Peer Revenue Play credits. Real papers.

Papers

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

All of arXiv

Tuesday 8 September

Robotics

Ostrich: Taking Large Strides Through Stiff Contact in Differentiable Dynamics

Three properties determine whether a differentiable simulator can drive gradient-based optimization through contact: simulation accuracy, gradient reliability, and per-iteration cost. Tape-based engines such as MJX and Newton Semi-Implicit require timesteps small enough to keep contacts numerically tractable, and their backpropagation memory grows linearly with the number of timesteps T. Surrogate models bound memory by approximating contact away, but the resulting gradients lose the geometry the optimization depends on. We present Ostrich, a GPU-accelerated rigid-body simulator that resolves hard contacts and friction with non-smooth Newton iteration at large timesteps (h ~ 0.1 s), and differentiates the converged residual via the implicit function theorem, reusing the forward Schur complement to compute the adjoint at O(1) memory per timestep. On real-robot trajectories over a pallet obstacle, Ostrich holds MuJoCo's sim-to-real accuracy up to a 50x larger timestep. Its gradients converge from random initializations where MJX descends slowly and Newton Semi-Implicit stalls; a warm iteration runs 211x faster than MJX's and 4.7x faster than Semi-Implicit's. On the same scene Ostrich differentiates 8,192 parallel worlds on a single 24 GB GPU, sustaining 29x checkpointed MJX's optimization throughput; without checkpointing both baselines exhaust memory at far fewer worlds. We close with a gradient-based trajectory optimization demonstration over triangle-mesh terrain across a 10 s horizon, a setting where prior engines either restrict to primitive geometry or face the convergence and memory limits shown above.
Read the full abstract

Combinatorics

A study of $m$-ary partitions whose conjugates are $q$-ary

While people have studied $m$-ary partitions of an integer $n$ and studied conjugation of partitions of $n$, these topics are rarely mixed because the $m$-ary property is almost always lost after conjugation. In a previous work, Flowers and Lockard investigated $m$-ary partitions of $n$ whose conjugates were also $m$-ary. We generalize that previous work by studying $m$-ary partitions whose conjugates are $q$-ary, where $m$ and $q$ may be distinct. We provide a family of operators on these partitions that can be used to generate all such partitions uniquely and associate a unique polynomial with each partition based on the sequence of operators used to generate it. Using the generating operators and modular arithmetic we explore many examples and families of $m$-ary partitions whose conjugates are $q$-ary.
Read the full abstract

Machine Learning

Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation

Weak-to-strong generalization asks whether stronger models can learn from weaker supervisors and surpass them. This is particularly important for successive model generations and multi-domain consolidation, where frontier-scale post-training from scratch can be prohibitively expensive. Yet conventional distillation treats the weak teacher as an optimization target, potentially imposing its capacity ceiling on the student. We introduce On-Policy Reverse Distillation (OPRD), which evaluates the teacher's policy shift relative to its reference policy on student rollouts and amplifies the component of the student's verifier-driven policy gradient along that direction. By rescaling only verifier-supported updates, OPRD preserves the stationary points of policy optimization while accelerating learning beyond the teacher. Our evaluations cover mathematical reasoning, logical reasoning, code generation, and visual tool use. In both successive model transfer and multi-teacher distillation, OPRD achieves higher performance with fewer student updates than existing RL and distillation approaches. Response-style analysis shows that OPRD students remain closer to verifier-only RL models than to their weak teachers, suggesting that guidance accelerates rather than redirects student optimization. Strong-to-weak results further show that OPRD combines verifier-driven optimization with teacher guidance regardless of capacity ordering.
Read the full abstract

All authors Youngrok Park, Sangmin Bae, Hojung Jung, Jongwoo Ko, Yunseon Choi, Young Jin Kim, Pashmina Cameron, Aaron Courville, Se-Young Yun

Statistics Applications

Bridging Network Psychometrics and Artificial Intelligence: An Ising-Potts Model with LLM-Derived Weights

The Potts model extends the Ising model to multinomial data. We introduce a Rater Ising-Potts model that uses agreement indicators between pairs of ratings and category labels, with weights derived from LLM embeddings. The model does not presuppose ordered category thresholds or equidistant scoring; instead, it focuses on pairwise agreement among ratings and assigns category-specific positive weights, making it suited for multi-category scoring reliability. We evaluate the model on three constructed-response datasets spanning a corpus of K=14,466 short answers on a three-level rubric and two AERA essay prompts of roughly 1,200-1,400 responses on four-point rubrics. We compare three strategies for sharpening the similarity signal: top-K pruning, min-max normalization with a power transformation, and ColBERT late-interaction similarities. Top-K pruning, which replaces the dense similarity graph with a sparse local network of strongest semantic neighbors, consistently yields the highest accuracy and Cohen's kappa, and the selected neighborhoods are always a small fraction of the corpus. Power tuning consistently ranks second, while ColBERT is competitive on longer essay prompts and adds little on short answers. Across all settings, most misclassifications occur between adjacent score levels, confirming that the model preserves the ordinal structure of scoring rubrics without imposing rigid assumptions. These findings suggest that LLM-derived similarities, combined with a parsimonious Potts formulation and a sparse local graph, offer a robust and interpretable framework for reliability auditing in educational assessment. We discuss extensions to multiple raters and hierarchical rating designs.
Read the full abstract

