Representation Theory
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Saturday 7 November
For each partition $\lambda$ of $n$ we construct an explicit basis of the unipotent Specht module $S^\lambda$ of $\mathrm{GL}_n(\mathbb{F}_q)$, defined recursively over horizontal strips. Each basis vector is $g\,x_\tau$, where $\tau$ is a semistandard tableau of shape $\lambda$, $x_\tau$ is obtained from a Steinberg-type vector by alternately applying a permutation matrix and a Fourier sum over a unitriangular group, and $g$ runs over an explicit set of coset representatives. Grouping the basis by the standardization of $\tau$ gives exactly $q^{\mathrm{maj}(s^T)}$ vectors for each standard tableau $s$, which is the numerical form of the Dipper-James standard basis problem. The same family is a basis over every field containing a nontrivial $p$th root of unity, including fields whose characteristic divides $|\mathrm{GL}_n(\mathbb{F}_q)|$, and a subfamily chosen layer by layer gives a basis of the simple quotient $D^\lambda$. We also give a basis of the $U$-fixed vectors of $S^\lambda$ and computed values of $\dim D^\lambda$ for $\mathrm{GL}_n(2)$.
Optimization and Control
A Low-Rank Augmented Lagrangian Frank-Wolfe-Based Method for SDPs with Unbounded Feasible Regions
In contrast to HALLaR (Monteiro et al. 2026), which is designed for solving problems with a trace-bounding constraint, this work develops a new augmented Lagrangian (AL) method, HALLaR+, for solving large-scale semidefinite programming problems without such constraint. Even though a trace constraint which preserves an optimal solution can always be added, choosing its bound is challenging: a value that is too small may exclude all optimal solutions, whereas one that is too large may substantially increase the computational effort. HALLaR+ is also a low-rank inexact AL method but, in contrast to HALLaR, it doubles the penalty parameter at every AL iteration and its AL subproblems have unbounded feasible regions that are approximately solved by a relaxed unbounded Frank–Wolfe (RUFW) variant. Its analysis relies on a uniform bound on the diameters of the sublevel sets of all generated AL subproblems instead of a bounded feasible set. Assuming that a primal–dual KKT pair exists and the primal optimal set is nonempty and bounded, it is shown that HALLaR+ returns a $\varepsilon$-primal-dual feasible solution in $\tilde {\mathcal O}(\varepsilon^{-2})$ total RUFW iterations. Numerical results suggest that HALLaR+ has very promising competitive performance against current state-of-the-art solvers.
Read the full abstract
Computer Vision
STAR-VF: Spatio-Temporal Aware Representation Condensation for Data-Efficient Video Fusion
Video fusion aims to integrate complementary cues from heterogeneous sources, producing visually informative videos with coherent spatio-temporal structures. Jointly learning cross-modal complementarity and temporal dependencies typically relies on large-scale, high-quality video datasets, whose storage and repeated use during optimization incur substantial costs, making data-efficient learning highly desirable. To address this challenge, we rethink video fusion from the perspective of dataset condensation and propose STAR-VF, the first framework that condenses spatio-temporal representations for data-efficient video fusion. STAR-VF condenses the original training set into a compact set of learnable spatio-temporal representations, which are decoded into synthetic video clips and optimized using feature-level supervision provided by a frozen proxy model. The resulting condensed data preserves fusion-relevant spatial and temporal cues without requiring ground-truth fused videos as supervision. To adapt fusion learning to condensed representations, we equip the final fusion model with flow-free inter-frame temporal modeling and query-driven intra-frame aggregation, enabling direct temporal interaction and adaptive selection of fusion-relevant information in representation space. Extensive experiments on infrared-visible, multi-exposure, multi-focus, and medical video fusion benchmarks demonstrate that STAR-VF, trained with only approximately 1/16 of the original temporal training set, achieves fusion quality comparable to methods trained on the full data. Further analyses validate the contributions of learned condensation and flow-free temporal modeling, and demonstrate the robustness of the condensed supervision across different proxy encoders and downstream fusion models. Project page: https://zju-xyc.github.io/STAR-VF-Project-Page/.
