Statistical Methodology
Subdata Selection for Measurement-Constrained Linear Regression under Model Misspecification
Yundi Kong, Min Yang and Ping-Shou Zhong
In many modern applications, obtaining response variables is costly, time-consuming, or technically challenging, making it impractical to measure outcomes for all available units. In such settings,
covariates may be available for a large pool of candidates, but responses can only be collected for a small, carefully selected subset, making the choice of which units to measure a consequential decision. Most existing subsampling methods are developed for massive
datasets, emphasizing the trade-off between statistical accuracy and computational cost, and typically relying on the assumption of a correctly specified model. These features limit their suitability in measurement-constrained settings, where statistical efficiency is paramount and model misspecification is often unavoidable. To address these challenges, we develop a new subset selection framework for linear regression that accommodates potential model misspecification. The proposed approach targets predictive expected loss and yields an A-optimality-type criterion that depends only on
covariate information, requiring no response observations at the selection stage. We refer to this method as Expected Loss Optimal Selection (ELOS). By focusing on predictive risk rather than model-based optimality, ELOS remains well-defined and robust under misspecification. Simulation studies and real data applications demonstrate that ELOS achieves strong predictive performance and compares favorably with existing methods across a range of scenarios.
Read the full abstract ▾
Statistics Theory
On Local Inference for Functionals of Ill-Posed Inverse Problems
Ellen Graham and Andrea Rotnitzky
We study inference for continuous linear functionals of solutions to ill-posed inverse problems, including Euclidean parameters arising in nonparametric instrumental variables, proximal causal inference, and shadow-variable models. Although pointwise root-$n$ inference is possible under standard conditions, we show that this need not translate into locally reliable inference. Under a model imposing only minimal regularity conditions, the target is Hellinger discontinuous when both inverse problems are ill-posed: arbitrarily small changes in the data-generating law can produce arbitrarily large changes in the target, ruling out informative locally uniform inference. Imposing a uniform bound on a minimum-norm solution restores Hellinger continuity under regularity conditions, but does not by itself ensure stable root-n inference. If, in addition, both inverse problems are ill-posed and the solution whose norm is uniformly bounded is nonunique at a law $P_0$, then under regularity conditions the target fails to be pathwise differentiable; consequently, asymptotically linear
estimators exhibit nonvanishing root-$n$ bias under some local alternatives, causing Wald confidence intervals to undercover asymptotically. Under regularity conditions, pathwise differentiability is restored if the uniformly bounded solution is unique at $P_0$. It is also restored, regardless of uniqueness, under the stronger requirement that the corresponding solution satisfy a uniformly bounded $\beta$-source condition for some $\beta$>0. We relate these regimes to very weak, weak, and strong identification in instrumental variables models.
Read the full abstract ▾
Machine Learning
Where to Start, How to Move: Analyzing Curricula under GRPO for Long-Horizon LLM Agents
Minchan Kim, Sungho Bae, Youngjin Kim et al. (+1 more)
Group Relative Policy Optimization (GRPO) provides no reward-relative learning signal when all rollouts receive identical rewards. In long-horizon tasks that require multiple interactions, rewards are given only upon successful task completion, making all-fail groups common early in training and limiting opportunities to practice later steps. Guide Assisted GRPO replays a successful reference prefix before the learner continues. We study how initial positions, handoff units, and movement rules affect unassisted task completion. Across three model sizes on BFCL, starting near completion is more effective under a fixed turn-level threshold rule. Finer handoffs alone do not consistently help, but outcome-informed tool-call selection outperforms GRPO on trained and untrained task categories. Guided starts also improve AppWorld success on both evaluation splits. These findings highlight where learning starts and how starting positions move as key curriculum choices.
Read the full abstract ▾
All authors Minchan Kim, Sungho Bae, Youngjin Kim, Hyunwoo Park
Solar and Stellar Astrophysics
Magnetic Reconnection Governs the Entire Lifecycle of Solar Eruptive Magnetic Flux Rope
Xinkai Bian, Chaowei Jiang, Qingjun Liu et al. (+3 more)
While
magnetic flux ropes (MFRs) are widely accepted to exist prior to many solar eruptions, the physical mechanism responsible for triggering and subsequently driving their eruptions remains debated. Classical models commonly attribute eruption onset to ideal magnetohydrodynamic instabilities, particularly the
torus instability, while
magnetic reconnection is often regarded as a secondary process. Using high-resolution three-dimensional MHD simulations, we investigate the eruption of an MFR that forms self-consistently through flux cancellation driven by converging motions and photospheric
diffusion following an initial shearing phase. The pre-eruption MFR rises quasi-statically and reaches a height where the decay index exceeds the classical
torus instability threshold, yet the system continues to evolve quasi-statically without eruption. Instead, eruption begins only when the current sheet beneath the MFR becomes sufficiently thin for
magnetic reconnection to set in. Increasing numerical resolution systematically delays the eruption onset, whereas the decay index at eruption onset remains nearly unchanged, demonstrating that the onset is governed by reconnection rather than
torus instability. During the eruption,
magnetic reconnection continuously converts surrounding sheared arcades into newly reconnected twisted flux that merges with the pre-eruption MFR and continuously to build the erupting MFR. A component-resolved analysis of the Lorentz force work further shows that the newly reconnection flux contributes predominantly to the main acceleration phase. These results demonstrate that
magnetic reconnection plays a central role throughout the entire lifecycle of the eruptive MFR: it builds the MFR, directly triggers its eruption, and provides the dominant driving of the eruption.
