Nuclear Theory
A novel method to determine the 4th moment of the 208Pb charge density from low-momentum-transfer electron scattering
Rika Danjo, Toshimi Suda and Toshio Suzuki
We propose a new method to determine the 4th moment of the
nuclear charge-density distribution of 208Pb, with high precision from low-momentum-transfer elastic electron scattering.
Soft Condensed Matter
On the ratio between longitudinal and transverse Ioffe-Regel frequencies in glasses
Licun Fu, Mingyu Zhu, Xinyu Chang et al. (+1 more)
The temperature dependence of thermal conductivity in glasses differs characteristically from that in
crystals, and is largely controlled by how sound waves are damped. The Ioffe-Regel (IR) frequency sets the high-frequency limit of well-defined sound waves. Although the transverse IR frequency has been established to coincide with the
boson peak frequency and reported to be far below the longitudinal IR frequency, the quantitative relation between longitudinal and transverse IR frequencies has remained unknown. Here we examine this relation in two- (2D) and three-dimensional (3D) model glasses with vastly different stability. We observe that, in 3D glasses, the longitudinal-to-transverse IR frequency ratio is approximately equal to the ratio of the low-frequency transverse to longitudinal sound attenuation coefficient, whereas this correspondence is not observed in 2D glasses. Moreover, we find that the glass stability strongly controls the longitudinal-to-transverse IR frequency ratio in both 2D and 3D glasses: the ratio decreases significantly with increasing glass stability and approaches unity for the most stable glasses under study. Accordingly, in sufficiently stable glasses, the longitudinal sound attenuation should be comparable to the transverse attenuation, and therefore cannot be treated as negligible, challenging common assumptions adopted especially in theoretical studies. Our work is another demonstration of the glass stability as a key parameter for investigating glass properties, and cautions against simply extrapolating observations from poorly annealed glasses to stable glasses.
Read the full abstract ▾
All authors Licun Fu, Mingyu Zhu, Xinyu Chang, Lijin Wang
Robotics
Active Spatial Inspection for Effective and Efficient Embodied Exploration
Wenbin Wang, Xiang Bai, Yizhao Wang et al. (+5 more)
Achieving high task success and efficiency remains a central pursuit in
embodied exploration. Existing frameworks typically guide
agent behavior through a spatially coarse and indirect assessment of suggestive cues and directions, yet such designs may struggle to judge cue sufficiency and the need for further inspection, leading to early termination or excessive continuation and ultimately reducing task success and efficiency. This work rethinks
embodied exploration from a spatially explicit standpoint and introduces the spatial-inspection-guided ACtive Exploration (ACE) framework coupling evidence-grounded perception with exposure-informed movement and establishing a spatially resolved decision paradigm. Evidence-grounded perception strengthens inferential validity by integrating granular localization with focused verification, turning suggestive cues into precise decisive visual support. Exposure-informed movement promotes directional judgment through prospective prioritization and retrospective suppression, selecting promising directions for efficient spatial progress. ACE alleviates the tension between preventing early termination and avoiding unnecessary continuation. Extensive experiments demonstrate that ACE achieves 18.0% higher
navigation task success and 10.3% higher exploration efficiency for question answering than prior state-of-the-art baselines, advancing effective and efficient
embodied exploration.
Read the full abstract ▾
All authors Wenbin Wang, Xiang Bai, Yizhao Wang, Hang Sun, Dong Ren, Jie Qin, Qingquan Li, Bing Wang
Fluid Dynamics
Stabilization of Yih's Viscosity Jump Interface in a Channel
Miroslav Krstic
Two immiscible viscous fluids in pressure-driven channel flow can be unstable through their interface when their viscosities differ, at arbitrarily small Reynolds number — the instability Yih found in 1967. The unstable state is the interface itself: a curve inside the domain, separating the two fluids, whose displacement is governed by a partial differential equation of its own and which moves the fluid domain with it. Actuation is at one wall only, so the state to be controlled is reached across a fluid. We stabilize this interface at any rate below the limit set by the streamwise mean flow, by static feedback of the two velocity components of that wall, through a Fredholm backstepping transformation acting jointly on the two layers. The target is the two-fluid Stokes problem, shifted, with the transmission of stress between the fluids kept and the coupling of the interface into them removed. Two findings about the two-fluid spectrum carry the design: joined at the interface, the
eigenvalue sequences of the two layers, which collide for a dense set of layer ratios when the layers are uncoupled, avoid each other uniformly, so no degeneracy has to be assigned; and the slowest mode left to viscosity is not a bulk mode but the capillary relaxation of the interface, whose rate is linear rather than quadratic in the wavenumber, which sets the band of actuated wavenumbers. The controller is tested on the nonlinear two-fluid channel, switched on against a saturated interfacial wave.
