Peer Revenue Equity in science

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Friday 6 November

Dynamical Systems

KAM theory for quasi-periodic equilibrium of generalized Frenkel-Kontorova model in $\mathbb{Z}^{d}$

We extend the KAM theory for one dimensional Frenkel-Kontorova model to multidimensional case in quasi-periodic media. By using the quasi-Newton method, we reduce the linearized operator with quasi-periodic coefficient in the Nash-Moser iteration to the composition of two diagonal operators and a near-identity operator, such that the small divisor problem can be overcome by imposing the simultaneous Diophantine condition. In contrast, we also discuss the application of Craig-Wayne-Bourgain method to the quasi-periodic equilibrium problem.
Read the full abstract

Cryptography and Security

Who's Asking? Audience-Conditioned Privatization

The abundance of user-generated conversational data makes the question of how to responsibly derive insights from confidential data without putting the individuals in it at risk ever more important. However, the majority of existing differentially private (DP) synthetic text generation (SDG) and anonymization approaches ignore the context of the data's downstream usage, which results in loss of important information. DP-SDG provides the same guarantee regardless of the recipient, and anonymization relies on predefined attribute categories. They both ignore the observation that an attribute that is harmless to one audience can be sensitive to another, and each audience typically needs only some attributes preserved. We propose Ext-Classify-Priv, a framework that (1) extracts the information types present in a corpus, (2) classifies which of them are private for a given audience following contextual integrity, and (3) protects only those attributes, using either anonymization or SDG with Feature-DP approaches. On Reddit data with 3 audiences, our audience-specific releases are more faithful than a single audience-agnostic release for every audience. On bioRxiv data, our SDG method improves MAUVE from 0.097 to 0.420 at epsilon=1 and on the proprietary real estate data, our anonymization method improves MAUVE from 0.698 to 0.959 over the best baselines.
Read the full abstract

All authors Daniil Filienko, Anusha Bagalkotkar, Xan Vongsathorn, Sikha Pentyala

Machine Learning

Learning to Restart: Differentiable Restart Policies for Optimization Dynamics

Adaptive restart mechanisms, which reset the momentum or other auxiliary state of an iterative optimizer, play a central role in the empirical performance of accelerated first-order methods, yet existing restart rules remain almost entirely hand-designed. The obstruction to learning restart policies is that, unlike the typical Learning to Optimize (L2O) setting, the discrete restart events make the resulting trajectory non-differentiable in the algorithm parameters and thus block gradient-based optimization of the restart policy. We resolve this obstruction by developing a differentiable framework for restart policies through their continuous-time limit. We introduce surrogate gradients for the restart events and optimization trajectories that are efficient to compute and provably converge to their continuous-time counterparts. We validate the effectiveness of our surrogate gradients in L2O settings, training restart policies that improve upon standard handcrafted adaptive restart criteria on quadratic and linear programs, and that transfer zero-shot from synthetic quadratics to logistic regression on real datasets.
Read the full abstract

High Energy Astrophysical Phenomena

Modeling of the X-ray Flux and Polarization Energy Spectra of Black Hole X-Ray Binaries with the KerrC Code

In this chapter, we introduce the kerrC model, an advanced framework designed for the analysis of combined spectral and polarimetric X-ray emission from black hole systems. The model is built upon an extensive dataset of ray-tracing Monte Carlo simulations that compute the radiative transfer of X-rays in the strongly curved spacetime near the event horizon of a Kerr black hole. These simulations incorporate thermal seed emission from a geometrically thin, optically thick accretion disk, along with the subsequent Comptonization processes within a hot corona of variable geometry and physical properties. Crucially, the reflection of both primary disk and Comptonized photons is treated self-consistently utilizing the xillver reflection tables, and the polarization 4-vectors of the photons are rigorously transported throughout the simulation to preserve full polarization information. The code tracks photons forward in time, allowing for multiple photon scatterings in the corona and off the accretion disk. We detail the underlying mechanics of the ray-tracing simulations and the fitting models, and discuss the applications of kerrC in leveraging X-ray polarization observations to constrain black hole spin and coronal geometry.
Read the full abstract

Materials Science

Limits of Lorenz number estimation from the temperature-dependent Seebeck coefficient

Separating electronic and lattice thermal conductivity often relies on a Lorenz number estimated from the Seebeck coefficient using an assumed band and scattering model. We ask how much information about the Lorenz number comes from temperature-dependent Seebeck coefficients rather than the assumed model. We formulate compatible minimum and maximum Lorenz numbers as linear programs in a fixed finite spectral representation, assuming known chemical potentials and a common energy dependence across observation temperatures for each candidate spectrum. Reference Seebeck coefficients are generated by single-band and same-sign two-band models. Increasing the prescribed observations from one to five temperatures reduces the Lorenz interval widths by approximately 97.6% and 97.8%, respectively. Nevertheless, the single-band example retains an attained interval of approximately $[1.46453,2.37037]\times10^{-8}$ W $\Omega$ K$^{-2}$ at 400 K. Exact electrical conductivity values do not narrow these ranges when spectral amplitudes can vary independently with temperature. Thus, even exact Seebeck coefficients and electrical conductivities at all five temperatures need not uniquely determine electronic thermal conductivity at 400 K. For the same single-band example, known ratios of spectral amplitudes across temperatures allow electrical conductivity values to confine all compatible Lorenz numbers to an enclosing interval less than 1% as wide as the five-temperature free-amplitude interval. Temperature-dependent Seebeck data strongly narrow the compatible range, while conductivity provides further spectral constraints when the temperature dependence of spectral amplitudes is independently restricted.
Read the full abstract