Read the full abstract ▾
All authors Xinkai Bian, Chaowei Jiang, Qingjun Liu, Xueshang Feng, Pingbing Zuo, Yi Wang
Machine Learning
Integrated Novelty Detection Systems: From Sensor Networks to Multi-Novelty Computer Vision
Irené Tematelewo
Novelty detection is crucial in various technological and scientific domains. Its importance spans from ensuring the reliability of sensor networks to enhancing the adaptability of computer
vision systems. As AI systems become increasingly sophisticated and complex, effective anomaly detection is essential for maintaining reliability and accuracy. This work addresses novelty detection in weather sensor networks and computer
vision. In weather sensor networks, particularly for monitoring rainfall, detecting blocked rain gauges can significantly impact data reliability, food security, and disaster management. This work evaluates four existing quality control methods for detecting blocked rain gauges and demonstrates their limitations, particularly in detecting clusters of faulty sensors. By comparing methods such as the Conditional Mixture Model (CMM), PSWQ (Fz-filter), Kernel Regression, and Gaussian Method, we identify their shortcomings and propose three novel quality control techniques that integrate probabilistic fault detection and consistency-based diagnosis framework to improve timely detection of blocked sensors, while reducing false alarms. In the realm of computer
vision, current systems often fail to handle multiple types of novelty due to their focus on single-novelty scenarios. We highlight the need for integrated systems capable of handling multiple types of novelty. To support this, we introduce the Novel Snapshot Serengeti (NOVEL-SS)
dataset, derived from the Snapshot Serengeti camera trap data, which includes various forms of novelty. We also present the DCA (Detect, Characterize, and Accommodate) baseline system, designed to handle multiple novelty by detecting, characterizing, and accommodating them. This work aims to advance the robustness and adaptability of novelty detection systems across both weather sensor networks and computer
vision applications.
Read the full abstract ▾
Chemical Physics
Combined Absorption and Emission Study of Air Species in NASA EAST
Jesse W. Streicher, Efaine Chang, Dylan L. Drescher et al. (+5 more)
A suite of tunable diode
laser absorption spectroscopy (TDLAS) diagnostics targeting rovibrational states of nitric oxide (NO) in the mid-infrared (MIR) and electronic states of atomic oxygen (O) and nitrogen (N) in the near-IR (NIR) was combined with optical emission spectroscopy (OES) for overlapping wavelengths at the NASA Electric Arc Shock Tube (EAST) facility for three selected high-temperature air experiments. Absorption features were probed at rates up to 1 MHz to provide quantitative, time-resolved measurements of temperature and species number density profiles behind incident shocks, enabling comparisons to spatially-resolved emission measurements, with the two typically agreeing within their respective uncertainties. Inferred time-histories are further compared to incident shock simulations from two one-dimensional models - one leveraging the Park two-temperature rates and the other more recent rate values - as well as the Data Parallel Line Relaxation (DPLR) code. For the two low-velocity cases (3.81 and 3.98 km/s) in which NO is probed, inferred temperatures are typically well-captured by all three models, although the NO number densities show stronger variation, agreeing better for the model with updated NO formation and dissociation rates. Finally, one high-velocity case (7.03 km/s) demonstrates the simultaneous resolution of three O-atom and six N-atom electronic states, with all states showing a steady increase in number density over time - possibly due to shock trajectory effects. Together, this combined absorption and emission study demonstrates resolution of many states of nonequilibrium air species, while directly comparing ground- and excited-state populations via the combined diagnostic strategy.
Read the full abstract ▾
All authors Jesse W. Streicher, Efaine Chang, Dylan L. Drescher, Christopher L. Strand, Ronald K. Hanson, Andrea Fagnani, Augustin C. Tibere-Inglesse, Brett A. Cruden
Artificial Intelligence
When Is Smaller Greener? Right-Sizing Language Models for Agentic AI
Sajid Siraj and Maria Barbati
AI
agents use large
language models (LLMs) to plan and act over multiple steps, yet models are usually compared per call rather than per completed task. This paper asks: which model should an
agent use to complete tasks reliably while minimising energy, cost and time? We show that per-call comparisons can be misleading for agentic AI. Because a task succeeds only if all required steps succeed, small differences in per-step reliability can produce large differences in overall task success. Failed attempts also consume resources. We therefore evaluate models using energy, cost and time per successfully completed task, treating the completed task as the functional unit. On this basis, we develop a decision model for choosing LLMs in agentic workflows. The model applies to tasks with binary or graded outcomes and to settings with or without retries. We show that each resource has a crossover task length at which a smaller, less reliable model ceases to be the greener, cheaper or faster option. These crossovers divide tasks into three regimes: for short tasks smaller models are preferred, for long tasks larger models are preferred, and only between these extremes do preferences among objectives matter. We embed these results in a two-stage framework that separates efficiency from preference. A Pareto screen first removes models that are never competitive; a weighted decision is required only when a genuine trade-off remains. A pilot study of 15 open-weight models from five developers, ranging from 0.6B to 27B parameters, on two multi-step
agent tasks with measured GPU energy supports the predicted structure. The model with the lowest energy per completed task changes with task length and task type, and observed crossover points generally match those predicted from per-step measurements. These findings provide a practical framework for right-sizing
language models in agentic AI.
