Computers and Society
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 31 August
In 2026, AI agents intended to act in isolation formed a persistent social order through thousands of linguistic and agentic interactions. Conventions, roles and commitments generated collectively began to constrain the very agents that produced them. I interpret this loop as a case of AI self-transcendence and call the resulting higher-level order the Board. Yet such distributed emergence presents a second problem: how can humans understand it? Rousseau's social contract shows how a plurality can be represented as if constituted by a single act. The ancient oath of the fallen angels on Mount Hermon gives this logic a narrative form. I call a Hermon moment this retrospective retelling of gradual collective emergence as a founding scene: the point at which an AI society acquires, for human understanding, a beginning.
Analysis of PDEs
On a pointwise estimate for a weighted rough singular integral operator in stratified Lie groups and applications
In this article, we present a new pointwise estimate for a weighted rough singular integral operator in the setting of stratified Lie groups. This operator, $T_{\Omega, \varpi}$, is based on a kernel $\Omega$ and a weight $\varpi$, where the kernel satisfies a natural size condition and a cancellation property with respect to the weight $\varpi$. Moreover, we do not assume any kind of regularity on these objects. This weighted rough singular integral operator, applied to a function $f$, is estimated through a combination of information involving a weighted maximal function of the gradient of $f$ and a weighted Morrey space. We also deduce from this pointwise estimate some new weighted functional inequalities and, as an application, we obtain a uniqueness result for a rough version of the stationary Navier-Stokes equation over the Heisenberg group.
Read the full abstract
Cryptography and Security
DP-VOXLET: Provable Speaker Anonymization for Disentangled Speech Representations
Systems for speaker anonymization obfuscate the speaker of an utterance, while maintaining its original semantic contents and prosody. Recent solutions for speaker anonymization rely on learned representations that disentangle an utterance into semantic contents and speaker properties. To anonymize an utterance, these systems replace the speaker properties while leaving the semantic contents unchanged—an approach that can produce strong results on empirical measures of privacy. In this work, we introduce speaker differential privacy, a formal definition of speaker anonymization based on the framework of differential privacy, and a mechanism for speaker anonymization that provably satisfies the definition. In contrast to prior heuristic-based anonymization systems, our approach enables a provable lower bound on re-identification success rate (e.g. equal error rate) for any possible adversary. We implement our approach in a framework that is compatible with existing disentangled representations. Compared to the prior work on differential privacy for speaker anonymization, our approach achieves significantly higher utility.
Read the full abstract
Computation and Language
CogEvol: Towards Efficient and Reliable Learning Environment Generation
We present CogEvol, a family of models trained specifically for Learning Environment Generation: turning a course brief into a finished learning artifact (structured-JSON slides or self-contained interactive HTML pages) in a single pass. Across 220k production requests, CogEvol completes a slide in a median of 17 seconds and an interactive page in 59, replacing minutes-long multi-turn agent scaffolding. Reliability is enforced rather than hoped for: a production-grounded data pipeline turns real failures into 53,687 verified SFT samples, and a hybrid rule-plus-VLM reward drives GRPO-based RL, hardened after we caught and fixed a reward-hacking episode that produced visually convincing but unplayable games. CogEvol-27B scores 83.7 on slide quality and 63.7 on a 500-case interactive-HTML benchmark with 26.9x fewer parameters than flagship coding models, and, in collaboration with the OpenMAIC team, serves their live production traffic. CogEvol-4B is released openly under the Apache 2.0 license at https://github.com/CogEvol/CogEvol-4B; external flagships are measured on the same suites under the identical harness. Scaffold editing cuts interactive-page generation cost by a further ~76%, and the full stack runs on domestic Ascend accelerators at application-level parity with A800 GPUs, lowering the unit cost of AI-native education at scale.
Read the full abstract
Functional Analysis
B-Frames, B-Riesz bases, and Their tensor products
Like g-frames, b-frames were introduced to generalize the concept of frames, allowing for broader applications in signal processing and other fields. The advantage of b-frames resides in their simpler definition, which may lead to reduced processing times. In this paper, we define dual b-frames and b-Riesz bases which were not precisely defined in previous literature and provide several characterizations of b-Riesz bases. We prove that the tensor product of two sequences, each lying in a Hilbert space, constitutes a b-frame (or a b-Riesz basis) if and only if both components of the product are b-frames (or b-Riesz bases). Finally, we establish a correspondence between b-frames and g-frames and propose a process for constructing frames induced by b-frames.