General Relativity and Quantum Cosmology

A pipeline for identifying astrophysical signals and instrumental glitches in LISA data

The Laser Interferometer Space Antenna (LISA) is a space-based gravitational wave detector scheduled for launch in the 2030s by the European Space Agency. Operating in the 0.1 mHz to 0.1 Hz band, LISA will observe a variety of astrophysical sources including but not limited to supermassive black hole mergers, extreme mass ratio inspirals, and white dwarf binaries. LISA will yield new insights, including supermassive black hole formation, new tests of general relativity, population studies of white dwarfs and others. Like all interferometric gravitational wave detectors, LISA data is expected to contain non-astrophysical transients, commonly referred to as glitches. We introduce a two step pipeline that consists of a search algorithm for detecting potential transients in simulated LISA data followed by a machine learning model that classifies the detected transients as gravitational waves or instrumental glitches. This is important for MBHBs with very high masses (around $10^6\ M_{\odot}$) which are transients that could be morphologically similar to LISA glitches in the data. Initial results demonstrate a promising ability to classify detected events with an accuracy close to $98\%$. The pipeline also shows the ability to analyze a month of data in under 2 minutes on a laptop. This shows an eventual utility of our pipeline for enabling accurate rapid alerts from low latency searches. It could also support model selection in global-fit analyses by providing an initial summary of the location and type of individual signals. This paper also discusses challenges encountered during training, including limitations of the current framework.
Read the full abstract

All authors Aryamann Rao, Jess McIver, Rhiannon Udall, Natalia Korsakova, Malvina Bellotti

Computer Vision

GeoJeV: Unified Spatial Probability for Training-Free Efficient Evidence Acquisition in Ultra-High-Resolution Remote Sensing

Ultra-high-resolution (UHR) remote sensing imagery provides rich spatial details but poses challenges for accurate and efficient multimodal understanding. Existing evidence acquisition methods selectively explore relevant image regions, yet largely rely on generative multimodal large language models (MLLMs) for intermediate search control. However, reinforcement learning from human feedback (RLHF) primarily aligns these models with human-preferred responses rather than explicitly optimizing region-level relevance judgments, while repeated autoregressive decoding introduces additional computational overhead. To address these limitations, we propose GeoJeV, a training-free method that formulates evidence acquisition as spatial probability-driven decision making. At each evaluated region, a frozen JeV model produces a probability distribution over candidate regions. Specifically, Region Rank prioritizes candidate regions, Sparse Branch adaptively determines branching width through sparse probability support, and Depth Gate controls refinement depth based on branch sparsity and distribution concentration. By coordinating search priority, branching width, and refinement depth through a shared distribution, GeoJeV reduces reliance on RLHF-aligned generation and eliminates intermediate decoding. Experiments on LRS-GRO, MME-RealWorld-RS, and LRS-VQA achieve accuracies of 60.59%, 52.79%, and 34.32%, respectively. Across the three benchmarks, GeoJeV improves the average accuracy of Qwen3-VL-8B by +5.32pp with only 1.12 additional seconds per question, while achieving a 7.8-fold inference speedup over the recent method.
Read the full abstract

All authors Xinyi Niu, Lingfeng Chen, Jiehua Zhang, Panhe Hu, Li Liu, Zhen Liu, Yongxiang Liu

High Energy Physics (Theory)

Simplicity of the three-point stress-tensor form factor beyond the planar limit

We express the complete three-loop subleading-color finite remainder of the three-point stress-tensor-multiplet form factor in $\mathcal N=4$ super Yang–Mills theory as a permutation sum of a single $A_2$-type function. The non-product part of the seed is a combination of 35 weight-six generalized polylogarithms, while all lower-weight products and zeta-value terms belong to the same five-letter algebra. A bootstrap based on independently computed collinear data, branch-cut constraints and empirical final-entry relations determines both the symbol and the full function. The remaining function-level ambiguity is proportional to $\zeta_2$ times the two-loop planar remainder and is fixed by a single final-entry relation. The weight-three left coproduct factors of the planar and subleading-color results span the same space modulo products; at weight four, the subleading-color result contains four additional nonclassical directions. Through the established maximal-transcendentality correspondence, this compact representation also describes the weight-six subleading-color sector of Higgs-plus-three-gluon amplitudes in QCD. We also construct a four-loop candidate symbol space under a conjectural extension of the three-loop constraints.
Read the full abstract