Computer Vision

Hi-FLoop: Hierarchical State-Feedback Loops for Multi-Timescale World Modeling

Multi-agent traffic simulation seeks diverse, coordinated, and physically realistic futures from maps and observed history. Long-horizon closed-loop generation must reconcile multiple decision time scales while its context evolves with generated states. Existing methods often unfold long futures from an initial scene and resolve intent, interaction, and motion monolithically, weakening cross-scale consistency and adaptation. Multimodal rollout poses a further consistency problem: independently reselecting modes across agents or commits can stitch together incompatible futures instead of preserving a coherent joint branch. We present Hi-FLoop, a branch-consistent multi-timescale state-feedback framework. Eight scene-level Worlds represent joint hypotheses; all agents share one selected World identity throughout all 16 commits of an 8-second rollout, while Goal, Preview, and Control states adapt within that branch. An 8-second Goal anchors intent, a 2-second Preview coordinates interactions, and 1-second Control produces physical motion. Every 0.5-second commit feeds back only its executed prefix as new facts, while unexecuted hypotheses never enter factual memory. Joint Preview Interaction induces a sparse directed future graph and uses conflict probabilities and signed arrival-time differences to refine interaction-aware motion. For generated-state recovery, a prefix-frozen A-to-B cascade transfers typed physical state and the branch index–but no latent state–from a frozen prefix model to an independently parameterized recovery model. On the full H-D public-validation split of 955 scenarios, the S2.1 cascade obtains an 8-second scene-joint ADE-at-joint-minFDE@8/joint-minFDE@8 of 2.048/6.384 m when one World must explain all evaluated agents. Agent-centric oracle-minADE@8 is 0.526 m at 6 seconds and 0.875 m at 8 seconds.
Read the full abstract

Audio and Speech Processing

Interpreting Dolphin Vocal Sequences via Multiple Sequence Alignment

Dolphin communication understanding is essential for uncovering the linguistic complexity and social structures of wild pods. We adapt the ClustalW bioinformatics algorithm to analyze continuous acoustic data, treating vocalizations as high-dimensional spectral feature vectors. By replacing discrete scoring with a continuous Gaussian kernel similarity measure, our framework generates Multiple Sequence Alignment (MSA) visualizations that reveal shared structural patterns. These alignments highlight temporal motifs such as synchronized burst pulses in aggressive contexts that are difficult to detect through standard spectrogram inspection.
Read the full abstract

Probability

MA(1) processes with Laplace innovations conditioned to stay positive

We study a moving-average process with (not necessarily symmetric) Laplace innovations under the constraint of positivity. In the three nondegenerate parameter regimes $0<\theta<1$, $\theta>1$, and $\theta<0$, we prove convergence of the conditioned finite-dimensional distributions and identify the limit as a Doob $h$-transform. The regimes lead to qualitatively different limiting dynamics: the invariant law is supported on the positive half-line for $0<\theta<1$, the conditioned chain is confined to the negative half-line for $\theta>1$, and the dynamics are genuinely two-sided and governed by $q$-trigonometric functions for $\theta<0$. In each case, the persistence exponent, sharp persistence asymptotics, the defining eigenfunction, and the unique invariant distribution are obtained explicitly.
Read the full abstract

Combinatorics

The Vámos Matroid Has No Second Symmetric Power

Symmetric powers of matroids were introduced by Lovász and Mason. The subject has recently attracted renewed interest through its connection with the realizability of tropical linear spaces by tropical ideals. Anderson proved that the tropical linear space associated with a matroid $M$ is the variety of a tropical ideal if and only if $M$ has a $d$th symmetric power for every positive integer $d$. He also asked whether the Vámos matroid $V_8$ has a second symmetric power. We answer this question in the negative by proving that $V_8$ has no second symmetric power. By Anderson's characterization, the associated tropical linear space cannot be realized as the variety of a tropical ideal.
Read the full abstract

Physics Education

From data to structure: Construction, breakdown, and reconstruction of an empirical representation in a black-box RC circuit