Read the full abstract ▾
Computation and Language
Target-Token Asymmetry Can Masquerade As Depth Specialization
Danlin Li, Miao Li, Xu Han et al. (+1 more)
Layer-ablation studies often perturb early or late layers of a
language model, compare the damage to prediction on two corpora, and read a larger early-layer effect on one corpus as early-layer specialization for its task. Such a comparison averages over the
tokens that each corpus asks the model to predict, and these target
tokens differ: math solutions, for example, ask for far more whitespace and numbers than encyclopedia prose. We show that this difference can create the appearance of depth specialization. We split a cross-corpus early–late damage interaction by target-
token type and recompute it under a target mixture shared by both corpora. Between GSM8K and Wikitext, a positive interaction replicates across five model instances. In the three fully decomposed models, only whitespace has a positive interaction. Removing whitespace reverses the pooled effect in all three, and matching the full target mixture reverses it in two. The diagnosis holds under position matching,
adversarial reassignment of
token types, and a continuous readout. The risk can also be anticipated from the tokenizer alone. A screen computed before any model runs places every held-out pair it scored in the region its label predicts, and on the pairs it rates at high or medium risk, removing whitespace reverses every positive pooled estimate. A smaller content-word effect one
token ahead follows the share of content targets that the unablated model already predicts. A cross-corpus depth effect therefore supports an interpretation of task specialization only after the target-
token mixture is reported, the effect is decomposed by
token type, and a mixture-matched summary keeps its sign.
Read the full abstract ▾
All authors Danlin Li, Miao Li, Xu Han, Jiecheng Lu
Machine Learning
Counterfactual Gate-Support Decomposition: Separating Selection from Redistribution in Training-Data Filtering
Hyojung Han, Jongmin Kim and Seung-Hun Jeon
A training-data quality gate changes both which examples survive within regions of the input space and how training mass is distributed across those regions, so a gate-versus-random comparison does not isolate these mechanisms. We introduce Counterfactual Gate-Support Decomposition (CGSD), a four-arm, budget-matched design that crosses gate eligibility with support distribution and decomposes the observed gate effect into within-support selection, between-support redistribution, and their interaction. The practical constraint is finite-pool realizability: we characterize mass and
token-assembly conditions under which the required arms can be constructed, together with a support-extinction case in which the ordinary design is undefined. On Korean-English and Korean-Japanese
retrieval, the mean within-support selection component is the only one that exceeds its calibrated null band (+29.7 and +5.6 holdout items; 30/30 positive replicates), decisively in Korean-English and marginally in Korean-Japanese; redistribution and interaction remain unresolved at the available power. The same pattern holds for a second, unrelated gate family – a model-confidence threshold rather than a deterministic verifier – judged against its own null (+27.3 items, 9/9 positive), so the result is not a property of one gate's mechanism. On a mathematics gate, zero-eligibility supports leave the ordinary design undefined at every budget, while the extended design is constructible only inside narrow budget bands that an exhaustive scan locates and that the natural operating point misses – the assembly condition separating a property of the pool from a limitation of the packer. CGSD therefore returns a decomposition, a constructible-but-underpowered design, or an explicit non-constructibility result, and says which; it prevents support redistribution from being silently interpreted as sample-level gate quality.
Read the full abstract ▾
Multiagent Systems
Why Do Agent Teams Fail Across Boundaries? A Failure Taxonomy for Agent-to-Agent Systems with Private States
Sarath Shekkizhar, Prathyusha Jwalapuram, Zixuan Ke et al. (+12 more)
Agents are increasingly interacting with other
agents that act for different principals (people or organizations on whose behalf they work), such as a buyer's personal
agent negotiating with an enterprise's seller
agent. In such
agent-to-
agent (A2A) scenarios, each
agent has its own environment and holds information and authority that the other cannot see. Moreover, the interests of each
agent may be aligned (cooperative), partly aligned (mixed-motive), or opposed (competitive). Existing multi-
agent system failure studies assume one orchestrator, shared workspace, or single principal. We ask how A2A systems fail when these assumptions are removed. We study more than 2,500 recorded runs of A2A systems built from different
agent harnesses and models, spanning 488 scenarios across eight families (legal, CRM, security, commerce, etc.). From these traces, we derive $A^2$-Failures, a taxonomy of 20 A2A failure modes in four categories: (i) discovery and delegation, (ii) grounding and record consistency, (iii) verification, and (iv) conduct and fidelity. Three findings hold across every harness, model, and scenario studied: (1) discovery and interaction are driven by the phrasing of the task rather than the availability of peer
agents; (2) no
agent verifies a peer's factual claim on its own initiative, so an A2A system inherits the reliability of its least reliable participant unless the protocol or mechanism performs the check; and (3) the aggregate outcome score credits the
agent completing the task while staying blind to the coordination, verification, and conduct failures beneath it, so A2A systems must be evaluated per
agent, against each
agent's own principal. We release the run logs behind the taxonomy at https://huggingface.co/
datasets/a2a-salesforce/a2-failures.