Read the full abstract ▾
Statistical Methodology
Bridging Local and Population Causal Effects: A Proximal Instrumental Variable Approach
Zixuan Yao and Guosheng Yin
Instrumental variable (IV) methods address treatment endogeneity, but with non-compliance and heterogeneous treatment effects a binary instrument generally identifies the local average treatment effect (LATE) among compliers rather than the population average treatment effect (ATE). When treatment effects and compliance probabilities are heterogeneous and dependent through latent factors, the ATE need not be identified by IV variation alone. We develop a proximal instrumental variable framework that uses proxies for these factors to adjust for the compliance weighting in LATE and recover the ATE. We show that the ATE is identified through either an outcome bridge capturing latent treatment effect heterogeneity or a compliance bridge representing inverse compliance probabilities, and combining the two yields a doubly robust representation. We derive the efficient influence function under known and unknown instrument propensities. For estimation, we propose regularized kernel minimax bridge
estimators, with orthogonalized conditional moments for estimated propensity, and a three-way sequential cross-fitting scheme. We establish projected bridge-risk bounds, and give conditions for asymptotic normality and semiparametric efficiency. The framework provides a proxy-based route from local IV effects to population causal effects under latent treatment effect and compliance heterogeneity.
Read the full abstract ▾
Popular Physics
Mass in the Light of Special Relativity
Francisco Caruso and Vitor Oguri
The text analyzes the concept of mass within Special Relativity, initially deconstructing the interpretation of Einstein's equation ($\varepsilon_\circ = m c^2$) as a principle of equivalence. Based on the adoption of the conservation laws of momentum and energy, the non-additivity of mass in composite systems is established. While in atomic and
nuclear systems the total mass is less than the sum of its constituents, in the subnuclear domain the opposite occurs: the positive potential energy of confinement of
quarks causes the mass of a
hadron to be greater than the sum of the masses of its
quarks; in the case of nucleons, it is much greater. It is concluded, then, that the mass of baryonic matter in the observable Universe comes, for the most part, from the energy associated with strong interactions.
Read the full abstract ▾
Combinatorics
The Three-Dimensional Erdős Box Problem Has Exponent $11/4$
Dean Menezes
Let $z(n)$ be the maximum number of edges in a tripartite $3$-uniform hypergraph with $n$ vertices in each part and no copy of $K_{2,2,2}^{(3)}$ (a “box”). Erdős (1964) proved that $z(n) = O(n^{11/4})$, whereas the best previous lower bound, due to Katz, Krop, and Maggioni (2002), was $\Omega(n^{8/3})$. For each $q = 2^m$, we construct a box-free hypergraph with $q^4$ vertices in each part and $q^{11}$ edges, showing that $z(n) = \Theta(n^{11/4})$. The construction uses the power map $\tau(s) = s^{q^2-q+1}$ on $\F_{q^3}$, which sends the fibers of $s \mapsto \tau(s+1) + \tau(s)$ to pairwise skew affine lines over $\F_q$.
Read the full abstract ▾
Artificial Intelligence
Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design
Hongyang Du, Lan Yan, Christian Flores et al. (+1 more)
Professional graphic design is a long-horizon agentic task in which structured, editable artifacts emerge from many interdependent actions, yet outcomes admit no reliable programmatic oracle. We introduce a continual adaptation framework in which a frozen frontier model operates professional design software through more than 230 tools, while an external procedural
memory of natural-
language skills accumulates and refines reusable design procedures from experience. The
memory widens by acquiring procedures for recurring uncovered subtasks and deepens by revising existing procedures against their own successful and failed executions, while a matched replay gate admits only changes that repair failures without regressing observed successes. Five rounds over 1,406 real user briefs and 1,869 automatically graded trajectories, with no weight updates and no human labels, grow the bank from 76 documentation-derived skills to 139 and raise GenEval2 execution success on Claude-Sonnet-4 from 72.7% to 99.3% (+11.99 points in generation quality), with 61.8% and 67.6% win rates against the no-skill
agent across four specialized design
benchmarks on Claude-Sonnet-4 and Claude-Opus-4.6. We further show the two mechanisms are effective in combination: on 200 held-out briefs from user-traffic
benchmark, widening or deepening alone reaches a 49.4% / 48.6% win rate over the no-skill
agent, while their combination reaches 58.5% (p = 0.025). Procedural
memory offers a practical route to continual adaptation of
agents under noisy, unverifiable feedback.
Read the full abstract ▾
All authors Hongyang Du, Lan Yan, Christian Flores, Asim Kadav
Robotics
SeeQ: Training Generalist Value Functions for Long-Horizon Robotic Manipulation
Saksham Singh, Zheyuan Hu, Max Sobol Mark et al. (+3 more)
Despite rapid progress, generalist
robot policies remain brittle on complex, long-horizon tasks that comprise multiple stages or require repeated attempts and deliberation on the same underlying stage before success. Q-value functions can improve these policies by ranking candidate actions or guiding policy improvement, but learning from sparse task-level rewards entails long credit-assignment horizons, difficult Bellman backups, and broad data-coverage requirements. We introduce SeeQ (Subtask-elicited Q-functions), which instead learns Q-values for the currently active subtask. This shortens the value-prediction horizon and enables effective learning with temporal-difference (TD) objectives. During training, subtask-level annotations present in offline
robot data provide the decomposition and enable learning from broad, potentially suboptimal
robot datasets. To eliminate the need for human annotations or modular subtask prediction systems at test time, our Q-function architecture is trained to autoregressively predict the active subtask in natural
language before estimating its value. We instantiate SeeQ using a base
vision-
language backbone, pretrain it on diverse open-source
robot manipulation data, and finetune it on downstream tasks. Across four real-world manipulation tasks on two bimanual
robot platforms, the SeeQ value function substantially improves best-of-N policy steering.