Read the full abstract
Databases
Extending DBLog-Based Incremental Snapshots to Handle Orphan Rows in CDC Pipelines
Applications use heterogeneous datastores where a source feeds sinks such as search indexes or data warehouses via change data capture (CDC). As transaction logs have limited retention, CDC alone cannot bootstrap sinks, and traditional approaches either stall source writes or hold back log events. DBLog-based incremental snapshots, adopted by Debezium and Flink CDC, read tables in chunks interleaved with the log without blocking writes or long log stalls and resume where they stopped. After the initial bootstrap, a sink can drift when losing change events or receiving unintended writes. Even a repeated bootstrap can leave orphan rows that exist in the sink and not at the source. A common remedy is to populate a new empty sink, but repairing an existing sink avoids parallel pipelines, a switch and serving traffic from a drifted sink until the new sink is ready. In DBLog's original design, each chunk only emits the latest state of the rows the table still contains, so it does not handle orphan rows. We extend DBLog to delete them using what a chunk read already knows, namely the keys that the range select did not return, emitting a new kind of event that sinks apply as deletes. We also add a repair request that an external process calls with orphan candidates, which the connector checks with its own read, emitting these events for keys confirmed absent. This enables three modes. In the fully integrated mode, every chunk pass removes the orphans it finds during full or periodic repairs. In the second mode, an external drift process hands suspected orphans to the connector. The third mode combines both and is what we recommend. Together, they turn DBLog into a complete data synchronization engine whether a sink catches up on live changes or missing rows, holds wrong column values, or contains orphan rows. Their safety and convergence are formally verified in Isabelle/HOL and Lean 4.
Read the full abstract
Optimization and Control
Gradient Estimation for LLM-Embedded Simulation Optimization
Large language models (LLMs) are increasingly embedded in operational simulators to represent context-dependent human decisions, which opens new possibilities for simulation optimization. However, the LLM acts as a black box, so derivatives through its decisions are unavailable for gradient-based policy search. Finite difference estimation can avoid this requirement, but the required paired simulations that sample LLM actions independently often disagree, and such action switches persist under small perturbations, which inflates the gradient variance to order $c^{-2}$ for perturbation radius $c$. To control this variance, we develop a behavioral common random number framework. It first applies maximal coupling to the LLM's action distributions, preserving each simulation's behavioral law while making action switches as rare as possible, which reduces the variance bound to $O(c^{-1})$. It then applies first-switch conditioning, which averages over whether a switch occurs, together with an impact-aware extension that prioritizes influential switches. Under suitable stability conditions, both keep the variance bounded as $c$ shrinks. We further establish improved complexity bounds for projected stochastic approximation. Experiments on synthesized and LLM-embedded simulators confirm the theory and show substantial performance gains.
Read the full abstract
Human-Computer Interaction
CompAIssionate Other: Identity Assumptions in LLM Responses to Distress
Individuals experiencing suicidal ideation (SI) or psychological distress may disclose their social identities when seeking support from LLMs. Attending to identity-related stressors can be important for providing contextually appropriate and empathetic support. However, disclosing an identity does not necessarily mean that it is a cause of distress. In this work, we audit how LLMs interpret distress through disclosed identities. Using paired counterfactual interventions, we evaluate six LLMs across two SI and distress datasets, 16 identity and control conditions, and two prompting protocols. Our findings suggest that models can transform identity disclosures into explanatory narratives, attributing distress to presumed identity-related adversity. These assumptions frequently appear within otherwise empathetic responses, particularly when marginalized identities are disclosed. We contextualize our findings through the lens of compassionate othering to characterize these patterns as a form of harm in which empathetic responses impose identity-linked assumptions on individual suffering. These responses can misrepresent individuals' experiences by treating their identities as explanations for distress without supporting evidence. Our findings highlight the need to mitigate these harms by grounding interpretations of distress in individuals' disclosures and avoiding unsupported identity-linked assumptions.
Read the full abstract
Complex Variables
Holomorphic interpolation and the Levi problem
On a nonsingular locally Stein Riemann domain over a Stein space, we prove that holomorphic forms of top degree interpolate jets of any fixed order on every closed discrete set finite over the base. For unramified domains, this yields jet interpolation by holomorphic functions, answering a question of Colţoiu and Joiţa for nonsingular sources. More generally, we establish a twisted interpolation theorem over weakly pseudoconvex spaces admitting injective holomorphic immersions into projective space. We also give negative answers to the Levi problem over singular Stein spaces: the Coeuré-Loeb space yields a non-Stein locally Stein unramified Riemann domain over a normal Stein threefold, and a related construction gives a relatively compact, non-Stein locally Stein open subset of a normal Stein fourfold. Finally, the punctured anticanonical bundle of Fornaess' ramified surface is a holomorphically separable increasing union of Stein manifolds that is not Stein, answering a question of Vigna Suria.