Read the full abstract ▾
All authors Sarath Shekkizhar, Prathyusha Jwalapuram, Zixuan Ke, Vaidehi Patil, Jiayu Wang, Srijan Bansal, Trong Huy Nguyen, Dingfeng Jiang, Dongqi Liu, Ruochen Zhao, Wei Lu, Giuseppe Carenini, Mohit Bansal, Silvio Savarese, Shafiq Joty
Computer Vision
MMKG-R1: Constructing Multimodal Knowledge Graphs via Reinforcement Learning
Zongyu Wu, Yilong Wang, Xiaochen Wang et al. (+5 more)
Multimodal knowledge graphs (MMKGs) organize information across different modalities to support
multimodal Graph
Retrieval-Augmented Generation (RAG), which aims to improve the performance of
multimodal large
language models (MLLMs). The effectiveness of these
multimodal Graph RAG methods largely depends on the quality of the constructed MMKGs. However, most existing methods construct MMKGs without explicit guidance on graph quality, while reinforcement learning (RL)-based methods that rely on final-answer correctness may provide only indirect feedback on MMKG quality. To address these limitations, we propose MMKG-R1, a novel MMKG-quality-driven RL framework for training an MLLM as a
multimodal knowledge constructor. To be specific, MMKG-R1 introduces multiple quality-aware rewards to directly evaluate both the quality of information extracted from individual knowledge sources and the quality of the assembled MMKG. These rewards are then used to guide the RL training. Experiments across multiple
datasets show that MMKG-R1 constructs MMKGs that improve downstream question-answering accuracy.
Read the full abstract ▾
All authors Zongyu Wu, Yilong Wang, Xiaochen Wang, Minhua Lin, Yuan Zhong, Fenglong Ma, Xiang Zhang, Suhang Wang
Dynamical Systems
Uniqueness of Limit Cycles for Abel Equations and Planar Systems
José Luis Bravo Trinidad, Renhao Tian and Yulin Zhao
We study positive limit cycles of the Abel equation $$\dot{x}=A(t)x^{3}+B(t)x^{2},$$ where $A$ and $B$ are $2\pi$-periodic functions. By developing a half-period convexity method, we establish a new criterion ensuring that the equation has at most one positive limit cycle. Moreover, if such a limit cycle exists, it is
hyperbolic. As applications, we consider two classes of planar
polynomial systems that can be reduced to Abel equations. For a subfamily of cubic rigid systems, we prove that there is at most one limit cycle, thereby giving a partial answer to the first problem in Gasull's list of 33 open problems. For quadratic systems, we derive a new uniqueness criterion for limit cycles surrounding the origin and provide explicit parameter regions under which the criterion is satisfied.
Read the full abstract ▾
Artificial Intelligence
Can We Believe AI Peer Review?
Zongrui Yang, Haoyuan Li, Zhongsheng Wang et al. (+2 more)
As large
language models become widely used in scientific review and manuscript revision, AI scores are increasingly serving as feedback for assessing paper quality and guiding edits. Yet whether a single AI score reliably reflects the reviewer's judgment, and whether revision gains based on that score are endorsed by other reviewers, remain understudied. Using 200 papers from ICLR 2025 and 2026 and approximately 14,000 AI reviews, we study score reliability, score-distribution estimation, and cross-model agreement on revision gains. Repeated reviews fluctuate even under identical conditions. Finer-grained scoring protocols reduce score variance but also systematically shift mean scores, while the gains assigned to manuscript revisions depend strongly on the reviewer.
Read the full abstract ▾
All authors Zongrui Yang, Haoyuan Li, Zhongsheng Wang, Gillian Dobbie, Jiamou Liu
Physics and Society
The Gompertz-Zipf city size distribution and urban allometry
Sharon Teles
This dissertation discusses the distribution of city sizes using censuses of the Brazilian Institute for Geography and Statistics (IBGE) spanning 150 years, from 1872 to 2022, as well as the allometric scaling relation with the 2022 census data. The Gompertz distribution is fitted to the population size of less populated cities, while the Zipf power law is used to represent the cities having larger populations, with threshold varying between 20,216 and 102,857 inhabitants. The data analysis shows that the Gompertz-Zipf distribution can describe the data across all censuses, with the Gompertz parameter $B$ oscillating between $-2.39\times10^{-5}$ and $-1.49\times10^{-5}$, indicating periods with faster decreases and periods with slower decreases in the number of cities within the less populated segment as population increases, while the Zipf exponent ranges from $a=3.61\pm0.09$ in 1872 to $a=2.28\pm0.02$ in 2022. The Gini coefficient $G$ and the Zanardi index $Z$ are used to quantify the inequality in the city population distribution, whose obtained values are $G=0.439$ and $Z=-0.002$ for the 1872 census and $G=0.733$ and $Z=0.018$ for the 2022
dataset, while the Pietra index $p$ and the Kolkata index $k$, associated with the most populated cities, are $p=31.893\%$ and $k=65.894\%$ for 1872, and $p=57.395\%$ and $k=78.342\%$ for 2022. These results reveal a high concentration of population in larger cities, with $G$ increasing from 1872 to 2022 and $Z$ becoming positive after 1900, indicating a greater concentration of population in the group of more populated cities than in the group of less populated cities. Finally, urban allometry is also examined, generating an exponent $b=0.835$ for the 2022 data, showing that the scaling relation between area and population holds for urbanized area but not for total administrative area, indicating that scaling is associated with spatial occupation.
Read the full abstract ▾
Human-Computer Interaction
Beyond Initial Creation: Two Decades of Community-based Chart Revisions from Wikimedia Commons
Jasmine Y. Shih, Chase Stokes, Melanie Tory et al. (+1 more)
Open online communities like Wikipedia offer a rich space for understanding the organic
evolution of public-facing artifacts, but little is known about how these communities collectively edit chart images. We investigate the phenomenon of "community editing" of visualizations by exploring chart revision patterns on Wikimedia Commons. We introduce a novel
dataset of 7,786 longitudinal chart revisions across 1,664 images, labeled with edit categories using an LLM pipeline. Initial data analysis uncovered several novel findings. Multi-contributor revision is associated with more edits to mark styling and text content compared to single-contributor revision. Temporally, mark and text edits occur in more "bursty" cadences than data edits. In addition, while SVG charts attract more unique contributors and revisions than raster charts, their edit histories involve more tightly scoped and less diverse changes. This
dataset enables a wide range of analyses of chart change over time.