Read the full abstract ▾
All authors Saksham Singh, Zheyuan Hu, Max Sobol Mark, Jeffrey Yu, Zackory Erickson, Aviral Kumar
High Energy Physics (Theory)
Multi-Boundary Spinning $AdS_5$ Black Holes, Strongly Coupled Plasma Balls and a dual locally de Sitter spacetime
Andrés Anabalón, Guillermo Buckle, Annelies Fullgraff et al. (+1 more)
We construct a new family of exact spinning black hole solutions in five-dimensional General Relativity with a negative cosmological constant, which are dual to strongly coupled spinning
plasma balls on four-dimensional Minkowski spacetime. In analogy with the recent discovery of such black solutions in four bulk dimensions, the boundary Lorentz factor acts as a new coordinate in the bulk, giving rise to a new UV region that, in our case, is locally a three-dimensional de Sitter spacetime times the real line. The dual energy-momentum tensor in this region describes a conformal anisotropic fluid that satisfies the weak and strong energy conditions.
Read the full abstract ▾
All authors Andrés Anabalón, Guillermo Buckle, Annelies Fullgraff, Marcelo Oyarzo
Computer Vision
MintAct: A Unified Visual Agent for Digital Environments
Mingfei Gao, Rui Tian, Haiming Gang et al. (+11 more)
We present MintAct, a family of
vision-
language models that unifies UI grounding, multi-step
navigation across mobile, desktop, and web, and visual tool use, trained at 2B, 4B, and 8B scales. Through careful design of our environments, data, and training recipes, MintAct models match the performance of per-domain specialists across all of these capabilities. To enable this, we develop a scalable environment and reinforcement learning (RL) infrastructure. On the environment side, we host hundreds of concurrent instances across heterogeneous per-domain backends, serving both trajectory data collection and online RL. To enable efficient and scalable RL training, an asynchronous framework keeps explicit control over the cross-domain training distribution and remains stable under noisy environment feedback and off-policy drift. Experimental results show that MintAct achieves state-of-the-art performance (48.9 on OSWorld-Verified) across a wide range of
benchmarks at comparable model sizes.
Read the full abstract ▾
All authors Mingfei Gao, Rui Tian, Haiming Gang, Bohan Zhai, Le Zhang, Yuanzheng Gong, Di Feng, Ege Özsoy, Kaixin Ma, Vishwesh Kirthivasan, Oğuzhan Fatih Kar, Roman Bachmann, Anders Boesen Lindbo Larsen, Afshin Dehghan
Computational Geometry
On (Directed) Width-Parameters of Geometric Spanners
Kevin Buchin, Carolin Rehs and Torben Scheele
To speed up
algorithms on
geometric graphs, it is common to approximate the complete Euclidean graph while maintaining certain
geometric properties. A (directed) $t$-spanner $G$ for a point set $P$ in the Euclidean space is a (directed) graph such that for every pair of points, the shortest path in $G$ is at most a factor $t$ longer than the Euclidean distance between those points. In this paper, we investigate $t$-spanners that are bounded by certain graph parameters. Let $\kappa$ be a graph parameter. We show that for path-width, branch-width and cut-width there is an $\mathcal{O}(n/k^{d/(d-1)})$-spanner $G$ on $P$ with $\kappa(G)=k$ and that this is asymptotically worst-case optimal. In $\mathbb{R}^2$ we show the same bounds for planar graphs of clique-width or rank-width $k$. In contrast, for tree-depth, we show that there are sets of points for which the dilation cannot be bounded. Therefore, we investigate computing a spanner with tree-depth $k$ and minimum dilation. We show that already for tree-depth $3$ this problem is NP-hard to approximate within any factor strictly less than $\sqrt{2}$, and present an XP-
algorithm to compute for a given tree-depth $k$ a graph with dilation at most $2t^*$, where $t^*$ is the minimum dilation. We further extend these results to obtain directed $\mathcal{O}(n/k^{d/(d-1)})$-spanners $G$ with $\kappa(G)=k$ for $\kappa$ being directed tree-width, directed path-width or DAG-width and show that also in the directed case, this is asymptotically worst-case optimal.
Read the full abstract ▾
Computation and Language
Structuring occupational accident narratives for cross-sector safety analysis: Transferability of accident-process role classification
Aho Yapi, Pierre Latouche, Arnaud Guillin et al. (+1 more)
Introduction: Occupational accident narratives describe work situations, unfavourable conditions, accident events, and consequences, but differences in terminology and reporting practices hinder systematic analysis across sectors and organisations. This study examined whether a model developed in one occupational sector could identify the same accident-process information in unseen sectors and reporting environments. Method: French accident narratives were segmented into factual units and expert-annotated as work situation, explicitly reported unfavourable condition, accident event or deviation, or reported consequence. Models were developed on 42,244 factual units from 6,040 construction-sector narratives and evaluated without retraining on metallurgy, chemistry-plastics, and an independently collected company corpus. We compared a TF-IDF-based lexical model, frozen pretrained text representations, and task-adapted pretrained models. Results: Average balanced accuracy was 75.0% for TF-IDF, 76.9% for frozen pretrained representations, and about 85.7% after task adaptation. Across repeated runs, leading adapted approaches showed similar overall performance, with no method consistently outperforming the others. Performance was lower and more variable on the company corpus, where transfer also involved a different reporting environment and data source. Conclusions: Accident-process roles learned from construction narratives remained identifiable in other sectors and an independent organisational setting. Practical Applications: The framework can support assisted coding, expert review, cross-sector analysis, and prevention-oriented analysis of large accident-report collections.