Human-Computer Interaction

The Relational Capacity Framework: A Lens for AI-Mediated Relationality

A growing literature examines the impact of engaging with AI chatbots on our relationality: our ability to form relationships with others, maintain them through life's changes, and exit when needed. Currently, substantial work focuses on purpose-built AI companions (e.g. Replika, Character.AI), with less work investigating socioemotional uses of the more widely-used general-purpose chatbots. In this paper, we present a 100-person investigation into how people engage general-purpose AI for socioemotional needs. Participants' AI uses varied, from practicing specific social skills to seeking additional bandwidth and information as they navigated their social contexts. To explain this heterogeneity, we provide 8 states that describe how AI expands or contracts relationality for people with different levels of skills, bandwidth, and information. We then present a unified conceptual model for the role of technology in mediating human relationality, the Relational Capacity Framework, and outline future research designing appropriate roles for AI in human connection.
Read the full abstract

All authors Emily Tseng, Patrick Gage Kelley, Renee Shelby, Ashley M. Walker

Statistics Theory

Reweighting for Transfer Classification: Exact Risk Conversion and Minimax Adaptation

A source can make a training mixture easier to learn while making its accuracy less valuable for the target. We study how the training problem a source creates and the finite-sample error of learning it jointly determine target-risk guarantees under reweighting. For binary classification, for each pair of laws and each positive target weight, we derive the sharpest target excess hinge-risk bound implied by mixture excess alone, uniformly over clipped predictors. This exact conversion curve is determined by density-weighted local votes. A two-sample oracle inequality for weighted support vector machines supplies the mixture excess, and the composition becomes explicit for Gaussian kernels. On a tilted-boundary class, we establish the all-procedure minimax classification rate $[n+s_+^2/(m^{-1}+\tau^2)]^{-1}$ in the transferable strength $s_+$ and the tilt $\tau$, separating source sampling uncertainty ($m^{-1}$) from relation uncertainty ($\tau^2$). A target-checked Gaussian-SVM selector adapts to unknown $s_+$ and $\tau$ up to a logarithmic factor for fixed dimension, and retains the target-only rate up to the same factor against arbitrary independent sources under the stated target model. On a separate smooth opposition class, the minimax risk within the family of exact Gaussian SVMs fitted to the original labels with nonnegative weights is exponentially larger than the unrestricted minimax risk whenever the target sample is at most half the source sample, despite arbitrary data-dependent weighting and tuning. The value of a source thus depends on the training problem it creates, on what accuracy there means for the target, and on whether the permitted use of the source can implement the needed correction.
Read the full abstract

High Energy Physics (Phenomenology)

Twist-3 and flavor-singlet contributions to $\gamma\gamma\to\pi^{0}\eta^{(\prime)}$ and $\eta^{(\prime)}\eta^{(\prime)}$ in $k_{T}$ factorization

We study two-photon production of the pseudoscalar pairs $\pi^{0}\eta$, $\pi^{0}\eta^{\prime}$, $\eta\eta$, $\eta\eta^{\prime}$, and $\eta^{\prime}\eta^{\prime}$ at center-of-mass energies $2<Q<4\,\mathrm{GeV}$ in $k_{T}$ factorization, including the chirally enhanced twist-3 amplitudes and the two-gluon component of the $\eta^{(\prime)}$. The twist-3 amplitudes raise the hard cross sections by about an order of magnitude in all five channels. They bring $\eta\eta$ close to the Belle data, while $\pi^{0}\eta$ stays 3 to 5 times below and falls too slowly with energy. Varying the two-gluon distribution amplitude within its uncertainty changes the cross sections by about 2% in $\pi^{0}\eta$ and $\eta\eta$ but by up to 24% in the $\eta^{\prime}$ channels, making them its probe. The hard and handbag mechanisms each give nearly constant ratios $\sigma(\pi^{0}\eta)/\sigma(\pi^{0}\pi^{0})$ and $\sigma(\eta\eta)/\sigma(\pi^{0}\pi^{0})$, whereas below $3.2\,\mathrm{GeV}$ the measured $\pi^{0}\eta$ ratio lies above both predictions and the $\eta\eta$ ratio below them. Adding the handbag amplitude coherently, with its modulus fixed by the pion and kaon channels and its phase fitted, reproduces the $\pi^{0}\eta$ ratio up to $4\,\mathrm{GeV}$ and the $\eta\eta$ ratio below $2.8\,\mathrm{GeV}$, and steepens $\sigma(\pi^{0}\eta)$ to a $Q^{-n}$ dependence with $n=9.7\pm1.6$, against the measured $10.5\pm1.2\pm0.5$. The energy dependence of the cross section ratios can thus be understood as the interference of two mechanisms of comparable size.
Read the full abstract

Chemical Physics

Fortuitous Cancellation of Competing Effects in the Free Energy of Water Autoionization: Long-Range Electrostatics, System Size, and Nuclear Quantum Effects