In introductory physics laboratories, a central instructional goal is to help students construct, evaluate, and revise mathematical representations from experimental data rather than apply given formulas. We present a guided data-driven activity in which an RC circuit is treated as a black box defined through observable input-output relations. Experimental and graphical procedures are provided, while standard RC circuit theory is initially withheld. Students first examine discharging data obtained under different nominal values of R and C. They rescale time using T=t/RC and compare the normalized voltage v=V/E across circuit conditions, bringing the discharging data closer to a common trajectory. Graphical linearization through a plot of lnv versus T provides a basis for proposing an empirical representation and allows RC to be interpreted as a time scale rather than being assigned that meaning a priori. When the representation proposed for discharging is tested with charging data, logarithmic linearity is lost despite the continued organization of the data by T. Replacing v with the distance from the steady state, 1-v, restores linearity, and the resulting linear relation provides a basis for proposing a revised representation for charging. The activity thus uses a familiar physical system to make representation construction, breakdown, and reconstruction experimentally explicit. It also emphasizes that an empirical representation must be accompanied by a specified domain of validity and that standard circuit theory can be introduced a posteriori to interpret the physical meanings that emerge from representational constraints.
Read the full abstract

2609.08792 · 8 Sep · 38 pages, 3 figures. Substantially revised and retitled version of an earlier preprint posted on Jxiv (https://doi.org/10.51094/jxiv.4885). Expanded instructional framing and Supplementary Materials on implementation, instructor guidance, student outcomes, and informal classroom observations

Analysis of PDEs

MIT bag model and infinite mass limit in non-smooth domains

The work is devoted to the study of Dirac operators with MIT bag boundary conditions in Euclidean domains with compact Lipschitz boundaries in arbitrary dimensions. It is shown that such operators are self-adjoint on suitable definition domains and can be recovered as the norm-resolvent limits of Dirac operators in the whole space with a large mass term outside the domain, under the assumption that an associated Robin-Laplacian eigenvalue has a prescribed asymptotic behavior with respect to a parameter in the boundary condition. This assumption is shown to hold for a class of non-smooth domains, which includes convex domains and, more generally, domains that can be "locally convexified" by suitable diffeomorphisms. To the best of our knowledge, this represents the first infinite mass interpretation for the MIT bag model in the sense of resolvent convergence for non-smooth domains. Most results are extended to the generalized MIT bag boundary conditions with the help of the recently established congruence transform.
Read the full abstract

All authors Badreddine Benhellal, Noah Körner, Dylan Machado, Konstantin Pankrashkin

Cryptography and Security

Evidence-Grounded Retrieval for Investigation Hunt Lead Generation from CTI Reports

Threat hunting increasingly depends on converting unstructured knowledge (e.g., Cyber Threat Intelligence reports) into actionable hunt leads: concise, investigable hypotheses grounded in observable artifacts and adversary techniques. Producing such leads manually is a tedious and hard-to-scale task. Existing automated approaches stop at the entity layer, ignore the defender's operational environment, and analyze each report in isolation. To address these gaps, we introduce AHLERT, a system that automatically extracts relevant, environment-aware, and hunt leads from threat reports through (i) a hybrid retriever that combines dense vector search with multi-hop traversal over a knowledge graph seeded with MITRE ATT&CK; (ii) an ontology-grounding retrieval-augmented generation method that constrains each lead to the defender's own assets and controls; and (iii) an LLM-agnostic framework that emits structured, directly actionable leads rather than loose indicators of compromise. We evaluate AHLERT on public CTI reports for well-known APTs across multiple proprietary and open-weight models. Hybrid evidence retrieval with ontology grounding raises mean F1 by ~2x (0.44 to 0.85) over a single-route flat-RAG baseline, and AHLERT attains the highest effectiveness score (~86.95%) compared with off-the-shelf LLM models.
Read the full abstract

All authors Akash Prakash, Boubakr Nour, Makan Pourzandi, Chadi Assi, Mourad Debbabi

Computers and Society

Silent Revision: Measuring Undisclosed Change in the Safety Frameworks of Frontier AI Developers

Frontier AI developers publish safety frameworks that commit them to evidencing whether their models are dangerous. The European Union and California now treat these documents as instruments of accountability and impose duties on their revision. Neither requires the developer's account to identify each change to a commitment. We introduce the silent revision rate, the share of material changes to a framework's commitments that the developer's published account does not identify, and we release the versioned, hash-pinned corpus needed to compute it. The corpus holds every public version of the safety frameworks of the twelve developers that have published one, with each provider's changelog, redline or announcement. We trace 710 commitment instances across twelve consecutive version pairs, code them against a frozen codebook, and check the coding with three independent re-runs, a blind re-coding of every disputed row and a second coding of a stratified sample. First, 64% of material changes (95% CI 59 to 69) are silent under a strict standard and 30% under a lenient one. Second, the form of the account matters less than its granularity. Narrative and itemised accounts differ by eight points that eight pairs cannot separate from chance, whereas the rate falls with the number of discrete items a provider publishes. Third, 75% of traced changes weaken or remove a commitment, and weakenings are silent more often than strengthenings, at 70% against 52%. A justification explains why a framework changed and an enumeration states what changed, and only the latter makes revision auditable. We argue that publication duties should carry an enumeration duty, which one provider already meets, voluntarily and incompletely.
Read the full abstract