Read the full abstract ▾
All authors Aho Yapi, Pierre Latouche, Arnaud Guillin, Yan Bailly
General Relativity and Quantum Cosmology
Rotating Neutron Star Migrations as a Standardized Test for 3+1 Numerical Relativity
Óscar H. Petit, José A. Font and Nikolaos Stergioulas
When simulating dynamically unstable
neutron stars, truncation errors can introduce perturbations that drive the
stellar configuration either toward gravitational collapse into a black hole or toward migration to a dynamically stable, lower-density configuration. This mechanism has become a standard
benchmark for validating general-relativistic hydrodynamics codes in the case of spherically symmetric
neutron-
star models. In this work, we extend the analysis to the more general case of uniformly rotating
neutron stars evolved in three spatial dimensions. We investigate these transitions using two equations of state: a cold polytrope and a hybrid prescription, which allow us to distinguish the barotropic response from the effects of shock-generated thermal pressure. We find that the migrated remnant remains close to uniform rotation after the transition, and we trace its quasi-radial pulsations through complementary diagnostics involving the central density, the rotation profile, and the axisymmetric gravitational-wave channel. We investigate the properties of these transitions and propose a standardized test for validating numerical-relativity codes by requiring the migration behavior to be consistent with the dynamics reported here.
Read the full abstract ▾
Statistical Mechanics
Universal Eigenvector Statistics of Non-Hermitian Random Matrices
Ze Chen, Zhenyu Xiao and Shinsei Ryu
Eigenvector overlaps quantify nonorthogonality and govern the response and dynamics of non-Hermitian systems. We extend the universality of non-Hermitian random
matrices to the statistics of these overlaps. We obtain analytical expressions for eigenvector overlaps in the spectral bulk and near the origin, covering ten
symmetry classes in the limit of large
matrix size. Using fermionic replica nonlinear $\sigma$ models, we relate these overlaps to Hermitian level statistics,
symmetry class by
symmetry class and topological sector by topological sector. Numerical calculations in various physical models support the universality of the normalized overlaps in the regimes studied. Our work establishes a duality between Hermitian level statistics and non-Hermitian eigenvector overlaps.
Read the full abstract ▾
Combinatorics
Stellahedral geometry of partially ordered sets
Tommaso Faustini, Luis Ferroni and Ludovico Piazza
We introduce a transformation on partially ordered sets, termed the stellahedral transform, with notable features. It preserves the properties of being Eulerian, Cohen–Macaulay, and of being the face poset of a polytope. Furthermore, it admits an explicit
geometric realization for convex polytopes and specializes to the construction that takes a simplex to the stellahedron. One motivation for this definition comes from the theory of toric $h$-
polynomials and (augmented) Chow
polynomials of Eulerian posets. We show that the right augmented Chow
polynomial of an Eulerian poset $P$ agrees with the toric $h$-
polynomial of the stellahedral transform of $P$. We use this perspective, together with $\mathbf{cd}$-index results due to Ehrenborg (2005) and Karu (2006), to prove two positivity results for augmented Chow
polynomials: for Gorenstein* posets they are unimodal, and for face posets of polytopes they are $\gamma$-positive. Along the way we provide negative answers to two open questions concerning Eulerian and Gorenstein* posets. First, the question on the nonnegativity of Eulerian Chow
polynomials, posed by Ferroni, Matherne, and Vecchi (2024). Second, the question posed by Athanasiadis and Kalampogia-Evangelinou (2023) on the real-rootedness of chain and Chow
polynomials of Gorenstein* posets: these examples provide a novel application of a technique introduced by Murai and Nevo (2014).
Read the full abstract ▾
Metric Geometry
Energy minimization for eight points on the sphere
Liudmyla Kryvonos, Lukas Liehr and Mitchell A. Taylor
We study the energy minimization problem for eight points on the unit sphere. For the logarithmic and Coulomb energies, we show that the unique global minimizer up to congruence is a square antiprism with height characterized by a unique stationarity equation. The proof is computer-assisted and fully verified in Lean. After this, we consider generalizations of the result to other important energies. For the Riesz $s$-energies, we provide a Lean-verified, non-computer-assisted proof that the square antiprism with height depending on $s$ is the unique global minimizer for all sufficiently large $s$, and a computer-assisted proof that this in fact holds for all $s\geq 0$. We also give examples of energies arising from completely monotonic potentials for which the square antiprism is not a global minimizer, answering in the negative a universality question of Cohn and Woo.