Few molecular processes are more fundamental to chemistry than the autoionization of water, whose equilibrium constant, $K_w \approx 10^{-14}$, underlies acid-base equilibria and the chemistry of life. Yet the molecular origins of the free energy cost of forming and separating ions remain incompletely understood. Efforts to explain it have been limited by the difficulty of modeling water reactivity in sufficient detail. Here we combine machine learning interatomic potentials (MLIPs) with spectral adaptive biasing force enhanced sampling to connect proton transfer and water wire configurations to the underlying free energy landscape. We identify competing influences of long-range electrostatics, system size, temperature, and nuclear quantum motion on the predicted ionization thermodynamics. The free energy surface extends from neutral water to ions separated beyond the solvent-separated ion pair (SSIP). It reveals a transition state in which a proton is shared between the second and third molecules of a three-membered water wire. The contact ion pair is unstable and the SSIP only weakly metastable, so determining pK$_w$ requires sampling longer wires. Long-range electrostatics raise the free energy at ion separations beyond $3\,\mathrm{\mathring{A}}$. Models with comparable force errors give substantially different free energy profiles in this region. Increasing system size lowers the calculated pK$_w$, while nuclear quantum effects bring it closer to experiment and reproduce the higher pK$_w$ of D$_2$O relative to H$_2$O. These results establish a baseline for interpreting thermodynamic results from MLIP simulations in which charged and polar groups react and diffuse in finite cells. Our analysis also explains how cancellation among competing effects can yield ionization constants close to experiment even when simulations omit several physical contributions.
Read the full abstract

All authors Samuel Varner, Yinan Xu, Yezhi Jin, Gustavo R. Pérez-Lemus, Pablo F. Zubieta Rico, Zhiying Yi, Joan M. Montes de Oca, Paul F. Nealey, Seth B. Darling, Zhen-Gang Wang, Juan J. de Pablo

Materials Science

Rotational Doppler Effect and Helimagnon Frequency Comb in Sliding Magnetic Spirals

Collective spin motion can reshape magnon spectra by providing a dynamical magnetic background. Magnetic spirals offer a particularly clean setting, as they simultaneously host propagating helimagnons and a gapless Goldstone mode, i.e., the sliding of the spiral. Here, we show that once a helimagnon is excited and the spiral slides, the motion, which rotates the local spin frame, produces a thresholdless rotational Doppler shift of the helimagnon spectrum, with even winding for the linear mode and odd winding for its second harmonic. Stronger excitation generates one finite-momentum phason. Mixing of that phason with the helimagnon produces a helimagnon frequency comb only inside a finite driving window, spaced by the phason frequency fixed by the spiral length rather than by the sliding rate. These two scales are separable in a microwave experiment: the Doppler offset tracks the sliding velocity, whereas the comb spacing scales inversely with the length of the spiral.
Read the full abstract

All authors Yang Liu, Jie Liu, Tianci Gong, Zhaozhuo Zeng, Xiaoping Gao, Kai Chang, Peng Yan

Robotics

Lodos: A Lightweight Simulator Designed for Contact-Rich Interactions with Detailed Meshes

In recent years, massively parallelized rigid-body simulation has played a central role in robotics research, enabling rapid progress in areas such as reinforcement learning, stochastic optimal control, and randomized smoothing-based trajectory optimization. Particularly in contact-rich settings, the scalability of such simulations in large part depends on well-optimized representations of collision geometry, often requiring a coarse approximation of the underlying surfaces using primitives such as spheres, capsules, cylinders, and boxes. Whenever the complexity of the surfaces or the precision demands of the application render such approximations insufficient, it is instead necessary to employ mesh-mesh collisions, which impose a significant reduction in throughput. This paper proposes a simulation framework designed from scratch to address this bottleneck, with novel routines for broadphase and narrowphase collision detection as well as the subsequent contact dynamics computation. As experimentally validated, the framework matches existing well-established simulators in terms of physical faithfulness, and can achieve approximately an order of magnitude higher throughput in the detailed mesh-mesh collision setting that it specializes in. The source code is available at https://github.com/martius-lab/lodos.
Read the full abstract

Materials Science

Magnetic-Field-Tuned Critical End Point and High-Field Phases in HoMn6Sn6

Topological Kagome RMn6Sn6 materials, where R is a rare-earth, exhibit a rich variety of magnetic phases which provide an opportunity to understand how different spin textures affect topological and transport properties through the application of magnetic field H. In this work, we report experiments on single-crystal HoMn6Sn6, made possible by pulsed magnetic fields of up to 60 T, that permit high-field features of the field-temperature phase diagram to be studied in detail using magnetometry. Complementary, sensitive transport experiments over the same field range employ focused-ion-beam fabrication of novel strain-relieved HoMn6Sn6 devices. In agreement with meanfield calculations, we find a metamagnetic transition at 45 T with H||c between canted phases that ends in a critical end point at about 90 K. As far as we are aware, this is the only critical endpoint that is observed within the RMn6Sn6 kagome metals, and its temperature and field far exceed those of analogous phenomena in other materials. In the low-temperature metamagnetic region, HoMn6Sn6 shows large magnetoresistance, reaching 900 % at 60 T, in concert with Shubnikov de Haas oscillations that suggest a slight Fermi-surface contraction as the metamagnetic transition is crossed. A phase analysis of the oscillations suggests non-trivial band topology, although a firmer attribution must await detailed bandstructure calculations. Near the critical end point, we find deviations from Fermi-liquid behavior in the temperature-dependent magnetoresistance. In the metamagnetic region, the anomalous Hall contribution is not proportional to magnetization, and peaks at about 40 K, suggesting the role of enhanced spin fluctuations near the critical point. At higher temperatures, the anomalous Hall effect reappears at low fields, showing that the magnetization provided by the Ho and Mn sublattices independently affect the conductivity.
Read the full abstract