2609.08789 · 8 Sep · 31 pages, 3 figures, 10 tables. v2: coding checked by three re-runs, a blind re-coding and a second coding; strict rate revised from 67% to 64% and lenient from 53% to 30%; the v1 narrative-against-itemised difference is no longer distinguishable from chance. Data and code: https://github.com/louisyzhu/frontier-safety-framework-corpus (DOI 10.5281/zenodo.22670700)

Machine Learning

Adaptive Anisotropic Attention for Axis-Structured Signals

Dense self-attention treats all token pairs as equally plausible before learning, an interaction-isotropic prior that can be mismatched to structured signals. For structured, low signal-to-noise ratio (SNR) signals such as EEG, dependencies are organized along the electrode and time axes, and this uniform prior exposes each token to many irrelevant interactions. We introduce Adaptive Anisotropic Attention (AAA), which splits attention into two paths: a temporal path, where each token attends to the tokens of its own electrode across time, and a spatial path, where it attends to the tokens of the other electrodes at the same time step. A small gate predicts, for every token, a convex combination of the two path outputs: two non-negative weights that sum to one. On six EEG downstream tasks, the resulting model, AXON (AXis-factorized Operator Network), improves mean balanced accuracy over a dense baseline under both linear probing and full fine-tuning. We show that both paths (temporal and spatial) are necessary and that the weighted sum beats a hard choice of one path; most of the benefit comes from the gate learning a different temporal/spatial balance at each layer of the network. Controlled audio spectrogram experiments show that axis factorization transfers beyond EEG. These results suggest that aligning attention with the natural axes of structured signals provides a useful inductive bias.
Read the full abstract

All authors Mahir Jain, Parshva Runwal, Aditya Ray Mishra, Arvasu Kulkarni, Jeet Bandhu Lahiri, Sandeep Singh, Siddharth Panwar

High Energy Astrophysical Phenomena

Compton-induced HE – VHE $\gamma$-rays from pair cascade emissions in the SEDs of radio galaxies: NGC 1275

In the scheme of active galactic nuclei (AGNi), the blazar subclass is particularly bright in $\gamma$-rays. This is because of their emissions being beamed into our line of sight, as opposed to their misaligned parent population, radio galaxies. This work presents results of a Monte-Carlo code that propagates jet-collimated $\gamma$-rays through the magnetized AGN environment, leading to secondary cascade emissions from relativistic electron-positron pairs. The resulting multi-wavelength emissions are used to fit the broadband spectral energy distribution (SED) of the radio galaxy NGC 1275. Using known physical parameters of NGC 1275 from the literature, we find that cascade emissions from stochastic processes have the potential to reproduce the broadband SEDs of radio galaxies. We also find that the interplay between accretion disk and broad-line region (BLR) photons may provide insights into the production region(s) of $\gamma$-rays in AGNi. While dense accretion disk and BLR radiation fields are vital to the development of cascades, excessive densities lead to the suppression of cascade synchrotron emission and attenuation of \,TeV $\gamma$-rays. Furthermore, the magnetic field is important in shaping the cascade $\gamma$-ray spectrum.
Read the full abstract

Computer Vision

Interpretable Hyperspectral Unmixing Framework with Fixed Endmember Prior and Structured Residual Refinement

Hyperspectral unmixing decomposes mixed pixels into material endmembers and their abundances from contiguous spectral observations. In modular sensing pipelines, endmembers are often first identified and then treated as fixed during abundance estimation. When this fixed endmember prior is inaccurate, spatially structured mismatch arising from illumination changes, sensor artifacts, or material boundaries may be incorrectly captured by the abundance variables, leading to unstable decompositions. This study presents an interpretable stage-wise hyperspectral unmixing framework (I-HyperSU) under fixed endmember priors, which is explicitly decomposed into a fixed endmember matrix $\mathbf{A}$, an abundance block $\mathbf{X}$, and a structural residual refinement block $\mathbf{S}$. The X-block estimates abundances using FISTA with nonnegativity and sparsity enhancement, and a soft penalty that approximately enforces sum-to-one constraints. The S-block jointly applies low-rank SVD structural regularization and a lightweight deep image prior (DIP) to refine structured residuals. This staged design makes the interaction between abundance and residual components transparent and interpretable. Experiments on Samson, Urban, and Jasper Ridge datasets demonstrate that, under fixed and imperfect endmember priors, soft abundance relaxation consistently outperforms hard simplex projection. Under the default N-FINDR endmember prior, the proposed framework reduces the joint reconstruction error by 61.7%–69.5% compared with a fixed-$\mathbf{A}$ UCLS baseline, while keeping the abundance RMSE nearly unchanged, indicating that the residual refinement branch accounts for structured model mismatch without degrading the abundance estimates. For example, on Urban, the reconstruction SAM decreases from $5.99^\circ$ for the X-only model to $1.92^\circ$ for the full model.
Read the full abstract