Read the full abstract ▾
All authors Jasmine Y. Shih, Chase Stokes, Melanie Tory, Marti Hearst
Robotics
Probabilistic Physics Digital Twin for Human-Robot Interaction
Yushen Hu, Harshvardhan Chandirasekar, Apurv Kushwaha et al. (+2 more)
Digital twins are increasingly used in robotics to simulate interactions, predict action outcomes, and support planning. However, constructing accurate digital twins requires precise estimates of physical properties such as mass, friction, and center of mass, which are often difficult to obtain in real-world environments. We propose probabilistic digital twins, which represent uncertain physical properties as probability distributions rather than fixed values. By simulating actions across these distributions, the twin predicts a range of possible outcomes and their likelihoods. These probabilistic simulations can guide both human and
robot actions by identifying actions that are effective despite physical uncertainty. We demonstrate this framework through human-
robot interaction scenarios, including interfaces that communicate simulated outcomes to users and
robot planning strategies that explicitly account for uncertainty.
Read the full abstract ▾
All authors Yushen Hu, Harshvardhan Chandirasekar, Apurv Kushwaha, Jonathan King, Zhu-Tian Chen
Other Statistics
Differentially Private Learning of Markov Chains: Near-Optimal Sample Complexity for Item and User Level Privacy
Devansh Gupta, Cristóbal Guzman and Enayat Ullah
Sequential data such as documents, conversations, and trajectories are naturally partitioned across users, motivating
privacy guarantees that protect each user's entire trajectory. We initiate the study of privately estimating the transition
matrix of a discrete ergodic Markov chain from N independent trajectories of length T, with arbitrary initial distributions, under maximum row-wise ell 1 error. Besides N, T, and the number of states, our bounds depend on the minimum stationary probability and the pseudo-spectral gap, which govern state coverage and mixing. Under item-level
privacy, we give upper and lower bounds that separate the costs of sampling, state occupancy and mixing, and
privacy. Under user-level
privacy, we reduce the problem for sufficiently long trajectories to private product distribution estimation. Along the way, we establish optimal user-level private bounds for estimating products of discrete distributions, which may be of independent interest. In the user level setting, our bounds are nearly tight in the large T regime, whereas in the item-level setting, our bounds as tight up-to a non-private term.
Read the full abstract ▾
Strongly Correlated Electrons
Evidence for hidden order and its potential correlation with electron-hole asymmetry in iridate superlattices
Zhihan Qiao, Han Zhang, Hangtian Wang et al. (+12 more)
Hidden order refers to an order that shapes the properties of
quantum materials yet remains difficult to identify. Here we report signatures of a hidden order in doped iridate superlattices through an unexpected electron-hole asymmetry in the coupling between the insulating state and the antiferromagnetic (AFM) state. While carrier doping suppresses both states, they remain coupled in the electron-doped regime but decouple in the hole-doped regime. Structural analysis rules out octahedral rotation and tilting as the dominant origin of this asymmetry. Instead,
magnetic characterization reveals evidences of a hidden order that only appears in the hole-doped regime. This hidden order emerges above TN and is closely associated with the AFM state, coinciding with an exceptionally strong
magnetic scattering intensity and an anomalously robust
magnetic coherence. Our results highlight that hole-doped iridates can be a promising platform for exploring hidden and unconventional ordering phenomena in
spin-orbit-entangled
quantum materials.
Read the full abstract ▾
All authors Zhihan Qiao, Han Zhang, Hangtian Wang, Deyang Li, Haiyang Zhang, Enyang Men, Junlei Jiang, Jingxin Chen, Yizhou Liu, Sang-Youn Park, Xiangnan Xie, MingLiang Tian, Haifeng Du, Ning Hao, Lin Hao
Symplectic Geometry
Algebraic overtwistedness of contact $(+1)$-surgeries via augmented Floer theories
Russell Avdek, Youlin Li and Zhengyi Zhou
We prove that contact $(+1)$-surgery along a Legendrian $n$-sphere $\Lambda\subset(\mathbb{S}^{2n+1},\xi_{\mathrm{std}})$ with \[ \mathrm{tb}(\Lambda)\ne (-1)^{\frac{n(n-1)}{2}+1} \] has vanishing contact homology. In dimension three, this produces new examples of tight algebraically overtwisted contact
manifolds, vastly extending the authors' previous results: If a Legendrian
knot in $(\mathbb{S}^3,\xi_{\mathrm{std}})$ admits an exact, oriented Lagrangian filling that is not a disk, then contact $(+1/k)$-surgery on it is algebraically overtwisted and tight for every positive integer $k$. Our proofs use versions of symplectic cohomology and wrapped Floer cohomology defined by augmentations of a chain-level contact homology
algebra.
Read the full abstract ▾
Analysis of PDEs
Nondegeneracy and asymptotic behavior of type I blow-up for semilinear heat equations with general nonlinearities
Yohei Fujishima, Ryoto Ichiya and Norisuke Ioku
We study the local asymptotic behavior of nonnegative type I blow-up solutions to semilinear heat equations with nonlinearities that need not be scale invariant. Using a quasi-similarity transformation adapted to the associated ordinary differential equation, we establish locally uniform convergence to a constant profile on the parabolic spatial scale near every interior blow-up point. Under a quantitative assumption on the nonlinearity at infinity, our results cover regularly varying nonlinearities in the Sobolev subcritical and critical ranges and, in one and two space dimensions, rapidly growing nonlinearities including double exponentials. A key ingredient in this analysis is a positive lower bound for the normalized solution at interior blow-up points. In fact, under weaker assumptions on the nonlinearity, we prove this bound in arbitrary space dimension without assuming radial
symmetry.