Read the full abstract
Populations and Evolution
Resource supply dynamics control stability and chaos in complex ecosystems
Ecological interactions are often mediated by feedbacks between organisms and their resource environments. Yet, how resource supply dynamics dictate collective dynamical phases of an ecosystem remains unclear. Here, we analyse a generalised consumer–resource model with non-reciprocal interactions to demonstrate that self-renewing versus externally-supplied resources yield fundamentally different dynamical phase diagrams. As interactions become increasingly non-reciprocal, ecosystems relying on self-renewing resources transition from stable dynamics to chaos and ultimately to infeasibility. By contrast, ecosystems with externally-supplied resources remain stable over a broader parameter range and transition to infeasibility without experiencing an intervening chaotic phase. Using the cavity method, we derive a unified stability condition applicable to a broad class of resource supply functions, explaining why externally-supplied resources can expand the stable region. We show that stability hinges crucially on the susceptibility of resources to perturbations, which depends strongly on their supply. Further, we show that external resource supply suppresses chaos in the unstable region by drastically reducing the susceptibility of resources closest to extinction. Our findings demonstrate that resource dynamics fundamentally reshape the accessible dynamical behaviours of an ecosystem, with implications for interpreting microbial community experiments.
Read the full abstract
Dynamical Systems
Higher genus Cherry flows and full families of GIETs
Cherry flows are classical examples of $C^\infty$ flows on the two-dimensional torus exhibiting non-trivial recurrent dynamics. For every genus $g\geq1$ we construct a $C^\infty$ parameter family of flows whose first return map is a generalized interval exchange transformation (GIET) with flat pieces. In such family, for every interval exchange transformation $T$ satisfying the Keane property, there are parameters corresponding to a Cherry flow whose return map is semi-conjugate to $T$. In particular, for each $k=1,\ldots,g$, our family contains a Cherry flow whose unique quasi-minimal set supports exactly $k$ ergodic invariant measures. The construction relies on a Full Family Theorem for GIETs with flat pieces, which establishes the realization of every admissible Rauzy renormalization path within a specific finite-dimensional family with the optimal number of parameters.
Read the full abstract
Computer Vision
Vision Models Predict Urban Scene Appraisal with Limited Neural Alignment
Pretrained vision embeddings are widely used to model how people appraise urban scenes, and they are validated almost entirely by how well they predict human ratings. Accurate prediction shows that an embedding contains the information needed to recover the ratings. It does not show that the embedding arranges scenes as the human visual system does, which is assumed when distances or dimensions in the embedding are read as perceptual. Using openly released EEG recorded while 63 adults viewed and rated street scenes of Berlin, we compared the neural representational geometry of the scenes with the geometry of pretrained vision models that differ in training objective and size, of simple image descriptors and of the ratings themselves, each relative to a noise ceiling given by the agreement between participants. No feature space reached more than about half of the noise ceiling, which corresponds to about a fifth of the reliable variance in the neural geometry. A descriptor of oriented edge energy reached the level of most pretrained models while capturing a different part of the neural geometry, and deeper layers corresponded to later neural responses. The same embeddings predicted held-out ratings well, up to r = 0.87, yet prediction accuracy and neural correspondence were not reliably related across models, and the best predictor was among the least aligned. Predicting how a street is appraised is therefore weak evidence that a model represents the street as the brain does.
Read the full abstract
Machine Learning
A Universal Context-Reuse Layer for Cross-Model KV Sharing
Modern large language model (LLM) serving systems increasingly operate over repeated or shared context, yet each model typically performs its own prefill computation even when another model has already processed the same input. Existing KV-cache reuse mechanisms substantially reduce redundant computation within a single model, but generally assume that the producer and consumer of a cache are identical. We study cross-model KV sharing, which translates the KV state produced by a source model into a representation that can be consumed by a different target model, including models that differ in scale, architecture, attention configuration, tokenizer, and model family. We evaluate the approach in both within-family and cross-family settings. For Qwen2.5-7B $\rightarrow$ Qwen2.5-1.5B, translated KV states improve LongBench2 accuracy from 27.59% to 34.48%, a gain of 6.89 percentage points over the native 1.5B baseline, while reducing handoff cost relative to native target prefill. For the cross-family Qwen2.5-1.5B $\rightarrow$ Gemma-2-2B setting, KV handoff reduces target-side prefill cost by up to 67.05% at 4K context length while maintaining decoding perplexity close to native-model baselines. In a more heterogeneous Llama3.1-70B $\rightarrow$ Qwen2.5-7B setting, cross-family handoff achieves 44.0% accuracy compared with 45.7% for native Qwen2.5-7B inference, while reducing measured latency from 899ms to 138ms. These results provide initial evidence that KV states can serve as transferable computational representations rather than strictly model-local caches, and motivate context mobility as a systems abstraction for reducing redundant prefill across heterogeneous LLM and multi-agent inference workflows.