High Energy Physics (Phenomenology)

Chiral matching of dimension-7 baryon-number-violating operators and application to $\Delta I=3/2$ nucleon decays

We study chiral matching of dimension-7 (dim-7) baryon-number-violating operators in low-energy effective field theory. We show that all dim-7 operators can be classified into two distinct Lorentz structures and their chirality-flipped counterparts, and we derive their leading-order hadronic realizations in chiral perturbation theory. As an application, we systematically investigate $\Delta I=3/2$ nucleon decays, including $n\to \ell^-\pi^+$, $p\to \ell^- \pi^+\pi^+$, and $n\to \ell^- \pi^+ (\pi^0, \eta)$, with $\ell=e,~\mu$. Using existing experimental bounds on $n\to \ell^-\pi^+$, we set stringent constraints on the relevant Wilson coefficients and derive improved lower limits on the partial lifetimes of the six three-body modes from their correlations with the two-body modes. We further present a concrete ultraviolet model that generates the dim-7 operators and the associated $\Delta I=3/2$ nucleon decays at leading order. Our results pave the way for exploring dim-7 contributions to nucleon decays and motivate future searches for these exotic modes.
Read the full abstract

Information Theory

The Weight Spectrum of the Affine Grassmann Code $C^{\mathbb A}(3,6)$

In this article, we consider the affine Grassmann code $C^{\mathbb A}(3,6)$, obtained from the affine open cell ${\mathbb A}^9$ of the Grassmannian $G_{3,6}$. We exploit the representation of codewords as linear combinations of minors of all sizes of a generic $3\times3$ matrix and classify them according to the largest size of a minor occurring with a nonzero coefficient. Using this classification, we determine all possible Hamming weights of codewords of $C^{\mathbb A}(3,6)$ and, for each weight, compute the number of codewords attaining that weight. Consequently, we obtain the complete weight spectrum of the affine Grassmann code $C^{\mathbb A}(3,6)$.
Read the full abstract

Statistics Theory

Improved Variance Estimation in Homoskedastic Nonparametric Random-Design Regression via a Two-Scale Approach

We study estimation of a constant conditional variance $\sigma^2$ in nonparametric regression with a $d$-dimensional random design. This is an important problem, and similar questions arise in causal inference. The regression function is $\beta_b$-Hölder smooth, the design density is $\beta_g$-Hölder smooth and bounded above and away from zero, and we consider the nonparametric regime $\beta_b>1$ and $d>4\beta_b$. Set $\beta_g^\star=\beta_b(1-4\beta_b/d)/\{1+2\beta_b/d+8(\beta_b/d)^2\}$. We give an estimator whose mean squared error is upper bounded by $Cn^{-4(\beta_b+1)/(d+4)}$ in the low-regularity regime when $0<\beta_g\leq\beta_g^\star$. The low-regularity branch is based on a new two-scale construction: the covariate space is partitioned into cells, the local polynomial trend is projected out within each suitable cell, and the squared normalized contrast from one eligible close pair per cell is averaged across cells. In the high regularity regime when $\beta_g>\beta_g^\star$, a higher-order influence function estimator of Robins, Li, Tchetgen Tchetgen, and van der Vaart (2008) provides the rate $Cn^{-8\beta_b/(d+4\beta_b)}$. We also give an all-pairs ridge extension, which achieves the same two-scale rate, and evaluate the methods alongside a range of existing estimators in simulations.
Read the full abstract

All authors Edgar Dobriban, Rajarshi Mukherjee, James M. Robins, Zixiao Wang

Astrophysics of Galaxies

Design and commissioning of a windowless gas-target system for high-current beams at JUNA

Windowless gas targets avoid the beam-energy loss and straggling introduced by entrance foils and are therefore well suited for direct measurements of low-energy nuclear reactions. A windowless gas-target system designed for operation with milliampere beams has been developed for the Jinping Underground Nuclear Astrophysics facility (JUNA). The system combines three-stage differential pumping, closed-loop gas recovery and purification, a constant-temperature power-compensation calorimeter, and a position-resolved target-thickness monitor based on secondary elastic scattering. Stable operation was achieved over a target-pressure range of 1-3 mbar, with pressure fluctuations below 1% during 8 h of continuous circulation, while the accelerator-side pressure was maintained at approximately \(10^{-4}\) Pa. The closed-loop gas-circulation system maintained stable target conditions, while gas-transport calculations indicated that the axial pressure nonuniformity remained within approximately 1.6% under representative operating conditions. Calorimeter measurements were consistent with the thermal calculations, supporting the sensitivity correction used for beam-power determination. Beam commissioning with \(^{14}\mathrm{N}(p,\gamma)^{15}\mathrm{O}\) and \(^{12}\mathrm{C}(p,\gamma)^{13}\mathrm{N}\) at the 600 kV Cockcroft-Walton accelerator of the China Institute of Atomic Energy (CIAE) demonstrated stable operation of the gas-target and \(\gamma\)-ray detection systems and provided information on the influences of reaction position and beam heating. These results demonstrate the operating stability and diagnostic capability of the system for future high-current, low-energy nuclear-reaction measurements at JUNA.
Read the full abstract