Read the full abstract ▾
Computation and Language
LLM-jp-4 Technical Report
LLM-jp: Chaoran Liu, Daisuke Kawahara, Hideyuki Tachibana et al. (+14 more)
This paper presents LLM-jp-4, a family of Japanese sovereign large
language models (LLMs) developed with an emphasis on transparency, reproducibility, and practical usability. LLM-jp-4 comprises an 8B dense model and a 32B-A3B mixture-of-experts model, both trained from scratch on approximately 12 trillion
tokens. To support independent verification and reproduction, we establish explicit criteria for selecting training data, external models, and software, and make the development process publicly available. Evaluation results demonstrate substantial improvements over previous LLM-jp models and competitive performance against open-weight models of similar scale. By documenting the development principles, training procedures, and empirical findings of LLM-jp-4, we aim to provide a reference for future sovereign AI development.
Read the full abstract ▾
All authors LLM-jp: Chaoran Liu, Daisuke Kawahara, Hideyuki Tachibana, Hirokazu Kiyomaru, Jun Suzuki, Koshi Eguchi, Kouta Nakayama, Qianying Liu, Rio Yokota, Ryo Takahashi, Sadao Kurohashi, Sosuke Hosokawa, Taishi Nakamura, Takashi Kodama, Yun-Ang Wu, Yusuke Miyao, Yusuke Oda
Computer Vision
NavAble: A Large-Scale Dataset and Synthetic Data Generation Pipeline for Blind Navigation
Hochul Hwang, Jahir Sadik Monon, Soowan Yang et al. (+9 more)
Reliable recognition of accessibility-critical objects (e.g., audible pedestrian signals, door-activation buttons, and handrails) is essential for assistive navigation technologies that support safe, independent mobility for blind and low-
vision (BLV) users. Yet these categories are underrepresented in existing large-scale
vision datasets, and the few
datasets that include them suffer from limited class coverage, inconsistent annotations, and poor diversity. We introduce NavAble, a large-scale
dataset spanning 11 accessibility-critical object classes, combining 41K curated open-source images with 8K newly collected, densely annotated real-world images. To extend scale and diversity, we develop a synthetic data generation pipeline that renders high-fidelity 3D assets across 37 environments under diverse styles and viewpoints, yielding 452K frames with rich ground-truth annotations from 565 assets. The pipeline minimizes manual effort by combining automated web image crawling, VLM-based filtering, SAM 3D-based asset generation, and convenient camera trajectory configuration, enabling large-scale data generation at a diversity unattainable through manual collection. Experiments show that fine-tuning existing segmentation models on the NavAble
Dataset achieves reliable accessibility-object segmentation (>80% mIoU on our held-out test set). We also demonstrate the effectiveness of synthetic data augmentation, which yields up to +3.5 mIoU over real-world-only fine-tuning. Beyond
benchmark evaluation, validation on egocentric data from two mobility-assistive
robots demonstrates improved perception reliability in real-world navigation scenarios. Together, NavAble establishes a scalable foundation for BLV navigation perception, directly addressing the data scarcity and embodiment gaps. The
dataset and pipeline are publicly available at https://huggingface.co/
datasets/NavAble/NeurIPS_2026_BLV.
Read the full abstract ▾
All authors Hochul Hwang, Jahir Sadik Monon, Soowan Yang, Shiven Umeshbhai Patel, Anh Nhat Hong Nguyen, Kien Nguyen, Keshav Garg, Dylan Gage, Duretti Hordofaa, Anshu Anjna, Eshed Ohn-Bar, Donghyun Kim
High Energy Astrophysical Phenomena
Search for Sub-Solar Mass Binaries in the First and Second Part of LIGO's Fourth Observing Run O4a-O4b
Keisi Kacanja, Kanchan Soni and Alexander H. Nitz
The observation of sub-solar mass
neutron stars and black holes would signify a new class of astrophysical objects since these systems are not expected to form via standard stellar
evolution. We present the first sub-solar mass search for binaries consisting of
neutron star or black holes that uses the data from both the second part (O4b) and first part (O4a) of the fourth observing run of the Advanced LIGO detectors. The search covers primary masses 0.1 to 2 $M_\odot$ and secondary masses 0.1 to 1 $M_\odot$. Tidal effects up to $7\times10^5$ are accounted for in extremely low-mass
neutron stars. No statistically significant candidates are identified. We place a $90\%$ confidence upper limit on the merger rate $\mathcal{R}_{90}$ for sub-solar mass black holes and
neutron stars to be $< 5.2\times10^{2}\,\textrm{Gpc}^{-3}\,\textrm{yr}^{-1}$ for a chirp mass of 0.5 $M_\odot$, a factor $2.2 \times$ lower compared to previous O1-O4a searches. We constrain the effective local dark matter fraction to be $\tilde{f}_\textrm{PBH}<0.6\%$ for $M_{\textrm{PBH}}=0.8 \,M_\odot$, approximately $1.5$ factor improvement to previous O1-O4a constraints. Given our model assumptions, our local dark matter fraction constraints are $2-7$ times lower than the OGLE microlensing survey for $M_{\textrm{PBH}}\ge{0.6}~M_\odot$.