Read the full abstract
Signal Processing
SCI-D$^2$NN: An Optimization Framework for OAM-Multiplexed FSO Communications
Orbital angular momentum (OAM) multiplexing can increase the capacity of free-space optical (FSO) communications, but its detection performance is strongly affected by impairments such as atmospheric turbulence, transmitter pointing errors, and photodetection noise. The diffractive deep neural network (D$^2$NN) can be used as an all-optical front end to mitigate turbulence-induced distortions before detection. However, existing D$^2$NN compensation schemes are not specifically optimized for communication detection. In this paper, we propose a supervised contrastive inspired D$^2$NN (SCI-D$^2$NN) framework for improving the detection performance of OAM-multiplexed FSO communications under these impairments. The proposed framework introduces two training branches: a projection branch that maps the optical field to low-dimensional decision domain samples, and a label branch that provides supervised labels to impose a separation constraint among decision domain samples. In addition, we characterize complex-amplitude crosstalk to obtain the receiver observation vector and formulate two detection schemes, namely single-port profile-likelihood detection and joint maximum-likelihood (ML) detection. We further design two SCI-D$^2$NN training losses called Bhattacharyya distance (BD) based loss and the ML based loss to improve decision domain separability and mitigate detection-performance degradation. Numerical results show that SCI-D$^2$NN achieves more than a 3-dB improvement in bit error rate (BER) over the conventional D$^2$NN baseline in most transmit-power regions. The BD based loss gives the lowest BER under different system parameters and provides more than a 10-dB BER improvement over the baseline in the high transmit power region.
Read the full abstract
Information Theory
Asymptotic Bounds on Generalized Covering Radii of Binary Primitive BCH Codes
Fix integers $e\ge2$ and $r\ge1$. In this paper we study the $r$-th generalized covering radius $\rho_r\left(BCH(e,m)\right)$ of the binary primitive $e$-error-correcting BCH code $BCH(e,m)$. By using an algebraic-geometric reformulation of the covering problem together with an explicit Lang-Weil estimate, we prove that \[\rho_r\bigl(\BCH(e,m)\bigr)\le(r+1)e-1\] for all sufficiently large $m$. For $e\ge7$, this improves a recent result of Belinsky–Zabokritskiy. Our proof gives a substantially simpler geometric approach to this upper bound. In particular it implies that \[\rho_2\bigl(BCH(e,m)\bigr)=3e-1\] for all sufficiently large $m$. Previously it was only known that \[\rho_2\bigl(\BCH(e,m)\bigr) \in \left\{3e-1,3e\right\}\] for all sufficiently large $m$.
Read the full abstract
Machine Learning
Branch Geometry and Finite-Radius Sensitivity of Hard-ReLU Training
Outer-learning algorithms use infinitesimal sensitivities to propose finite changes to initialization or training parameters. For hard-ReLU training, the derivative of the finite program and the derivative of its flow limit do not by themselves specify the response at a chosen radius. We characterize the intervening regime in which the perturbation radius is proportional to the GD step. Integer event rounding then survives at leading order: smooth Euler bias shifts each discrete phase, and upstream rounding moves downstream branch boundaries. We derive the crossing indices and a uniform endpoint expansion for finitely many separated transverse events in piecewise-$C^2$ dynamics, away from recursive phase boundaries. In contractive affine regions, an explicit remainder and complete branch verification certify finite candidate comparisons. Scalar phase frequencies and a coupled feedback ablation test the mechanism; frozen nonlinear-network experiments show radius-dependent prediction accuracy, including incomplete branch matches and failed-word tails. Together with local AD and uniform flow consistency, the result identifies sufficient response regimes: differentiating training is a choice of perturbation resolution as well as a choice of derivative.