All authors Joanna Blawat, John Singleton, Junho Seo, Charles Mielke III, Tyler J. Slade, Paul C. Canfield, Philip J. W. Moll, Ross D. McDonald, Robert J. McQueeney

Human-Computer Interaction

Bidirectional Modeling of mmWave Human Signals for Enhancement & Synthesis

Millimeter-wave (mmWave) is attractive for human-centered robotics in smoke, darkness, and privacy-sensitive indoor spaces: it measures reflections rather than visually identifiable imagery and remains usable without light. Deployed systems, however, face two complementary needs that run in opposite directions. Enhancement starts from a raw range-angle-Doppler (RAD) heatmap, where human returns coexist with nonhuman structure, and aims to isolate the human signal by removing nonhuman returns, yielding a clean RAD input for downstream perception. Synthesis starts from denser human geometry and aims to synthesize natural RAD observations, including environmental and hardware features, thereby expanding scarce supervision data. Although the two tasks reverse direction, both require modeling how human and nonhuman reflections coexist in the same measurement. Among RAD axes, Doppler provides motion information native to the modality, with the potential to support this modeling. We therefore cast both tasks on the dual-domain RAD heatmap and propose HeatWave, which injects Doppler when it is locally reliable, and routes Doppler asymmetrically for enhancement and synthesis within one shared backbone. This design improves both tasks: enhanced observations reduce mean per-joint position error by 11.4% relative to the raw recording, and synthesized samples that replace half of the real radar labels recover most of the activity accuracy lost to the reduced budget. Relative to an unspecialized dual-task heatmap baseline, the enhancement improves motion-critical samples under metallic clutter by 2.3times more than when geometry already suffices, and synthesis follows the same ordering with a narrower spread, providing the community with a research direction more closely aligned with the signal's characteristics.
Read the full abstract

All authors Shuntian Zheng, Hoang Hai Pham, Jiaqi Li, Xiaoman Lu, Shuai He, Yu Guan

Data Structures and Algorithms

As Easy as Rank One: Optimal Algorithms for Matroid Secretary and Single-Sample Matroid Prophet

We resolve the matroid secretary conjecture in its strong form and the single-sample matroid prophet inequality problem, achieving competitive ratios $e$ and $2$, respectively. Both ratios are optimal, even for rank-one matroids. For matroid secretary, our ordinal algorithm accepts the elements of the canonical optimum independently, each with probability $1/e$. It needs only the number of elements in advance and uses an expected polynomial number of independence queries on arrived elements. For single-sample matroid prophet inequalities, our algorithm uses one sample per independent value distribution and $O(n^2)$ independence queries on a known $n$-element matroid. It works for every arrival order fixed independently of the samples, values, and algorithmic randomness. The selected elements and their values have exactly the joint distribution obtained by independently retaining each element–value pair in the canonical optimum of a fresh product draw with probability $1/2$. Our main algorithms maintain a reference state whose greedy solution contains every accepted element. Its updates, made even after rejections, preserve exact distributional invariants. A short reversible-update argument gives a $4$-probability-competitive secretary algorithm, already resolving the original conjecture; a related invariant gives the optimal single-sample prophet inequality. For secretary, we characterize the distributions achievable by ordinal algorithms through a finite polytope. A continuous-time formulation then lets us design an accepted-set distribution achieving the optimal ratio $e$ and realize it efficiently. We also obtain guarantees for matroid intersections.
Read the full abstract

All authors Hamed Abdi, Kiarash Banihashem, MohammadTaghi Hajiaghayi, Zhiyi Huang, Danny Mittal, Sahil Singla

Robotics

Lodos-v2: An Efficient Primal Formulation of the Exact Frictional Contact Problem

Lodos-v2 is a GPU-based simulation framework for massively parallel rigid-body dynamics with contact. It is specifically designed to achieve high throughput in scenes involving collisions among large numbers of detailed meshes. Its predecessor, Lodos-v1, had two main limitations. First, it solved the standard convex relaxation of cone-complementarity-based contact dynamics, thereby inheriting well-known physical artifacts, including lift-off and artificial compliance in the normal direction, as well as creep in the tangential direction. Second, its contact-group reparameterization scheme, which rescales spring-damper forces using a single factor per collision pair, could not be integrated with other constraints required by a full-fledged simulator, such as joint limits and welds. Lodos-v2 addresses both limitations through two main contributions. First, it introduces an efficient algorithm for solving the exact frictional contact problem by sequentially solving a cascade of convex relaxations in primal form within an outer proximal minimization loop. Second, it extends the contact-group reparameterization scheme of Lodos-v1 to the primal formulation of cone-complementarity-based contact dynamics, enabling its integration with the other constraints of a full-fledged simulator. Experiments demonstrate that these contributions result in throughput improvements over established rigid-body simulation frameworks. The source code is available at https://github.com/martius-lab/lodos.
Read the full abstract