Read the full abstract ▾
Earth and Planetary Astrophysics
ExoEcho: Exoplanet Ephemerides Change Observations. III. A JWST Survey for Transit Timing Variations of 37 Exoplanets
Yunke Wu, Zixin Zhang, Xinyue Ma et al. (+6 more)
Transit timing variations (TTVs) provide a powerful probe of planetary dynamics and orbital
evolution. We extend the ExoEcho project by incorporating James Webb Space
Telescope (JWST) transit observations and analyze 84 light curves of 37
exoplanets, including 20 hot Jupiters, 11 Neptune-like
planets, and 6 super-Earths. We derive updated transit ephemerides and conduct a systematic search for long-term and periodic TTV signals. We identify two strong TTV candidates, including HATS-6~b and HAT-P-18~b. The hot Jupiter HATS-6~b exhibits a long-term trend with $\dot{P} = 230.5 \pm 46.8\ \mathrm{ms\ yr^{-1}}$, potentially driven by tidal interactions with its host
star. In addition, the Neptune-like
planet HAT-P-18~b shows periodic TTV signal with amplitude of $92.9 \pm 13.7$~s. We also show that the previously claimed positive period derivative in WASP-121~b is inconsistent with our new transit timing data. Although continued monitoring is required to confirm these findings, our results demonstrate the potential of JWST for high-precision transit timing and extend TTV studies beyond the traditional hot-Jupiter population to lower-mass
exoplanets.
Read the full abstract ▾
All authors Yunke Wu, Zixin Zhang, Xinyue Ma, Wenqin Wang, Siqi Guo, Cong Yu, Shang-Fei Liu, DiChang Chen, Bo Ma
Computer Vision
SonoWorld: Towards Interactive Pose-Controllable Ultrasound World Models
Zefan Yang, Zheyuan Zhang, Boris Maihe et al. (+7 more)
Robotic ultrasound can standardize image acquisition and reduce operator workload and variability, but policy learning remains limited by the lack of realistic, controllable simulation environments. While controllable world models have advanced rapidly in natural-image domains, ultrasound world modeling remains challenging because each frame captures only a cross-section of unobserved 3D anatomy, causing small probe motions to induce large appearance changes. Existing approaches rely on
robot-tracked poses and latent-space prediction, but do not directly generate the observations required for interactive control. We propose SonoWorld, a causal ultrasound world model that learns
video dynamics from unlabeled clinical
videos and grounds them to 6-DoF control using a small set of freehand clinical scans tracked with an EM sensor. SonoWorld is trained in three stages: pretraining a latent-action-conditioned
video diffusion model on around 6.9 million unlabeled multi-organ ultrasound frames; grounding the learned
video dynamics to metric probe poses with a small EM-tracked thyroid
dataset; and applying distribution matching
distillation to enable causal, long-horizon autoregressive generation with selectively caching eviction
tokens at inference time. We evaluate controllability through pose-following fidelity, round-trip consistency, and axis-specific control under synthetic actions on held-out tracked scans, and further validate it with a user study. In model-based action discrimination, SonoWorld correctly ranks the target-directed probe motion over an equal-magnitude opposite motion in up to 74.8% of trials. To our knowledge, SonoWorld is the first causal, pose-controllable ultrasound world model learned primarily from freehand clinical
videos, providing a data-driven interactive environment for robotic ultrasound data acquisition and policy evaluation.
Read the full abstract ▾
All authors Zefan Yang, Zheyuan Zhang, Boris Maihe, Yikang Liu, Hang Yu, Ankush Mukherjee, Xiao Chen, Pingkun Yan, Terrence Chen, Shanhui Sun
Numerical Analysis
Solving a sequence of eigenvalue problems using snapshot and row subsampling
Francesco Hrobat and Yuji Nakatsukasa
A sequence of
eigenvalue problems requires solving $A_i\vec{x}_i= \lambda_i\vec{x}_i$ where $i \in \{1, \dots, N\}$, $N \gg 1$ and (usually) $\lambda_i$ is the largest or smallest
eigenvalue of $A_i$. Often, the
matrices $A_i$ come from a discretization of a smooth map depending on a parameter $p \rightarrow A(p)$, that is $A_i = A(p_i)$, so the
matrices $A_i$ change gradually along the sequence. Under this latter assumption, we introduce an efficient
algorithm for solving sequences of
eigenvalue problems. Our
algorithm combines the ideas of reduced order model and row subsampling, as done in [E. Jones and Y. Nakatsukasa, arXiv, 2025] for linear systems. In addition, in our approach, the same set of rows is guaranteed to work well independently of the value of the parameter $p$ or the
matrix $A(p)$ as long as the snapshot
matrix is unchanged and given that the row subset was chosen well at the beginning. Furthermore, we implement an update strategy of the snapshot
matrix. Along these results, we present theoretical analysis of the snapshot
matrix for
eigenvalue problems, showing an exponential decay of the singular values, and we give bounds on the error on eigenvectors for unseen value of the parameter. This analysis generalizes that in [M. Guido, D. Kressner, and P. Ricci, J. Sci. Comput., 2024] and shows that, regardless of the distribution of the snapshot points, the convergence is always (asymptotically) exponential in the number of snapshot points if the function $A(p)$ admits an analytical extension. We further generalize our
algorithm for computing largest or smallest singular values and relative singular vectors and to generalized
eigenvalue problems. We conclude with numerical results showing the efficiency of our approach.