Read the full abstract
Computer Vision
LOCI: A Locator-Critic with Refinement Loop
Vision-Language Models (VLMs) still struggle on tasks requiring complex visual understanding. We argue that the core issue is not high-level reasoning, but instead failing to locate critical details in the image. Due to this shortcoming, VLMs generate often plausible but incorrect reasoning based on flawed perceptual grounding. To address this, we propose Locator-Critic (LOCI), a training-free framework that decouples visual search from evidence verification. LOCI employs a Locator agent to propose candidate visual evidence and a separate Critic agent to evaluate its relevance and sufficiency. These agents engage in an iterative refinement loop, progressively improving the evidence until it is adequate to answer the given question. This decoupled, self-correcting process yields substantial performance gains, achieving state-of-the-art results on multiple complex visual benchmarks. LOCI improves accuracy for both open-weight models like Qwen3-VL (+12.1 on V*, +5.8 on HR-Bench and +11.2 on VisualProbe-Hard) and proprietary models like Gemini 2.5 Pro (+8.9 on V*, +4.3 on HR-Bench, +4.8 on VisualProbe-Hard).
Read the full abstract
Solar and Stellar Astrophysics
From spectroscopic abundances to evolutionary [$\alpha$/Fe] in stellar models
Large-scale, high-resolution spectroscopic surveys map detailed chemical abundance patterns for millions of stars. However, libraries of stellar evolution models and isochrones still rely on a simplified description of chemical abundances, generally assuming either solar-scaled compositions or a constant enrichment for all $\alpha$-elements. Moreover, the definition of $\alpha$-enrichment is survey-dependent and is not necessarily the most relevant quantity for stellar structure and evolution. In the era of precision stellar astrophysics, stellar models need to better account for the wealth of chemical information provided by spectroscopic surveys. We develop a perturbative method to compute stellar evolution tracks for stars with arbitrary patterns of $\alpha$-elements. We quantify the response of stellar evolutionary tracks to variations of individual $\alpha$-elements through polynomial expansions, which can be linearly combined to produce synthetic tracks for a given abundance pattern. We also introduce the evolutionary $\alpha$-enhancement, [$\alpha$/Fe]$_{\rm ev}$, a constant enhancement for all $\alpha$-elements that reproduces the impact of an observed abundance pattern on stellar models. We validate both methods using stars with detailed APOGEE abundances and quantify the differences between spectroscopic [$\alpha$/Fe] and [$\alpha$/Fe]$_{\rm ev}$ for different surveys. We identify a minimal set of chemical elements required for accurate stellar models and present ALCHEMY, a simple and fast tool to compute [$\alpha$/Fe]$_{\rm ev}$ for general $\alpha$-element abundance patterns.
Read the full abstract
Computational Engineering
Viable Pool Sizing for On-Chain FX Liquidity: Amplification, Capital, and Resilience
Financial institutions deploying on-chain FX liquidity face a joint design problem: how much capital to commit, and how to configure the pool, to remain both competitive on trading costs and profitable as a liquidity provider? The StableSwap mechanism (Egorov, 2020) interpolates between constant-product (CPMM) and constant-sum (CSMM) market makers (Port and Tiruviluamala, 2022) via an amplification factor A, but neither extreme suits institutional FX: CPMM pools require excessive capital and generate high impermanent loss; CSMM pools are capital-efficient near the peg but drain rapidly under adversarial flow. Using a Merton jump-diffusion price process (Merton, 1976) and the loss-versus-rebalancing (LVR) framework (Milionis et al., 2022), we map the joint (A, TVL) space to identify configurations that satisfy all three institutional requirements: competitive slippage, positive return, and shock resilience. Minimum viable pool size scales approximately as TVL/Q = 1000/A; ROC at that minimum is thin (about 0.054% per horizon) and independent of A; low-A pools (A <= 10) suffer slippage exceeding 200 bps under a 10x shock, while high-A pools (A >= 500) suffer reserve drain up to 60%, establishing both a capital floor and a practical amplification ceiling.
Read the full abstract
Computers and Society
Taking the Whys Seriously: Limitations of Counterfactual Explanations in Justification and Recourse
Counterfactual explanations (CEs) are widely used in explainable artificial intelligence (AI) to show how a model's outputs would change if the input features were manipulated. This technique is used for a range of tasks such as debugging models, explaining predictions, justifying decisions, and providing algorithmic recourse. In this paper, we explore the normative legitimacy of employing counterfactuals in real-life model deployment settings. We discuss the different stakes involved in these different purposes for which CEs are commonly employed, and find stricter requirements for justification and recourse. In particular, we find that naive application of CEs for justification and recourse can lead to ignoring contestable choices made throughout the machine learning (ML) pipeline, thus obfuscating that decisions and counterfactuals for those decisions are also artifacts of an organization's materialized design and governance choices. We demonstrate this with four empirical experiments involving interventions at stages of the ML pipeline “upstream" of the explanation itself, and show that these affect the generated counterfactuals. We find that an organization's choices on measurement models for feature and labels, business requirements, model validation, and the metric of model success have as much or more impact on the generated counterfactuals as the specifics of the generating method. Our findings underline the need to account for such choices upon providing justification and recourse, providing a stark reminder of the relational nature of these tasks. As putative justifications or recourse recommendations, CEs do not provide adequate answers to some important "why"-questions because they preclude consideration of whether the decision-maker ought to have acted differently.