Read the full abstract
Artificial Intelligence
WARP-JEPA: World-Model-in-the-Loop Amortized Rollout Planning for Frozen Hierarchical Joint-Embedding Predictive Architectures
Hierarchical Joint-Embedding Predictive Architectures (H-JEPA) enable long-horizon visual control but rely on costly test-time gradient descent for planning. Our analysis reveals that iterative optimization consumes up to 97.7% of episode runtime and that low predicted rollout costs do not necessarily translate into successful control. We further identify a planning-horizon asymmetry between long-horizon top-level composition and short two-step subordinate planning. To address these challenges, we propose WARP-JEPA, a hierarchical amortized planning framework that replaces online gradient optimization with learned multi-candidate trajectory proposers while keeping the world models frozen. WARP-JEPA combines a lightweight world-model-in-the-loop causal Transformer at the top level with parallel residual MLPs at subordinate levels. The proposers are trained using model-based rollout costs and best-of-K behavior cloning, enabling efficient candidate selection without test-time gradient optimization. Across 30 frozen H-JEPA models on four benchmarks, WARP-JEPA achieves 9.6-18.8x faster episode inference on long-horizon tasks (2.9-3.0x on Cube) and improves task success by up to 33.3 percentage points. These results demonstrate that amortized planning can substantially improve both inference efficiency and control performance without retraining the world models. Code is available at https://github.com/hideinsewer/WARP-JEPA.
Read the full abstract
Category Theory
Abelian categories and double décalage
We show that an additive category is abelian if and only if its functorial nerve is a double décalage of a pseudo-simplicial object in Cat. In this characterization, the canonical isomorphism between the image and coimage is encoded by a coherent pseudo-simplicial identity.
Distributed Computing
DICOMPUTE: An Economic and Privacy-Aware Framework for Distributed Artificial Intelligence Inference
The prevailing design of large-scale artificial intelligence inference concentrates accelerators, model weights, and request processing in centralized computing facilities. Centralization offers important benefits in accelerator utilization, high-bandwidth interconnects, operational control, and serving efficiency. Yet centralized infrastructure is not always the least-cost or most governance-compatible choice for every inference workload. This paper develops a research framework for DICOMPUTE, a heterogeneous distributed inference architecture in which policy-constrained scheduling can allocate suitable requests across privately controlled, regional, and independently operated computing resources. We derive a full-cost comparison that accounts for capital amortization, energy, network transfer, orchestration, utilization, failures, and trust-related expenses; characterize conditions under which distributed inference can outperform centralized alternatives; and specify a privacy model separating data residency, transport confidentiality, execution confidentiality, and verifiable accounting. We propose falsifiable hypotheses and a reproducible experimental design, without presenting simulated examples as measured performance. Our central conclusion is conditional: distributed inference is most promising for divisible, latency-tolerant or locality-sensitive workloads on underutilized compatible hardware, while centralized clusters retain structural advantages for tightly coupled, memory-intensive, high-throughput inference. The framework is intended to support rigorous evaluation rather than to assert universal superiority.
Read the full abstract
Machine Learning (Statistics)
On Generalization in Bilevel Least Squares with Overparameterized Neural Networks
Bilevel optimization addresses machine learning problems in which the outer objective is a value function implicitly defined through the solution of an inner problem. While its algorithmic aspects are well developed, its generalization properties remain much less understood. In this paper, we study the generalization of inexact gradient methods when both objectives are least-squares regression problems and when the estimator of the value function's gradient, or hypergradient, is obtained by implicit differentiation in function space. This approach allows us to derive stationarity bounds for these methods in terms of the uniform generalization errors of the two regression functions needed to estimate the hypergradient. When these regression functions are defined as overparameterized two-layer ReLU networks trained by gradient flow with early stopping, we obtain explicit finite-sample bounds on these generalization errors in the Neural Tangent Kernel (NTK) regime. Combining these with our general inexact gradient analysis shows the effect of sample size on the overall convergence of the bilevel method. Numerical experiments on synthetic data illustrate our results. Beyond the NTK regime, further experiments reveal an initialization-dependent implicit bias induced by the bilevel structure, which can lead to feature learning at the inner level and overall performance improvements.
