distortion measurements from the
galaxy density field. We create a dedicated simulation to test growth rate measurement from the combination of Type Ia
supernova peculiar velocities from the Zwicky Transient Facility (ZTF) with the Bright
Galaxy Survey (BGS) of the Dark Energy Spectroscopic Instrument (DESI), using projections for 6 years of ZTF data and the Data Release 3 (DR3) of DESI. We apply a wide-angle likelihood-based field-level inference to combine ZTF velocity and DESI BGS density fields simultaneously. This procedure is performed while varying the different density and velocity field configurations. Adding DESI
galaxy density information to the ZTF
supernovae improves the constraint on nuisance parameters and yields up to a 50 % reduction in the growth-rate uncertainty without introducing bias. If we consider the conservatively best DESI density configuration, the mean error bar achievable by a DESI ZTF measurement is $\sigma_{f\sigma_8} = {}^{+0.047}_{-0.048}$ corresponding to a 10 % error. These measurements will play a critical role for testing modified
gravity theories.
Read the full abstract ▾
All authors Corentin Ravoux, Bastien Carreres, Julian Bautista, Damiano Rosselli, Jessica Aguilar, Steven Ahlen, Melissa Amenouche, Uendert Andrade, Eric Armengaud, Alejandro Aviles, Florian Beutler, Davide Bianchi, David Brooks, Umut Burgaz, Aurelio Carnero Rosell, Edmond Chaussidon, Todd Claybaugh, Andrei Cuceu, Axel de la Macorra, Georgios Dimitriadis, Andreu Font-Ribera, Daniel Felipe Forero Sánchez, Jaime E. Forero-Romero, Enrique Gaztañaga, Satya Gontcho A Gontcho, Ariel Goobar, Gaston Gutierrez, Hiram K. Herrera-Alcantar, Cullan Howlett, Mustapha Ishak, Joel Johansson, Stephanie Juneau, Ofer Lahav, Martin Landriau, Laurent Le Guillou, Alexie Leauthaud, Marc Manera, Ramon Miquel, Seshadri Nadathur, Jakob Nordin, Hernan Enrique Noriega, Enrique Paillas, Nathalie Palanque-Delabrouille, Will Percival, Francisco Prada, Ignasi Pérez-Ràfols, Mickael Rigault, Antoine Rocher, Philippe Rosnet, Graziano Rossi, Rossana Ruggeri, Lado Samushia, Eusebio Sanchez, Christoph Saulder, David Schlegel, Michael Schubnell, Hee-Jong Seo, Joseph Silber, Małgorzata Siudek, Mathew Smith, Gregory Tarlé, Benjamin A. Weaver, Pauline Zarrouk
Machine Learning
KiT: A Foundation Model for Financial Time-Series Forecasting using DiffusionTransformers
Boyu Zhang and Haorui Li
Financial candlestick
forecasting is fundamental to quantitative investment, yet it remains exceptionally challenging due to extremely low signal-to-noise ratios and vast heterogeneity across
markets and instruments. Existing approaches have largely attempted to introduce deep learning to capture hidden temporal features, but most adopt an auto-regressive formulation, which leads to error accumulation during inference. Meanwhile, general-purpose time-series foundation models are not tailored to the unique structure of k-line data and yield unsatisfactory performance on downstream candlestick
forecasting tasks. To tackle these problems, we introduce KiT, a K-line
Diffusion Transformer foundation model, and reformulate future prediction as conditional path generation via flow matching: given a historical context window, the model generates an ensemble of plausible future OHLCV trajectories. We pre-train KiT at multiple parameter scales on billions of candlestick bars spanning multiple
markets and timescales. Across three
markets and seven resolutions, KiT attains a mean return RankIC of 0.057 and a mean
volatility RankIC of 0.66, leading at every timescale and outperforming both task-specific financial forecasters and general time-series foundation models. Code will be available at: https://github.com/Luciferbobo/KiT.
Read the full abstract ▾