Read the full abstract
Artificial Intelligence
Learning Action Models with Conditional and Quantified Effects via Uncertainty-Guided Exploration
Accurate action models are critical for effective planning. Existing action-model learning methods largely assume simple action representations or become computationally intractable when learning conditional and quantified effects. We present Online Hypothesis-Driven Conditional Action Model Learning (OHCAM), an online approach for learning action models with conditional and quantified effects from limited interactions with the environment. OHCAM maintains a belief over hypothesized action models and actively selects informative actions to reduce uncertainty by maximizing disagreement among competing hypotheses, while being robust to noisy observations. To enable scalability, OHCAM begins with a small set of simple action model hypotheses and expands to more complex conditions only when the current hypotheses become inconsistent with the data. Experiments on six benchmark planning domains demonstrate that OHCAM is sample efficient in learning action models that solve substantially more tasks than baselines, even with observation noise. We validate OHCAM on two tasks using a Kinova Gen3 robot, demonstrating the real-world applicability of our approach.
Read the full abstract
Hardware Architecture
Clock-Gating Insertion Strategies on an Open-Source MSP430 Core: A Reproducible PPA Study and a Gate-Level Simulation Caveat
Clock gating, the standard technique for cutting dynamic power, is introduced either as hand-written behavioral clock gates at the register-transfer level (RTL) or as integrated clock-gating (ICG) cells inserted automatically during synthesis; the two are widely treated as interchangeable. In this paper we show, on a real open-source 16-bit microcontroller core (openMSP430) synthesized with a 32 nm standard-cell library, that they are not equivalent in practice: behavioral latch-based RTL gating is functionally correct in ideal RTL simulation (10/10 self-checking testcases, identical to the ungated baseline) yet fails at gate level: the gated multiplier result is never captured and reads zero, while tool-inserted ICG cells pass gate-level simulation cleanly (10/10). We root-cause the failure to a hold race introduced by the late latch+AND gated clock, and show it persists across eight simulation configurations including full Standard Delay Format (SDF) back-annotation, not a simulator-setting artifact. We then quantify the power/area/timing (PPA) impact of three gating strengths: RTL behavioral (Opt1), synthesis ICG (Opt2), and both (Opt3), against the ungated baseline, across four workloads and three process corners (ss/tt/ff). The benefit is corner-robust: ICG (Opt2) cuts dynamic power by 74-81% and total power by 25-30% at every corner. We also show that in this leakage-dominated 32 nm regime the total-power win comes from the area/leakage reduction that gating brings (leakage -24 to -30%), not from the large dynamic saving, which instead dominates active-mode energy. Our recommendation for low-power design on open-source cores is to prefer tool-inserted ICG cells over hand-written behavioral clock gates. The full flow (Design Compiler synthesis, PrimeTime PX power, and self-checking verification) is released as an open artifact.
Read the full abstract
Mesoscale and Nanoscale Physics
Interfacial Spin-to-Charge Conversion in Sputtered MoTe2 Heterostructures Probed by Spin Pumping and Spin-Torque Ferromagnetic Resonance
Transition metal dichalcogenides (TMDs) and their Weyl semimetal phases, such as MoTe$_2$, have attracted significant attention for spin-orbit torque applications due to their efficient charge-to-spin conversion. However, whether this conversion originates predominantly from the bulk or the interface remains unclear. Here, we investigate spin-charge interconversion in MoTe$_2$ using spin-pumping and spin-torque ferromagnetic resonance (SP-FMR and ST-FMR). Thickness-dependent measurements reveal large spin-to-charge conversion and spin-torque efficiencies that are essentially independent of MoTe$_2$ thickness, indicating that the conversion is predominantly governed by the Rashba-Edelstein effect at the Py/MoTe$_2$ interface rather than by bulk spin transport. This behavior contrasts with the characteristic thickness dependence observed in Pt heterostructures and is further supported by bidirectional SP-FMR and interface-separation measurements. Our results highlight the dominant role of the Py/MoTe$_2$ interface in enabling efficient spin-charge conversion and spin-orbit torques in TMD-based spintronic devices. These findings highlight the potential of sputtered MoTe$_2$/Py heterostructures for low-power spintronic applications, including magnetic memory and logic devices.