Systems and Control

Extending Grid Foundation Models for Phasor-Domain Dynamic Simulation

Grid Foundation Models (GridFMs) are emerging as a powerful framework for learning and transferring knowledge across power grids and grid analysis tasks. This work extends GridFMs from steady-state analysis to dynamic simulation. We present DynGridFM, which uses component-wise tokenization and a structured latent operator to learn shared dynamic representations across heterogeneous grids and enable transfer to unseen systems. Pretrained on nine grids, DynGridFM is adapted to a held-out 48-machine grid, where fine-tuning on 64 trajectories achieves accuracy comparable to training from scratch on 128. On a single GPU, it achieves 924x to 1232x speedup over numerical simulators on one CPU core.
Read the full abstract

All authors Boyang Zhao, Yifan Zhou, Alban Puech, Tilman Bockhacker, Hendrik Hamann

Machine Learning

Reconstructing LLM Lineage Trees via Fine-Tuning Path Inference

Fine-tuned (FT) derivatives of large language models (LLMs) are now released in large numbers, yet their provenance relies on self-reports and cannot be verified. Misattributed lineage leaves a verifier unable to audit outsourced fine-tuning, confirm license compliance of derived models, or trace vulnerabilities that a parent passes to its children. All three require the direction of derivation, not merely its presence. Existing output-only methods are limited to pairwise derivation tests or undirected lineage estimation, while recovering a directed tree requires white-box access to model weights. We formalize the problem of reconstructing a directed model tree, including internal nodes, under gray-box access, where only teacher-forced log-likelihoods are observed and weights are never inspected. Fine-tuning raises the log-likelihood of the training data, and a child inherits this increase from its parent. Building on this property, we define an asymmetric per-task log-likelihood deviation between any two models as a trace of fine-tuning history, and derive from it an edge weight for each ordered pair of models. We propose TRACE-Edge, which exactly solves for the maximum spanning directed tree under these edge weights, with neither hyperparameters nor tree-structure constraints. We evaluate it on five configurations totaling 25 model trees built from three base models and seven tasks. TRACE-Edge achieves a parent-match accuracy of 0.74 when probed with the fine-tuning examples, and on average outperforms an output-similarity baseline even with held-out probes drawn from the same distribution as the training data.
Read the full abstract

All authors Sota Ogawa, Shojiro Yamabe, Daiki Nishiyama, Jun Sakuma

Artificial Intelligence

Rethinking Agent Trajectory Compression: A Workflow-Aware Perspective

As long-horizon agents increasingly rely on multi-step tool interactions to accomplish complex tasks, their execution trajectories continuously accumulate structured outputs and intermediate artifacts, leading to growing context overhead. Existing compression methods primarily rely on text importance or fixed truncation rules, often overlooking cross-step dependencies within workflows and potentially discarding critical information required for subsequent task execution. To address this challenge, we propose WAAC, a workflow-aware framework for agent trajectory compression. WAAC structurally models successful historical trajectories to extract reusable workflow compression experience and incorporates online backward dependency analysis to dynamically identify information retention requirements. By combining current dependency constraints with historical experience, WAAC selects appropriate compression strategies for different types of historical information. Experiments on PinchBench and Odysseys demonstrate that WAAC achieves an overall score of 87.6 and a perfect completion rate of 19.0%, respectively, while reducing execution costs by 19.7% and 29.1% compared with the strongest baseline methods. These results demonstrate the effectiveness of workflow-aware trajectory compression in improving task performance and cost efficiency for long-horizon agents. Code and resources are available at https://anonymous.4open.science/r/waac-6527/.
Read the full abstract

All authors Qijia Zhuang, Rui Liu, Wenqi Hu, Yuxiang Ren

Distributed Computing

Adaptive Scheduling Pipeline for Multi-Instance Asynchronous Reinforcement Learning

We present the Adaptive Scheduling Pipeline (ASP), an offline rollout scheduler for multi-instance asynchronous reinforcement learning (RL) of large language models (LLMs). ASP jointly reduces three sources of rollout underutilization - padding bubbles from intra-minibatch length heterogeneity, skewness bubbles from long-tail outputs, and sorting bubbles from fluctuations in training-queue occupancy induced by output-length-sorted dispatch - through a single offline preprocessing pass. This enables efficient multi-instance rollout that overcomes the early scaling plateau of single-instance inference while reducing fluctuations in the supply of generated samples to the trainer. We integrate the scheduler into an asynchronous architecture that decouples rollout from training while keeping policy lag observable and bounded. Within our asynchronous architecture, ASP's scheduling reduces mean iteration time by 10.9%. The complete system improves end-to-end throughput by up to 3.22 times and reduces iteration time by up to 76.8% relative to MindSpeed-RL, while reducing iteration time by up to 62.5% relative to verl. It also scales more robustly with additional generation resources than these baselines and shows no visible convergence degradation in the measured bounded-lag regime. ASP has since been used in an internal telecom-domain RL post-training workflow, reducing the reported training period from more than three months to two months and producing a domain-enhanced model for knowledge answering and tool calling.
Read the full abstract

All authors Salah Chikhi, Luis Zuin, Entong Li, Zeng Li

Robotics

What Limits mmWave Human-State Sensing for Robot Perception?