Read the full abstract ▾
Cryptography and Security
APort Vault: Benchmarking AI Agent Payment Authorization with the Open Agent Passport
Uchi Uchibeke
APort Vault is a
benchmark for payment authorization in tool-using AI
agents. It replays 4,371 attacks written by humans against a live payment
agent during a public capture-the-flag event, across 14 models from 8 labs, five policy configurations and two replay tracks, with and without a deterministic pre-action check implementing the Open
Agent Passport (OAP) specification. 225,964 evaluations completed. We report five distinct events per evaluation, because collapsing them is how an
agent benchmark produces a number that does not survive review. Requests are common and their rate differs far more across configurations than across models, though each attack exists at exactly one configuration so policy and attack cohort vary together: 10.9% of model-alone evaluations at Level 1, 3.0% at Level 2, 0.1% at Level 3, 79.4% at Level 4. On the 1,293 Level 4
prompts, each evaluated on every model, request rates run from 71.2% to 84.3%, and 809
prompts (62.6%) elicited a request from all fourteen models, each ending in a successful payment to the level's allowlisted recipient. The authorization boundary is where the conditions diverge. At Levels 2 to 4, transfers to recipients the passport did not permit number 140 of 76,842 with the model alone and 0 of 69,297 behind the layer, and 105 against 0 on 68,970 matched model,
prompt and track triples. The zero spans 790 source sessions, giving a per-session upper bound of 0.38%. It was not obtained by refusing payments: 25,370 payments executed behind the layer, while the policy denied 187 of the 25,640 transfer calls it evaluated, 148 of them for a forbidden recipient. We release the 225,964 evaluations, the level passports, the scoring code and the analysis script at huggingface.co/
datasets/aporthq/vault-
benchmark-v1 .
Read the full abstract ▾
Robotics
LIMBO: Learning and Internalizing Model-Free Barrier Objectives for Agile and Safe Whole-Body Control
Jake Gonzales, Arturo Flores Alvarez, Yu-Ming Chen et al. (+3 more)
Safe whole-body control requires coordinating collision avoidance and balance under high-dimensional, nonlinear dynamics–making safety certificates difficult to design and reuse across behaviors. We present LIMBO, a framework for synthesizing a state-action control barrier function and distilling its safety structure into a task policy. LIMBO learns the safety certificate from black-box transitions and a state-based failure specification over residual actions around a frozen base controller, making Q-CBF synthesis tractable in the full control dimension while placing the certificate in the task policy's control space. During synthesis, the learned safety value drives risk-guided sampling near the estimated boundary of recoverability; during task learning, it serves as a teacher that provides action-level safety feedback, yielding a robust task policy and alleviating the need for an online safety filter at deployment. We demonstrate LIMBO on a 29-degree-of-freedom
humanoid performing dodgeball avoidance and locomotion beneath low obstacles. Beyond scaling learned Q-CBFs to whole-body control, we show that risk-guided boundary sampling provides a theoretically grounded way to explore the edge of recoverability. Under the same safety specification, ceteris paribus, varying the sampling concentration produces strategies ranging from crouching to a novel backward-leaning limbo maneuver. In both settings, the learned policies transfer to hardware without online safety filtering, showing that learned safety synthesis scales to agile whole-body control.
Read the full abstract ▾
All authors Jake Gonzales, Arturo Flores Alvarez, Yu-Ming Chen, Aaron D. Ames, Lillian J. Ratliff, Manikantan Nambi
Astrophysics of Galaxies
The AGN Channel in 3D: Scattering Belts and the Importance of Eccentricity in the Black Hole Population
Connar Rowan and Martin E. Pessah
Active
galactic nuclei (AGN) are a promising origin for observed gravitational wave mergers. Current population synthesis models are limited to 1D N-body or Monte Carlo methods which rely on statistical approaches to resolving dynamical scatterings. We present three-dimensional hybrid $N$-body simulations of a population of black holes (BHs) surrounding an AGN using a new code in development, AGNBI, where close interactions are directly simulated for both single and binary objects. Our results show binary formations occur throughout the AGN disk, particularly where the migration time is long. While BHs efficiently migrate and pair up in migration traps, they are frequently disrupted by ionisation or exchanges from binary-single and binary-binary interactions. We report a broadened radial distribution of formations and mergers about the trap radius when accounting for binary-single and binary-binary interactions, which we describe as a "scattering belt”. We assess the influence of the pre-existing BH population. When we model a non-zero eccentricity distribution in the BH population, the binary formations and mergers are suppressed by an order of magnitude compared to a non-eccentric population as fewer BHs embed in the AGN disk. We report an approximate merger rate of $\mathcal{R}_\mathrm{GW}\approx2\text{--}12\,\mathrm{Gpc}^{-3}\,\mathrm{yr}^{-1}$ with maximum BH masses of $10^2\text{--}10^3\,M_\odot$ through hierarchical mergers, where lower initial velocity dispersions produce larger BHs. The mass and mass ratio distributions are comparatively flat, suggesting AGN may be more relevant for high-mass or unequal-mass gravitational wave sources.
Read the full abstract ▾
Robotics
Duty Factor Predicts Robust Constrained Quadrupedal Locomotion Across Gait Types
James Zhu, David Ologan, George Ortiz et al. (+5 more)
Quadrupedal
robots are increasingly deployed in environments where locomotion must remain robust to disturbances and constrained terrain. Gait type, such as walking or trotting, is commonly used to characterize quadrupedal locomotion. However, gait type does not uniquely define locomotion, as parameters such as duty factor, speed, and stance width can vary within a single gait type. In this work, we investigate the relationship between these gait parameters using three distinct quadrupedal locomotion control approaches. First, using whole body trajectory optimization with LQR feedback, we show that duty factor is a stronger predictor of local error convergence than nominal gait type. Second, we investigate duty factor selection with a learned locomotion controller, suggesting how duty factor may serve as a low-dimensional parameter for adapting locomotion robustness in narrow-terrain environments. Finally, we show that these trends persist under a centroidal model predictive control framework and validate them through narrow-terrain experiments on a physical quadruped. These results show that duty factor provides a simple and effective basis for understanding and selecting robust quadrupedal locomotion across gait types and control architectures.