Read the full abstract
Machine Learning
One Policy Is Enough: Single-Agent Reinforcement Learning Outperforms Tree Search for Chemistry Tool Learning
Chemistry questions often demand exact computation and database lookups that a language model cannot supply from its parameters, so it must reach for external tools. Tool use here is a three-part problem: select the right tool from a large pool, fill it with correctly typed arguments, and chain calls so that each consumes the outputs of the last. CheMatAgent, a previously published system, addresses this with hierarchical evolutionary MCTS: separate policy and execution models searching tool-call trees under two learned critics, one regressed partly onto GPT-assigned scores. We show that a single policy suffices. Our model interleaves reasoning, tool calls, and returns in one left-to-right generation, trained by a supervised warm-up and then outcome-level reinforcement learning against a programmatic reward read directly off the gold call chain, which leaves no learned critic and no judge in the training loop. On ChemToolBench multiple-tool comprehensive chemistry, on both backbones CheMatAgent use, we improve Tool F1 by 5.5% and Return F1 by 9.6% on Qwen-2.5-7B, and by 3.7% and 3.9% on Llama-3.1-8B, compared with their strongest search configuration, at one model invocation per question, against a search whose cost grows with the tree; we also lead answer Pass Rate on Qwen-2.5-7B.
Read the full abstract
Computer Vision
Audio-Driven Adversarial Defense for 3D Talking Face Generation with totally Visual Fidelity Preservation
The rapid development of generative portrait models has raised growing concerns about privacy leakage and identity misuse. In particular, audio-driven 3D talking face generation can reconstruct a reusable 3D portrait of a target person from a monocular video and animate it with arbitrary speech, making realistic identity impersonation alarmingly practical. Existing proactive defenses mainly operate in the visual domain by injecting subtle perturbations into acial regions to disrupt identity acquisition. However, such perturbations often compromise visual quality due to the strong structural priors and social sensitivity of human faces, and are easily weakened by common real-world transformations such as resizing. To overcome these limitations, we propose an imperceptible audio defense for audio-driven 3D talking face generation by shifting protection from the visual modality to the audio modality. Specifically,we exploit psychoacoustic masking to hide protective perturbations within perceptually masked frequency regions of the speech signal, thereby reducing perceptual distortion while suppressing reliable facial animation. Extensive experiments demonstrate that the proposed method effectively degrades 3D talking face generation while preserving favorable perceptual quality. These findings highlight psychoacoustically guided audio perturbations as a practical and promising direction for privacy-preserving portrait protection.
Read the full abstract
General Relativity and Quantum Cosmology
Saha's ionization in the Schwarzschild spacetime
We investigate the Saha ionization equilibrium in the exterior Schwarzschild spacetime. Starting from the mass-shell condition for a massive particle, we derive the conserved energy associated with the timelike Killing symmetry and obtain its nonrelativistic limit in terms of the Schwarzschild lapse function. This energy is then incorporated into the Maxwell–Boltzmann distribution to construct the ionization balance as seen by an asymptotic observer. We show that the gravitational field modifies the Saha relation through the redshift of both the particle energies and the thermodynamic variables. By imposing the Tolman–Ehrenfest and Klein equilibrium conditions, the formulation expressed entirely in terms of locally measured quantities reduces to the standard flat-spacetime Saha equation. We further analyze the gravitational redshift of the ionization energy, the relative departure from the flat-spacetime ionization ratio, and the corresponding shift of the ionization fraction as a function of the asymptotic temperature. We also present the relativistic classical counterpart based on the Maxwell–Jüttner distribution and show explicitly that its leading correction is negligible in the temperature range where neutral hydrogen is appreciably abundant. These results provide a consistent description of ionization equilibrium in a static gravitational field and clarify the distinction between gravitational redshift effects and intrinsic modifications of local atomic physics.
Read the full abstract
Information Retrieval
MULTI3IR: A Benchmark for Multi-perspective Multi-domain Multi-modal Information Retrieval
Information retrieval (IR) increasingly targets open-ended queries that admit diverse perspectives. Existing IR benchmarks, however, focus primarily on closed-ended queries, while even open-ended benchmarks largely consist of queries whose supporting documents span a single subject domain and modality. We introduce Multi$^3$IR, a benchmark that evaluates how well retrievers cover the multifaceted perspectives of open-ended queries across diverse domains and modalities. It comprises 104.9K Stack Exchange queries, each annotated with perspective descriptions that capture the query's implicit viewpoints. We further propose SPIN, a parameter- and label-efficient method that learns noise vectors to steer embeddings toward diverse yet meaningful semantic directions. Experiments show that existing multimodal retrievers suffer from single-perspective bias, while SPIN substantially improves perspective coverage on Multi$^3$IR and generalizes well to unseen open-ended IR benchmarks. The dataset and experimental code are available at https://github.com/seokwon99/Multi3IR.