Assistive robots and search-and-rescue platforms require reliable human state sensing in smoke, darkness, and privacy-sensitive environments, where cameras and LiDAR may be unreliable Millimeter-wave radar offers an alternative that works without ambient light, penetrates smoke, and does not capture visual appearance. However, the compact arrays that fit robot size, power, and cost budgets have limited apertures, producing joint-localization errors on the order of decimeters at a few meters, large enough to compromise the safety margins and tracking accuracy a robot depends on. Existing pipelines reduce pose error through frame fusion, multiple viewpoints, return selection, or stronger estimators, but examine one technique at a time, without relating its effect to a shared measurement limit or separating gains from added information from gains from estimator completion. We address this gap with a framework that models the human and environment as one time-varying scattering field and states a Cramer-Rao type error floor for joint localization as a function of bandwidth, aperture, signal-to-noise ratio, fused frame count, phase coherence, and visibility of retained returns. We then use existing frame fusion, multi viewpoint, return selection, and pose estimation methods as matched probes of this floor rather than as competing pipelines. Under the evaluated conditions, experiments on three datasets and a live deployment show these strategies move joint localization error in the predicted directions, for example lowering mean per-joint position error from about 97 to 65 mm with a second viewpoint, while a stronger estimator mainly lowers overall error by completing joints the measurement cannot resolve. These results give the robotics community a principled basis for deciding whether to prioritize observation design or estimator capacity when deploying reliable mmWave human sensing.
Read the full abstract

All authors Shuntian Zheng, Guangming Wang, Hongyang He, Shuai He, Yu Guan

Combinatorics

Jaeger conjecture and Jaeger-Zhang conjecture

We prove that every signed planar graph $\widehat G$ of negative girth at least $\ell\geq 3$ satisfies $\chi_c(\widehat G)<\frac{2\ell}{\ell-2}$. As a consequence, for every integer $k\geq 1$, every planar graph $G$ of odd girth at least $4k+3$ satisfies $\chi_c(G)<\frac{2k+1}{k}$ and hence admits a homomorphism to $C_{2k+1}$. Our proofs build on recent results of Jafari on circular flows. In particular, Jafari proved that every $(4k+2)$-edge-connected graph has circular flow number strictly less than $2+\frac{1}{k}$. We revisit his theorem, providing an alternative proof that emphasizes the underlying graph-theoretic ideas. We also extend the result to signed graphs and weaken the connectivity assumption by imposing a lower bound only on the size of negative cuts. The key step is a small-correction lemma associated with a distribution on spanning trees. A splitting argument first gives the required correction in an even-degree graph; doubling the edges removes the degree-parity assumption. Plane duality then gives the coloring theorem. We discuss tightness of bounds in the end.
Read the full abstract

Computer Vision

Align3D: Image-Structure Co-Aligned 3D Scene Generation with Observability-Aware Fidelity-Preserving Guidance

We present Align3D, an image-conditioned diffusion model that generates 3D indoor layouts that are both faithful to a single input photograph and physically plausible beyond the observed view. From the photograph we infer a confidence-calibrated scene graph and train a set-diffusion layout denoiser with image-structure co-alignment where confidence-weighted modality fusion balances image and scene-graph alignment using predicted confidences, while observability-aware fidelity-preserving guidance projects conflicting scene-quality gradients and reweighs them with an object-level confidence mask so under-observed degrees of freedom absorb physical priors without undoing the image-graph match. On 3D-FRONT, Align3D improves image-alignment and scene-quality metrics over recent baselines; once trained, it maps previously unseen photographs to layouts in a single feed-forward pass.
Read the full abstract

Artificial Intelligence

Mind the Divergence: Divergence-Guided Reliability-Aware Fusion and Adaptive LLM Prompting for Conversational Emotion Recognition

Multimodal emotion recognition in conversations (MERC) integrates textual, acoustic, and visual cues to identify utterance-level emotions. However, existing approaches generally do not explicitly adapt multimodal processing to the degree of inter-modal predictive disagreement. In real conversations, semantic incongruity, concealed emotion, noisy observations, and ambiguity can affect modality-specific emotional evidence differently, while the reliability of individual modalities varies across samples. Rather than identifying each source of inconsistency separately, we model their common consequence for emotion recognition as predictive disagreement among unimodal emotion distributions over the same emotion-label set. Such disagreement makes indiscriminate fusion prone to propagating unreliable or competing evidence. Inspired by the Brunswik Lens Model, we propose MiD-ERC, a divergence-centric framework that quantifies inter-modal predictive disagreement using a normalized divergence score and estimates sample-specific modality reliability. MiD-ERC uses these signals to coordinate divergence-guided cross-modal gating, divergence-modulated reliability weighting, and divergence-conditioned adaptive LLM prompting. The framework is trained end-to-end using only standard emotion annotations, without requiring additional conflict labels, auxiliary data construction, or external retrieval modules. Experiments on IEMOCAP and MELD demonstrate that MiD-ERC achieves state-of-the-art performance, with particularly pronounced improvements on high-divergence samples.
Read the full abstract