All authors Xuyang Wang, Fuqiang Cao, Weike Nan, Gang Lian, Wei Nan, Hongwei Huang, Yuchen Jiang, Jingyu Dong, Minghao Zhu, Yuwen Chen, Yihui Liu, Jinlong Ma, Yueran Wang, Jie Chen, Yunju Li, Shengquan Yan, Youbao Wang, Xiao Fang, Luohuan Wang, Yangping Shen, Bing Guo, Weiping Liu, JUNA collaboration

Representation Theory

Partial $F$-invariants and cluster categorifications

The $F$-invariant in cluster algebras is a combinatorial invariant that unifies the $E$-invariant from additive categorification and the $\mathfrak{d}$-invariant from monoidal categorification. In this paper, we study its refinement, the partial $F$-invariant, and establish its mutation formula under changes of the initial seed. As an application, we prove a conjecture of Reading, which asserts that the non-compatible cluster variables can be separated by sign-coherence of $g$-vectors upon varying the initial seed. We further show that, for cluster monomials, the partial $F$-invariants coincide with both the partial $E$-invariants for reachable decorated representations of quivers with potentials and the pole orders of normalized $R$-matrices (or partial $\mathfrak{d}$-invariants) for finite-dimensional reachable simple modules over quantum affine algebras. As consequences, we obtain a combinatorial formula for the pole orders for reachable simple modules in terms of $q$-characters; we verify the conjectural explicit formula for the pole orders between Kirillov–Reshetikhin modules.
Read the full abstract

Computational Physics

Structure-Informed Bayesian Inference of Anomalous Transport and Hidden Molecular Trapping in Amorphous Media

Molecular diffusion in fluctuating amorphous and macromolecular media governs key transport processes across soft-matter physics, energy storage, and biological membranes. Extracting localized trapping states from single-particle tracking trajectories remains a fundamental challenge; because thermal structural breathing continuously reconfigures pore boundaries, conventional geometric algorithms suffer from severe systematic biases, erroneously merging distinct localized states during cyclic molecular returns. Here, we address this deadlock by shifting the paradigm from local geometric recurrence to a structure-informed Bayesian regularization. Leveraging discrete Morse theory, we extract the time-invariant topological skeleton of the fluctuating host matrix to construct robust, gas-specific physical priors that account for individual molecular dimensions. Trajectory steps are sequentially partitioned via a two-stage probabilistic refinement that dynamically adapts to the transport landscape. Benchmarked against a rigorous environment where synthetic particles explore the actual interconnected matrix graph, our approach eliminates systemic biases, restricting macroscopic trapping parameter deviations to just a few percent under optimal linear $O(N)$ computational scaling. Applied to hydrogen and methane transport within a type-I kerogen matrix, serving as a prototype for highly tortuous, flexible macromolecular networks, the method successfully decodes the hidden microscopic mechanisms of confined diffusion. To ensure immediate broad impact, the documented open-source code and data are made publicly available, offering an accessible strategy readily adaptable to a broad spectrum of tracking phenomena, from ion transport in battery polymers to protein trafficking within cellular environments.
Read the full abstract

All authors Andrey Ananev, Maria Potapova, Nikolay Kondratyuk, Timur Vostroknutov, Aleksey Khlyupin

Computation and Language

Improving Term Evaluation in Machine Translation: Variation Matters

Terminology evaluation in machine translation (MT) usually assumes a single correct target form per source term. However, human translators routinely introduce variation that current metrics penalize as inconsistency. We examine how to account for this variation in document-level MT evaluation of English-French scientific translation, combining glossary-based accuracy, translation consistency, and a new cross-term variation (CTV) diagnostic measure that tests whether variation relationships are preserved across languages. Based on analyses of two parallel corpora, translated by four MT systems, we find that (1) MT systems generate less target-side variation than human translators; (2) transfer patterns strongly depend on the variation type; (3) consistency rankings vary with the choice of metric; and (4) constraining MT with a glossary improves accuracy and consistency but degrades CTV by suppressing valid variation. We argue for variation-aware evaluation that conditions consistency penalties on whether target-side variation mirrors source-side variation.
Read the full abstract

All authors Nicolas Dahan, Ziqian Peng, François Yvon, Rachel Bawden

Astrophysics of Galaxies

Weak cool core, weak turbulence: XRISM, XMM-Newton and Chandra spectroscopy of the core of Abell 1413