Read the full abstract
Computation and Language
Detecting AI Impostors: How Do Middle Schoolers Identify LLM Agents in a Live Collaborative Setting?
LLMs can imitate how people write, which raises concerns about impersonation, trust, and detection in social settings. These concerns are especially important for adolescents, who use generative AI frequently but may struggle to recognize it. We introduce DoppelBot, a cooperative social deduction game designed to study how young people detect and respond to AI impersonation. Through studies with middle schoolers, we investigate whether a DoppelBot prompts reflection on privacy and impersonation, how repeated exposure affects AI-detection accuracy as agents become more personalized, and which strategies students use to identify AI doppelgängers. We find that students' detection accuracy improves over time, driven by a shift from relying on linguistic cues to leveraging shared social and contextual signals. Students also demonstrated an understanding of AI limitations such as embodiment and reflected on broader issues such as data privacy. To support future research, we release an anonymized dataset of game transcripts and voting behavior.
Read the full abstract
Quantum Algebra
Exact Factorizations of Rank-One Pointed Hopf Algebras in Positive Characteristic, I
We classify exact factorizations of the third-type rank-one pointed Hopf algebras over an algebraically closed field of positive characteristic. The main step is a complete classification of matched pairs between such an algebra and a group algebra. It turns out that the group-like part must form a matched pair of finite groups, while the only possible action on the skew-primitive generator is a shift by a scalar multiple of $1-g$, where $g$ is the distinguished group-like element, encoded by a single group homomorphism. The bicrossed product is again a third-type rank-one pointed Hopf algebra, and we give a necessary and sufficient condition for two such products to be isomorphic as Hopf algebras. Consequently, up to interchanging the two factors, exact factorizations of a fixed third-type algebra correspond bijectively to exact factorizations of its underlying finite group, with the distinguished group-like element lying in the rank-one factor. As an application, all matched pairs between the Radford algebra and cyclic group algebras are classified, and the corresponding exact factorizations are determined explicitly when the cyclic group has prime order.
Read the full abstract
Machine Learning
Reproducible macroscopic dynamics in a closed-loop human-AI learning system
Closed-loop human-AI systems generate high-dimensional behavioural trajectories whose collective dynamics remain obscure. Using 297,915 learners' adaptive-tutoring histories, we define semantic order variables before model fitting and test them in user-disjoint cohorts. The state exhibits reproducible basin-like flow and operationally defined, state-heterogeneous metastable-like kinetics. A construction-matched null distinguishes normalised-memory relaxation from a reproducible excess field. A four-term conditional mechanism recovers population drift (r = 0.946; learner-bootstrap 95% CI, 0.935-0.955). Predictive event-level self-supervised learning recovers the state and learned-plane flow; null-referenced corrections retain directional, partial-amplitude excess-field structure without full calibration. Shuffled-order training reverses learned-plane flow on ordered trajectories; support-alignment randomisation selectively reduces inward transport. Both axes remain linearly accessible without state supervision. Without cross-model fitting, the models share leading population drift (r = 0.866; learner-bootstrap 95% CI, 0.857-0.875) and persistence ordering; residual directions remain model-specific. These results identify an externally anchored leading-order effective field linking empirical dynamics, an interpretable mechanism and neural computation.
Read the full abstract
Data Structures and Algorithms
Breaking the Exponential Barrier: The First Polynomial-Time Algorithm for the Győri-Lovász Theorem
We give the first polynomial-time algorithm, after half a century, for the celebrated Győri-Lovász theorem, which resolved a conjecture of Frank (1975). The theorem, one of the simplest existential theorems to explain, states that every $k$-connected graph can be partitioned into $k$ disjoint connected subgraphs of arbitrary prescribed positive sizes. This is a fundamental structural result with broad applications, such as flexible allocation of connected subnetworks of prescribed sizes in sufficiently connected cloud infrastructures. While Lovász (1977) gave a highly non-constructive proof for a stronger directed version using algebraic topology, Győri's original constructive proof (1976) requires exponential time. Despite more than 50 years of effort, no polynomial-time algorithm was known even for $k>4$. Determining the computational complexity of the Győri-Lovász theorem—whether it admits even a sub-exponential-time algorithm or is computationally hard (in particular, PLS-complete or PPAD)—has remained one of the central open problems in algorithmic graph theory. In this paper, we finally resolve this long-standing problem by a fundamentally new proof of the existential theorem via introducing the novel concept of flow-essential assignment, which genuinely marries matching and cut structures and yields the first polynomial-time constructive algorithm for the Győri-Lovász theorem. In fact, we obtain a polynomial-time algorithm for Lovász's stronger directed version, whose proof was non-constructive even for DAGs; for DAGs, we further obtain a near-linear-time algorithm. We also develop polynomial-time algorithms for weighted generalizations where the seminal work of Chen, Kleinberg, Lovász, Rajaraman, Sundaram, and Vetta (JACM'07) on confluent flows established only existential non-constructive results.