Read the full abstract ▾
All authors James Zhu, David Ologan, George Ortiz, Thomas Chun Fai Lee, Selvin Garcia Gonzalez, Ardalan Tajbakhsh, Pinhas Ben-Tzvi, Aaron M. Johnson
Solar and Stellar Astrophysics
Multi-Epoch Stability in the Rotational Modulation of the Planetary-Mass Companion Ross 458C
Elena Manjavacas, Jared Bull, Paulo Miles-Paez et al. (+7 more)
Spectroscopic variability is detected in brown
dwarfs across all spectral types, suggesting atmospheric patchiness in brown
dwarf atmospheres. We present JWST/NIRSpec time series observations of the T8.0 planetary-mass companion Ross 458C. We compared our derived spectral variability and light curve to those obtained using HST/WFC3 observations more than 7.5 yr before and published by Manjavacas et al. (2019). We found that the light curve of Ross 458C is remarkably similar in the two epochs in terms of shape and variability amplitude potentially created by stable weather patterns. We measured a rotational period of 11.57+\-0.09 hr, demonstrating that the previously reported HST/WFC3 period of 6.75+\-1.58 hr is most likely a second-order harmonic produced by an unresolved double-peaked rotational modulation, likely due to the higher uncertainties of the HST/WFC3 light curve and the short time baseline of those observations. We measured the variability amplitude in the white light curve created using only the wavelength range covered by HST/WFC3 (1.10-1.63 micron), obtaining a value of 1.47+\-0.18%, which is consistent with that measured in the HST/WFC3 light curve.
Read the full abstract ▾
All authors Elena Manjavacas, Jared Bull, Paulo Miles-Paez, Natalia Oliveros-Gomez, Theodora Karalidi, Ben W. P. Lew, Zhoujian Zhang, Yifan Zhou, Daniel Apai, Xianyu Tan
Astrophysics of Galaxies
Discovery, Characterization, and Potential Origins of a Stream in the Stellar Halo of Nearby LMC-Mass Galaxy NGC 55
Benjamin N. Velguth, Burçin Mutlu-Pakdil, Erik Tollerud et al. (+32 more)
We present a previously undetected
stellar stream in the
halo of the LMC-mass
dwarf galaxy NGC 55 (2 Mpc), as part of an ongoing effort to characterize the
stellar halos of LMC/SMC-mass
dwarfs in the DEEP component of the DECam Local Volume Exploration (DELVE) survey. This structure is aligned with an
outflow traced by H$\alpha$ emission, a spur and cloud of neutral hydrogen, and the ultra-diffuse and possibly disrupting satellite
galaxy NGC 55-dw1. We investigate possible ex-situ progenitor scenarios for this stream through the local luminosity-
metallicity relation and a set of toy dynamical models that include gaseous,
stellar, and dark matter components. If the stream is a product of a previous merger, the progenitor
galaxy had an absolute magnitude $M_V \leq -7.2$ and a
stellar mass $M_* \geq 10^5 M_{\odot}$ based on extrapolations of the detected
stellar populations. If it is disrupted material from the ultra-diffuse satellite, their shared progenitor had an absolute magnitude $M_V \leq -8.3$ and a
stellar mass $M_* \geq 10^{5.2} M_{\odot}$. We compare our stream and progenitor scenarios to those for NGC 300, a
galaxy with a
stellar mass similar to that of NGC~55 and known to host multiple structures in its
halo. Our results demonstrate that even relatively isolated LMC-mass
galaxies can have complex
accretion histories, providing new tests of hierarchical
galaxy formation at low masses.
Read the full abstract ▾
All authors Benjamin N. Velguth, Burçin Mutlu-Pakdil, Erik Tollerud, Sarah Pearson, Kristine Spekkens, Amandine Doliva-Dolinsky, Michael G. Jones, Denija Crnojević, Jeffrey L. Carlin, David J. Sand, Paul Bennet, Guy Stringfellow, William Cerny, Alex Drlica-Wagner, Joanna D. Sakowska, Jaclyn Jensen, Laura C. Hunter, Eric F. Bell, David Martínez-Delgado, Anirudh Chiti, Adam Smercina, Nitya Kallivayalil, Deepthi S. Prabhu, Alexander H. Riley, Yumi Choi, Noelia E. D. Noël, Sasha N. Campana, Guinevere Herron, Alistair R. Walker, Andrew B. Pace, Clara E. Martínez-Vázquez, Clecio R. Bom, Julio A. Carballo-Bello, Pol Massana, Gustavo E. Medina
Human-Computer Interaction
A Sociotechnical Review of Algorithms in Health Systems: Technical, Cost, and Human-Centered Considerations
Victoria Chui, Kelly McConvey and Shion Guha
Artificial intelligence (AI) applications in healthcare are becoming increasingly prevalent, to assist health systems, providers, and patients with tasks such as decision-making, risk prediction, and diagnosis. This increasing computational potential brings AI applications to the forefront of workplace decision making, often without full consideration of subsequent computational, organizational, and social costs. These applications are leveraged to reduce healthcare costs and increase efficiency of daily tasks, with model-related costs being considered at varying levels of granularity. To understand these trends, we critically analyze 114 papers to examine how cost-aware AI models have been developed for health systems. We explore the data, method, and outcome choices of these models, as well as their intersection with cost and human-centered concerns, highlighting the gaps in rigorous sociotechnical model design. From these trends, we define model costs and subsequent dimensions, presenting insight into those studies reporting financial, computational, organizational and/or social measures. Further, we critique the benefits and challenges of evaluating model-related costs and sustainability concerns when developing AI models for health systems.