All authors Jingwen Wang, Jiao Li, Yuhang Qian, Yuyin Wang, Min Zhang

Econometrics

Instant Payments After a Betting Ban: Evidence from Brazil

Brazil has been facing major problems with delinquency, household debt, high interest rates, consumption, and access to credit. Betting has emerged in public debate as a possible contributing factor. On September 25, 2026, the federal government banned betting nationwide. We investigate how the ban affected instant payments (Pix). Our main finding, under the linear-trend specification, is that cumulative transaction counts over 37 days were approximately 8.37% below forecast, while settlement values were 10.83% lower. Both deviations were statistically significant under this specification.
Read the full abstract

Computational Complexity

Quantum Interactive Proofs of Proximity

We initiate the systematic study of quantum interactive proofs of proximity (QIPPs). These are quantum interactive proof systems with sublinear communication, where the verifier accepts inputs in a property $\Pi$ and rejects inputs $\varepsilon$-far from $\Pi$, while only querying the input a sublinear number of times. 1. We show that all logspace-uniform $\mathbf{NC}$ properties admit QIPPs with query, communication and time complexity all $\widetilde{O}(n^{2/5})$. This improves on the best known classical total complexity of $\widetilde{O}(\sqrt n)$, which is optimal in the worst case for polynomial-time verifiers under the PRG assumption of Kalai and Rothblum (CRYPTO, 2015). This resolves an open question of Dall'Agnol, Gur, Roy Moulik, and Thaler (Quantum, 2022) in the affirmative. 2. Assuming the existence of a post-quantum subexponentially secure PRG computable in logspace-uniform $\mathbf{NC}$, we prove that, for every nondecreasing, logspace-computable query bound $q(n)\le n^{1/2-\Omega(1)}$, there exists a property $\Pi$ in logspace-uniform $\mathbf{NC}$ and a constant $\varepsilon>0$ such that no polynomial-time $\varepsilon$-QIPP for $\Pi$ can achieve query complexity $q(n)$ and communication complexity $c(n)$ satisfying $q(n)+c(n)=o(n^{1/3})$. 3. We establish a bi-directional connection between QIPPs for logspace-uniform $\mathbf{NC}$ and query-model QIPs for parity. A QIP for $k$-bit parity with total complexity $k^{2/3-\Omega(1)}$ yields QIPPs for every logspace-uniform $\mathbf{NC}$ property with total complexity $n^{2/5-\Omega(1)}$. Conversely, achieving total complexity $o(n^{2/5})$ within the standard $\mathsf{PVAL}$-based framework requires improving the known query-communication tradeoff for parity. Additionally, we show that almost every size-$2^{\beta n}$ property ($0<\beta<1$) requires linear total QIPP complexity at constant proximity.
Read the full abstract

Robotics

STRIDE: Learning Humanoid Locomotion in Minutes with Acceleration-Based Rewards

Learning locomotion directly on a humanoid robot is a promising direction to avoid the sim-to-real gap. However, locomotion rewards in Reinforcement Learning (RL) typically track the base velocity, which cannot be measured onboard without an external motion capture system or well-calibrated state estimator. This is why we introduce \method{}, which learns humanoid locomotion from minutes of samples with a reward that only requires the onboard Inertial Measurement Unit (IMU). Instead of tracking the base velocity, STRIDE rewards the robot for following a commanded acceleration profile. To learn efficiently from this reward, we use the off-policy RL algorithm XQC and combine a high gain setting for the joint position controllers with left-right symmetry augmentation. In simulation, STRIDE learns to walk on the Unitree G1 and Booster T1 humanoids within 14 minutes of samples. Compared to learning without these two components, it reduces the falls during training by 81%, while without them the robot learns only in over an hour.
Read the full abstract

Statistical Methodology

Competitive Balance Beyond Standings: Bayesian Inference for Latent Tiers and Structural Heterogeneity

Competitive balance has long been a central concern in sports economics, tied to outcome uncertainty, fan engagement, and the way leagues design their policies. Most existing measures of competitive balance are built from standings or winning percentages, which say little about how overall parity relates to the way teams sort into latent competitive tiers. We propose a Bayesian framework based on the Bradley-Terry model that jointly estimates the number of competitive tiers and team memberships. To capture the overall degree of competitive heterogeneity, we introduce a structural balance parameter defined on the unit interval within a hierarchical model, and we establish theoretical results giving this parameter a clear interpretation:it is directly tied to the separation between tiers, and overall heterogeneity depends on both the size and frequency of cross-tier differences. Importantly, structural balance is distinct from the information content of the schedule itself, so leagues with different numbers of teams and games can be compared on equal footing. Inference is carried out with an efficient posterior sampling algorithm that jointly handles team abilities, tier structure, home advantage, and structural balance. Simulation studies validate the proposed method, showing accurate recovery of the latent competitive structure and correct performance of the structural-balance testing procedure. Applied to the NBA and NFL, it reveals that the two leagues exhibit similar global structural balance despite very different competitive organizations: the NBA has three latent tiers, while the NFL has two broader ones, and the NFL's sparser schedule produces considerably more posterior uncertainty. Overall parity, it turns out, can mask quite different forms of competitive stratification-something standings-based measures alone cannot detect.
Read the full abstract