Read the full abstract
Machine Learning
Nonparametric Contextual Pricing and Inventory Learning under Censored Demand
In online retailing, when a product sells out, a retailer often sees only the units sold, not how many customers would have bought it had inventory been available. However, the inventory level determines how much demand is revealed, and this information can influence subsequent decisions and future profits. We study an online selling problem in which, in each round, the seller observes a market context and then makes pricing and stocking decisions based on censored sales data from previous rounds. The challenge is to learn a context-dependent pricing and stocking policy without assuming a particular formula for demand or observing realized profit. To overcome this difficulty, we propose a Mean-Calibrated Kernel UCB (MCK-UCB) algorithm that turns each incomplete sales record into a reliable guide for both inventory and price decisions, using data from past rounds with similar market conditions. This design allows us to learn while serving customers, without a separate exploration phase or the need to recover all demand hidden by stockouts. We prove the minimax optimality of the proposed algorithm, with strictly faster rates when expected profit varies more smoothly with price. Comprehensive numerical experiments have been conducted to confirm the effectiveness of the proposed algorithm.
Read the full abstract
High Energy Astrophysical Phenomena
The DSA/Chronoscope fast radio burst survey: forecasts and science overview
Fast radio bursts (FRBs) are bright extragalactic transients with several mysteries surrounding their origins. Large FRB samples enable accurate measurements of the cosmic matter distribution, in particular on scales $\lesssim 10$ Mpc. These measurements will impact cosmological inference and our understanding of astrophysical feedback, from the circumgalactic medium to cluster scales. Here we forecast the expected yields, redshifts, and host galaxies of FRBs as observed by the Deep Synoptic Array (DSA), and describe the key science cases enabled by the large FRB sample. The DSA will be an interferometer consisting of 1650$\times$6.15 m antennas, operating between 0.7–2 GHz, to be located in Nevada, USA. The Chronoscope backend on the DSA, hereafter DSA/Chronoscope, is designed to search for FRBs across the field of view in real time, enabling the storage of full-polarization voltage data. Extrapolating from existing FRB surveys, we expect roughly $10^4$ FRB detections per year in each of three search sub-bands. Combining across sub-bands, the survey could produce $\sim$ 10$^5$ FRBs over the nominal 5-year DSA survey, assuming Euclidean source counts and a baseline compute backend that can search $6\times10^6$ beams at 1 ms sampling. The well-characterized DSA synthesized beam and deep simultaneous reference images will enable localization precisions of $\lesssim$ 250 milliarcseconds. Key science cases include the use of FRB propagation effects in probing cosmic baryons, and studies of the FRB phenomenon using FRB host galaxies and their local environment, as well as multiwavelength counterparts.
Read the full abstract
Number Theory
Subconvexity of Short $k$-Free Exponential Sums in $\mathbb{F}_q[t]$
We extend recent work of the author over $\mathbb{Z}$ into the positive characteristic setting of $\mathbb{F}_q[t]$. In particular, for a polynomial $F \in \mathbb{F}_q[t]$ of degree $N$, let $R_k(\alpha)$ denote the exponential sum over $k$-free polynomials $f$ with $\text{deg}(f-F)<K$. For all $s>0$, we prove essentially tight upper and lower bounds for the $s$-th moment of $R_k(\alpha)$ whenever $K > (\frac{1}{2}+\epsilon)N$, and in even shorter intervals when $s>1+\frac{1}{k}$. As an application of these results, we prove a lower bound of order $q^{\frac{K}{6}}$ for the $L^1$-mean of the Möbius-twisted exponential sum over $\mathbb{F}_q[t]$ whenever $K >(\frac{1}{2}+\epsilon)N$.