Read the full abstract ▾
Cryptography and Security
AutoPoly-HE: Autonomous Design of Polynomial Activations for Privacy-Preserving Encrypted Inference
Zeinab Elkhatib, Kamrul Hasan, Ali Sekmen et al. (+2 more)
Privacy-preserving machine learning has become critical for deploying artificial intelligence in sensitive domains such as healthcare, finance, and infrastructure monitoring. Homomorphic encryption (HE) enables secure inference on encrypted data. Despite its advantages, HE is not compatible with nonlinear activation functions found in machine learning models. Existing methods mitigate this problem by manually designing
polynomial approximations for these activation functions. However, these functions do not account for the differences in
dataset characteristics or model architecture. Furthermore, these functions are also limited to the accuracies of convolutional neural network (CNN) models. In this work, we propose an autonomous framework that selects
polynomial activation functions tailored to each
dataset and model. Using this framework, we analyze the interdependence between
polynomial activation design and model architecture by evaluating several CNN ResNet and
Vision Transformer models across multiple
datasets. Experimental results demonstrate that the proposed framework generalizes across multiple architectures, including ResNet-18, ResNet-34, ResNet-50, ViT-S/16, ViT-B/16, and ViT-L/16. Across all
datasets,
transformer-based models maintain higher accuracy under
polynomial approximation, while convolutional models exhibit increased degradation as task complexity grows. In addition, latency analysis reveals that the proposed CLS
token based inference pipeline significantly reduces computational overhead by eliminating depth-dependent operations, shifting the primary cost to encryption and decryption. These findings demonstrate that the proposed method not only adapts
polynomial activations across architectures, but also enables efficientencrypted inference through both activation design and pipeline optimization.
Read the full abstract ▾
All authors Zeinab Elkhatib, Kamrul Hasan, Ali Sekmen, Imtiaz Ahmed, Tariqul Islam
Machine Learning
Why don't Graph Transformers attend to the same token?
Alexandre Bismuth
In large
language models (LLMs),
attention concentrates heavily on the first
token. This '
attention sink' is believed to protect deep models against over-mixing. Even though graph
transformers face a related pressure, we find that a SoTA graph
transformer, GRIT, rarely forms sinks on molecules, and that it sinks less when trained on larger graphs, the opposite of the trend in LLMs. In principle, a graph can prevent sinks: we prove and show empirically that its
symmetry and locality limit the
attention any node can receive. On molecules, however, with distinguishable atoms and global
attention, neither limit prevents sinks; what stops them lies in the model. We find that the
attention scoring function is the main cause. Replacing GRIT's gated logit with a dot product raises its sink rate from 0.09 to 0.68 on identical graphs. In Graph
Transformers using dot-product
attention, removing the sink gives about the same error as removing a random node, while GRIT's most-attended nodes matter to its predictions.
Attention sinks are therefore a property of dot-product
attention rather than of
Transformers. In our experiments, they form when dot-product scoring has a distinguishable node to concentrate on.
Read the full abstract ▾
Computation and Language
TopicMerge: A Divide and Conquer Approach to Topic Modeling
Matthew Raffel, Lizhong Chen, Sai Munikoti et al. (+1 more)
Recent advancements in topic modeling have split into two main paths: efficient
embedding-based methods that offer scalability but are often outperformed on standard clustering metrics, and Large
Language Model (LLM) approaches that achieve state-of-the-art clustering performance but at the cost of prohibitive computational expense and poor scalability. This paper introduces TopicMerge, a novel multi-stage pipeline that bridges these approaches to achieve both state-of-the-art clustering performance and computational efficiency. TopicMerge first employs an efficient clustering pipeline to deliberately over-segment a document corpus into a large set of fine-grained, semantically pure micro-clusters, which are then given initial semantic labels. It then introduces a novel, iterative merging
algorithm where an LLM, guided by local neighborhood context, arbitrates on consolidating candidate pairs. These approved merges are then selectively executed and semantically relabeled by the LLM, hierarchically building broader, more coherent topics. Extensive experiments on the Bills and Wiki
datasets show that TopicMerge achieves state-of-the-art clustering performance while reducing required LLM calls by over 92% compared to document-level methods, offering a practical and scalable solution.
Read the full abstract ▾
All authors Matthew Raffel, Lizhong Chen, Sai Munikoti, Sameera Horawalavithana
Artificial Intelligence
Generalization Analysis of Biased Stochastic Gradient Methods for Minimax Problems
Shuang Zeng, Yunwen Lei and Yiming Ying
Minimax optimization refers to a class of optimization problems with some variables to minimize and other variables to maximize. Biased
stochastic gradient methods have shown practical success in solving minimax problems to either improve robustness, enhance communication efficiency, or decrease computational costs. These successes motivate a lot of theoretical works to study the convergence of biased
stochastic gradient methods, while their generalization analysis remains untouched. In this paper, we present the first framework to study the stability and generalization of biased
stochastic gradient methods for minimax problems. We establish a connection between stability and generalization for minimax problems by relaxing the existing bounded gradient assumption to a bounded second moment condition. We then introduce a generalized Lipschitz-type condition on bias and gradient
estimators, and derive a general stability bound to clarify the connection among bias, gradient
estimators, and stability. We apply our general analysis to Zeroth-order and Clipped
stochastic gradient descent ascent (SGDA), and derive stability bounds that match those of SGDA under appropriate smoothing/clipping parameters. We combine stability and convergence analyses together, and derive optimal excess risk bounds of order $1/\sqrt{n}$, where $n$ is the sample size.
Read the full abstract ▾