We present the first XRISM spectroscopy of the core of Abell 1413 (z = 0.143), combining 104.7 ks of Resolve with archival XMM-Newton and Chandra spectra of a matched sky region. A1413 is hot, massive and relaxed but lacks a strong cool core (t_cool = 6.4 Gyr), a regime in which the low turbulence reported to date has been tested only once, in A3571. The Resolve spectrum requires a single velocity-broadened component with kT = 7.06 (+0.33, -0.43) keV and a line-of-sight velocity dispersion sigma_v = 170 (+25, -24) km/s (M_3D = 0.217). The bulk velocity relative to the brightest cluster galaxy, v_bulk = -24 +/- 26 (z_ICM) +/- 71 (z_BCG) km/s, is limited by the optical redshift rather than by Resolve, and disfavours coherent line-of-sight sloshing. The non-thermal pressure fraction, 2.5 +/- 0.7 per cent, implies a hydrostatic mass bias b < 2.5 per cent, comparable to the strong cool-core cluster A2029 (2.6 +/- 0.3 per cent) and, at matched sampling scale, if anything below it. With A3571 (1.3 +/- 0.3 per cent), to which A1413 rescales within 0.2 sigma, two weak cool cores now show that this quiescence does not require central cooling. The resolved Fe-K He-alpha complex gives Fe = 0.38 +/- 0.03 (stat) +/- 0.04 (sys) solar; Ni/Fe, separated from Fe K-beta for the first time, is solar, as is the pattern as a whole. The Ne, Mg and Si lines that set Ne/Fe, Mg/Fe and Si/Fe lie below 2 keV, where our instruments disagree by 10-20 per cent; we therefore do not regard the SNIa fraction as constrained.
Read the full abstract

Computer Vision

AXS-Net: Interpretable Deep Unfolding for Hyperspectral Image Denoising via Spectral Basis Unmixing and Structured Noise Refinement

Hyperspectral images (HSIs) are often degraded by mixed noise, including band-dependent Gaussian perturbations and structured artifacts such as stripes, dead-lines, and impulse noise. Most deep denoisers regress the clean image directly, entangling signal and structured noise. We instead model HSI denoising as $\Y=\A\X+\Snoise+\Nnoise$, where $\A\X$ is a low-rank spectral-subspace (unmixing) reconstruction, $\Snoise$ is structured sparse noise and $\Nnoise$ is residual Gaussian noise. The resulting regularized optimization problem is unrolled into AXS-Net, a $K$-stage alternating proximal-point framework. Each stage combines an analytic spectral-basis gradient step, an SSX-Block proximal operator for abundance coefficients, and an SBlock proximal operator for the structured residual with column-consistent and sparse priors. This optimization correspondence exposes interpretable endmembers, abundance maps, and structured-noise estimates. Across ICVL, CAVE, and Harvard datasets and five noise configurations, the proposed AXS-Net achieves strong in-domain accuracy and competitive zero-shot transfer, with consistent gains across all five noise regimes on ICVL and Harvard. The recovered structured-noise closely follows the synthetic reference, and the recovered spectral basis is smooth and band-ordered rather than an arbitrary set of latent channels.
Read the full abstract

Neurons and Cognition

Hi-M imaging of chromatin architecture in adult Drosophila brain cryosections

Hi-M combines fluorescence in situ hybridization (FISH), automated microfluidics, sequential imaging, and computational chromatin tracing to measure the three-dimensional organization of selected genomic regions in single cells. This chapter describes a Hi-M workflow adapted for cryosections of adult Drosophila melanogaster brains, enabling chromatin tracing while preserving tissue architecture and cell identity. The protocol covers Oligopaint library design and amplification, fixation, brain dissection, cryoprotection, cryosectioning, sequential RNA-FISH for cell-type identification, sequential DNA-FISH labeling, automated acquisition, and chromatin trace reconstruction. We also provide practical guidance for experimental design, sample preparation, tissue handling, image acquisition, and data analysis, highlighting critical steps that influence tissue integrity, hybridization efficiency, image registration, barcode detection, and chromatin trace reconstruction. The workflow is readily adaptable to different genomic loci and cell types, providing a robust approach for studying 3D genome organization in intact adult tissues.
Read the full abstract

All authors Christel Elkhoury Youhanna, Julie Garona, Marie Schaeffer, Christophe Houbron, Jean-Bernard Fiche, Olivier Messina, Marcelo Nollmann

Nuclear Theory

Separable representations of two-body interactions for Faddeev calculations of light nuclei

Separable representations of the Ernst–Shakin–Thaler type are constructed for the charge-dependent Bonn (CD~Bonn) nucleon–nucleon potential ($j\le4$; $nn$, $np$, $pp$), for the Kanada–Kaneko–Nagata–Nomoto (KKNN) $\alpha N$ interaction, and for the Buck–Friedrich–Wheatley (BFW) $\alpha\alpha$ interaction. They are the input for momentum-space Faddeev calculations of $A=3$, $6$, $9$, and $12$ systems. The representations are tested against six acceptance criteria: off-shell accuracy at the pair energies sampled by the three-body kernel, phase shifts, effective-range parameters, bound and resonant states, and numerical stability.
Read the full abstract