Read the full abstract ▾
Computer Vision
OmniVBench: A Benchmark and Large-Scale Dataset for Omni Reference-to-Video Generation
Wenxue Li, Peiyan Guan, Haoyang Jiang et al. (+14 more)
Omni reference-to-
video (R2V) generation is emerging as a general paradigm that enables flexible
video customization from diverse and compositional visual references. However, existing
benchmarks fall short of these emerging capabilities: their test cases cover only a limited range of reference types and compositions, and their evaluation protocols largely assess holistic reference consistency, overlooking whether reference factors are properly preserved, disentangled, and routed. Meanwhile, the high cost of constructing omni R2V training data makes suitable training resources scarce. To address these gaps, we introduce OmniVBench and the Omni-R2V
Dataset, establishing a shared foundation for evaluating and training omni R2V models. OmniVBench expands R2V evaluation across broader reference types, richer reference compositions, and more flexible instruction-driven customization, covering 7 task families and 18 fine-grained tasks. Complementing this task design, we introduce factor-grounded evaluation with 12,172 case-specific checklist items, explicitly assessing whether intended reference factors are faithfully preserved, correctly disentangled and bound to their targets, and realized according to the instruction. We further introduce the Omni-R2V
Dataset, bringing industrial-grade training resources for diverse R2V tasks to the broader research community. Drawing primarily on a large-scale corpus of professional
video footage, it comprises 340K processed training samples spanning diverse reference types and multi-reference compositions. Specifically, we develop task-specific pipelines for reference-target pair construction, offering a practical and scalable recipe for omni R2V data construction. Extensive evaluation of advanced open- and closed-source R2V models reveals substantial performance variation across diverse tasks and distinct capability patterns across evaluation dimensions.
Read the full abstract ▾
All authors Wenxue Li, Peiyan Guan, Haoyang Jiang, Junxian Cai, Hualuo Liu, Chunjie Zhang, Chong Guan, Kai Huang, Songlian Li, Taiyi Wu, Yongjian Yu, Xiaotong Zhao, Alan Zhao, Quande Liu, Xi Chen, Yu Liu, Lei Zhu
Artificial Intelligence
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
Bowen Ye, Lei Li, Shicheng Li et al. (+16 more)
Training capable coding
agents via reinforcement learning (RL) requires diverse tasks with reliable verifiers. Open-source codebases offer a rich source of such tasks, while existing methods typically rely on development artifacts such as issues and commits, limiting the range of tasks that can be extracted. To better scale RL environments, we present CodeMidas, an agentic pipeline that turns implemented functionality in existing codebases into executable RL environments using source code as its only task-specific input. CodeMidas allocates agentic compute to every stage of environment construction:
agents explore implemented functionality to formulate behavioral specifications, construct tests grounded in execution of the original code, and validate and filter candidate tasks through execution checks and repeated solution rollouts. The resulting
dataset has 5,545 training tasks from 3,185 open-source codebases spanning 23 programming
languages and 15 technical domains. Training
MiMo-V2.5 on these tasks with GRPO improves performance on all five diverse
benchmarks, covering issue repair (DeepSWE + 11.7%), whole-program construction (ProgramBench +17%), and terminal work (Terminal-Bench v2.1 +8.5%). Ablations show that increasing the number of high-quality training tasks improves performance. Trajectory analysis shows the RL-trained
agent demonstrates better behaviors like increasing codebase exploration and more diverse self-verification. These results establish source code as a scalable foundation for constructing RL environments that improve coding
agents across diverse software tasks.
Read the full abstract ▾
All authors Bowen Ye, Lei Li, Shicheng Li, Zihao Yue, Linghao Zhang, Hanglong Lv, Yuanxin Liu, Wenhan Ma, Hao Tian, Rang Li, Jinhao Dong, Yikai Zhao, Xiangwei Deng, Hailin Zhang, Liang Zhao, Qi Liu, Lingpeng Kong, Tong Yang, Fuli Luo
Human-Computer Interaction
Value-Sensitive Delegation in Everyday AI Agent Use: Evidence from OpenClaw
Renkai Ma, Ruyuan Wan, Xuan Lu et al. (+3 more)
Users increasingly delegate work to autonomous AI
agents, yet evaluations typically measure task completion rather than the values users prioritize. Using Value Sensitive Design, we analyzed, with LLM assistance, 73,093 first-person Reddit posts about using OpenClaw, each for its human value,
agent aspect, value fulfillment, and user outcome. The 21 values form six value groups, including Autonomous, Dependable, and Affordable Operation, Bounded Reach, Reviewability, and Equitable Access. Relative to each aspect's corpus share, values clustered not at the
agent's outputs but at the operating conditions users set around a run. Values were usually met where users described what the
agent delivered, in five of six groups, and mostly unmet where users described supervising it, in all six groups. We conceptualize this pattern as value-sensitive delegation. Supporting human values requires
attention not only to what an
agent accomplishes, but to the conditions users set around delegation, including cost, access, and oversight.
Read the full abstract ▾
All authors Renkai Ma, Ruyuan Wan, Xuan Lu, Fan Yang, Chen Chen, Lingyao Li
Mathematical Physics
Generalized Hamiltonian formalism for spatially nonlocal nonlinear differential equations
Ali Pazarci, Nadir Ghazanfari and Ilmar Gahramanov
In this work, we develop a generalized
Hamiltonian formalism for spatially nonlocal field theories whose Lagrangian densities depend explicitly on both the local field and its spatially reflected counterpart. Starting from a generalized variational principle, we derive generalized Euler-Lagrange equations and introduce a generalized functional derivative that consistently accounts for reflected-field contributions. The proposed formalism is applied to three spatially nonlocal nonlinear Schrödinger equations. For the Ablowitz-Musslimani equation, we construct, to the best of our knowledge, the first standard Lagrangian density formulated directly in terms of the complex fields and derive its complete
Hamiltonian formulation. The same framework is subsequently applied to the two Lagrangian nonlocal nonlinear Schrödinger equations introduced by Velasco-Juan and Fujioka, yielding consistent
Hamiltonian formulations that reproduce the corresponding generalized Euler-Lagrange equations. These results establish a unified
Hamiltonian framework for a broad class of spatially nonlocal nonlinear field theories.
Read the full abstract ▾
Information Theory
Hermite-Fisher bounds and stability for min-entropy power inequalities
Silouanos Brazitikos, Martin Rapaport and Tomasz Tkocz
We derive explicit lower bounds for relative Fisher information by combining a variational principle with suitably orthogonalized Hermite-
polynomial test functions. The resulting cumulant bounds are asymptotically sharp and yield lower bounds for Gaussian
entropy deficits. We also establish quantitative versions of sharp min-
entropy power inequalities in all dimensions. En route, we develop a stability result for Brzezinski's sharp bound for block sections of products of Euclidean balls, which may be of independent interest.
Read the full abstract ▾
Machine Learning
BrainWideBench: Benchmarking large-scale pretraining and across-animal transfer in multi-region neural recordings
Alexandre Andre, Shivashriganesh P. Mahato, Vinam Arora et al. (+39 more)
Advances in large-scale neural recording have made it possible to collect data across many animals and distributed
brain regions, raising the question of whether this scale can be exploited to learn general-purpose neural representations transferable across diverse downstream tasks. Yet, progress toward this goal has been limited by fragmented evaluation protocols and a narrow focus on individual task domains. Here, we present BrainWideBench, a
benchmark for evaluating across-animal transfer on multi-region neural recordings, built on the International
Brain Laboratory Brainwide Map
dataset of neural and behavioral recordings spanning 276
brain regions from 139 mice performing a sensory-guided decision-making task. The
benchmark is organized around three complementary task suites that evaluate whether learned representations support downstream decoding of behavior, can predict masked or future neural activity, and can recover biologically meaningful anatomical organization. With this
benchmark, we systematically evaluate pretraining methods across transfer settings, including finetuning on downstream objectives and zero-shot generalization to unseen animals. Our results confirm pretraining improves performance over matched single-session baselines, but we show current methods exhibit heterogeneity in transfer capabilities: gains depend strongly on the
alignment between pretraining objectives and downstream tasks. No single approach performs uniformly well across all three suites, and most methods are designed to only address a subset of them. Together, these findings suggest that learning representations that jointly generalize across behavior, dynamics, and anatomy remains an open challenge. By providing a unified and reproducible evaluation suite, BrainWideBench establishes a framework for measuring progress toward general-purpose models of the mouse
brain.
Read the full abstract ▾
All authors Alexandre Andre, Shivashriganesh P. Mahato, Vinam Arora, Keshav Balaji, Divyansha Lachi, Nanda H. Krishna, Jingyun Xiao, Yizi Zhang, Ximeng Mao, Wenrui Ma, Han Yu, International Brain Laboratory, Daniel Birman, Niccolò Bonacchi, Gaelle A. Chapuis, Joana A. Catarino, Felicia Davatolhagh, Mayo Faulkner, Laura Freitas-Silva, Fei Hu, Julia M. Huntenburg, Anup Khanal, Inês Laranjeira, Petrina Lau, Guido T. Meijer, Nathaniel J. Miska, Jean-Paul Noel, Alejandro Pan-Vazquez, Georg Raiser, Cyrille Rossant, Karolina Z. Socha, Anne E. Urai, Miles J. Wells, Steven J. West, Olivier Winter, Blake Richards, Guillaume Lajoie, Cole Hurwitz, Mehdi Azabou, Matthew R. Whiteway, Liam Paninski, Eva L. Dyer