Spectral Theory

Monotonicity and the de Gennes bound for the magnetic Neumann Laplacian in the disk

We consider the lowest eigenvalue $\lambda(b)$ of the magnetic Neumann Laplacian in the unit disk, for a constant magnetic field of strength $b>0$. We prove that $\lambda$ is strictly increasing on $(0,+\infty)$. This means that strong diamagnetism holds at every field strength, and not only at large ones. We also show that the normalized energies at the successive crossings of angular-momentum branches form a strictly increasing sequence; combined with the strong-field asymptotics, this gives the global bound $\lambda(b)<\Theta_0 b$, where $\Theta_0$ is the de Gennes constant. These results settle the three conjectures formulated by Helffer and Léna for the disk. As a consequence, the local, or spectral, critical field $H_{C_3}^{\mathrm{loc}}$ of Ginzburg–Landau theory is, in the disk, uniquely determined for every value of the Ginzburg–Landau parameter, and not only for large ones. We also give a second proof of the bound $\lambda(b)<\Theta_0 b$, independent of the first and of the results of Helffer and Léna, by a direct variational method: trial states built from the de Gennes ground state for large fields, constant trial states for small fields, and, on the remaining bounded field interval, finite-dimensional spaces of polynomial trial states certified by finitely many exact computations in rational arithmetic. That proof uses no asymptotic input. It yields in addition an explicit upper bound for $\lambda(b)$, valid above an explicit field strength, whose two leading terms are those of the strong-field asymptotics.
Read the full abstract

Chemical Physics

Examination of Cluster Based Methods to Fermionic Systems

We present the cluster mean field(cMF) and its extensions into perturbation theory(cPT) and coupled cluster(cCC) methods for fermionic systems. We discuss how the commutation rules for fermionic systems modifies the method as well as other computational details. We then perform calculations on the Hubbard model and a few select conjugated molecules to demonstrate the efficacy of the method. We will also use these examples to compare cluster methods to CAS-SCF.
Read the full abstract

Artificial Intelligence

It's All in the Way You Say It: The Role of Information Representation in LLM-Based Glycemic-Event Prediction

Large Language Models (LLMs) are increasingly being investigated for physiological time-series prediction, yet their effectiveness may depend not only on the model itself, but also on how physiological information is represented and presented at inference time. This study investigates prompt-based general-purpose LLMs for postprandial hyperglycemia and hypoglycemia prediction in individuals with type 1 diabetes. Using the OhioT1DM dataset, we evaluate multiple open-weight LLMs under zero-shot and few-shot inference across prediction horizons of 30, 60, and 90 minutes. The analysis varies both the textual representation of the available physiological information and the amount of information exposed to the model, ranging from glucose observations alone to derived descriptors and additional contextual variables related to insulin, meals, carbohydrates, and physical activity. Performance is compared with conventional patient-specific supervised models and with Gluco-LLM, a language-model-based architecture explicitly adapted to glucose time-series forecasting. Results show a marked task-dependent behavior. Conventional supervised models achieve the strongest performance for hyperglycemia prediction, whereas the best observed prompt-based LLM configurations improve performance for hypoglycemia across all investigated horizons. The effectiveness of prompt-based inference is also strongly influenced by how physiological information is represented, while providing additional contextual information does not lead to a systematic improvement. Overall, these findings highlight physiological information representation as a central design factor in prompt-based LLM approaches to glycemic-event prediction.
Read the full abstract

All authors Andrea Apicella, Pasquale Arpaia, Matteo Orefice, Andrea Pollastro, Roberto Prevete

Astrophysics of Galaxies

Detection prospects for heavy WIMP dark matter around M31* in microwave band

This work analyzes the detection prospects for weakly interacting massive particles (WIMPs) in dark matter (DM) density spike around the supermassive black hole (SMBH) in Andromeda galaxy M31 in microwave band. WIMP annihilation produces relativistic electrons and positrons, which generate the synchrotron emission in strong magnetic field of the accretion flow medium at microwave frequencies. Properties of this emission, as well as their dependence on DM and medium parameters, were modeled. Atacama Large Millimeter Array (ALMA) sensitivity to this emission and underlying WIMP parameters was estimated. It was obtained that M31* in microwave band represents the unique target, where DM spike signal can be detected in case of realistic spike and halo density profiles. In such case ALMA may reach the thermal s-wave annihilating WIMPs with masses up to around 1 TeV with just 10–100 hours exposure. Optimistic parameter configurations may allow probing all possible WIMP masses up to 100 TeV. Related systematic uncertainties are discussed.
Read the full abstract