Read the full abstract
Computation and Language
Evaluating Jev on Arabic: Comparing Decision Models and LLMs Across Arabic Tasks and Dialects
How does Jev compare with other systems on Arabic tasks, and how does its performance change between Modern Standard Arabic (MSA) and dialects? We evaluate 67 systems, including 47 decision models and 20 large language models (LLMs), on 43 tasks across 9 task families using a shared zero-shot protocol. The Arabic-focused frozen suite includes English controls in three parallel resources, which we also remove in a separate descriptive comparison. Jev 1.14.0 achieves a mean task-primary metric of 69.24 on a scale multiplied by 100, compared with 68.77 for Jev 1.13.0 and 71.81 for Gemma-4-31B, the highest-scoring system on this aggregate. Removing the recorded English controls, repeated cross-task items or both leaves the full system order unchanged. Relative performance and gains between Jev versions vary across task families. In content-matched comparisons within Jev 1.14.0, pooled dialect-minus-MSA accuracy gaps range from -7.94 to -0.35 percentage points across four datasets; dialect-specific results show larger variation, and SIB-200's source-clustered interval includes zero. Paired contrasts with Jev 1.13.0, Gemma-4-31B and Jais-2-8B show a larger Palestinian banking penalty for Jev 1.14.0 than for either LLM under simultaneous uncertainty; the other gap contrasts remain inconclusive. These descriptive findings show why aggregate scores should be read alongside task-family results, coverage and matched language comparisons. They do not establish a general model ranking for dialect robustness or explain the causes of the observed gaps.
Read the full abstract
Quantum Physics
Trained, Frozen, or Absent: Identifying What a Quantum Layer Contributes to Cross-Sectional Equity Ranking
Hybrid quantumclassical models are increasingly proposed for nancial prediction, but reported gains are rarely separated from the classical components around them. We study a hybrid quantum temporal convolutional network for daily cross-sectional ranking of 2000 Tokyo Stock Exchange equities under a protocol built to make the quantum contribution identiable: six conditions (circuit trained, frozen or absent, a parameter-matched classical surrogate, and the circuit alone, trained or frozen) through one code path; four walk-forward folds with ve seeds; bootstrap intervals, Deated Sharpe Ratios, backtest-overtting probabilities and family-wise error control; ve classical baselines including a gradient-boosted tree; and a locked holdout. On 500 out-of-sample days two one-sided tests place the trained-versus-frozen dierence inside 0.03 annualised Sharpe (p = 0.0055, minimum detectable eect 0.02): the circuit is equivalent to a frozen one, not merely indistinguishable from it. With the circuit as the head's only input the two are equivalent inside 0.25; against the network with no circuit the design resolves only 0.50. No condition is distinguishable from zero (best De-ated Sharpe Ratio 0.17). The holdout reproduces the ablation on a protable window, and 2000 US equities reproduce it where every condition loses money.
Read the full abstract
Computation and Language
Enhancing Mathematical Reasoning in Turkish Large Language Models with Search-Based Inference and Model Merging
We improve Turkish mathematical reasoning in Large Language Models by combining LoRA fine-tuning, weight-space model merging, and search-based inference. Using Turkish-Llama 8B-DPO with Turkish translations of MetaMath, NuminaMath, and TigerLab, we find data quality beats scale (Mix-30k best among fine-tuned; large MetaMath underperforms). Linear merging delivers the largest gains (GSM8K up to 0.63), while Ties is modest and Breadcrumbs/Slerp degrade. Search with MCTS and Tree of Thoughts improves mean LLM-judge scores for reasoning quality on a 0-100 scale. Limitations include machine translation, limited scale, and search stochasticity.
Read the full abstract
Classical Analysis and ODEs
Schauder exponential basis which is not Riesz
We show that there exists a Schauder basis $\{e^{i \lambda_n t}\}_{n \in \mathbb{N}}$ with real frequencies in $L^2(-\pi,\pi)$ which is not a Riesz basis.
Hardware Architecture
RTLREX: LLM-Guided Framework for Exact High-Level RTL Recovery
Hardware designs may be available only as gate-level netlists or input-output descriptions when the original register-transfer-level (RTL) design and documentation are not accessible. Low-level representations obscure word-level computation and its finite-width behavior. Recovering high-level RTL from such representations therefore helps engineers understand a circuit's functionality and enables RTL rewriting and resynthesis. A plausible function model is also insufficient unless the generated RTL is equivalent to the original circuit for all input combinations. In this paper, we present RTLREX, a large language model (LLM)-guided framework that coordinates domain-specific tools for function recovery, RTL generation, independent verification, and cost-guided rewriting. Independent verification establishes equivalence between the recovered RTL and the original circuit. The LLM uses tool feedback to select further analyses, revise candidate RTL, and propose syntax-cost reductions. Experiments on 30 benchmarks from Problem A of the ICCAD 2022 CAD Contest show that RTLREX recovers equivalent high-level RTL for all benchmarks, surpassing the state-of-the-art method WolFEx in recovery coverage. On benchmarks recovered by both methods, RTLREX reduces mean syntax cost by 20.5% relative to WolFEx. On complete truth-table benchmarks from the IWLS 2026 Contest, our framework reduces mean syntax cost by 95.7% relative to an LLM agent baseline on the benchmarks validated by both methods.
Read the full abstract
Computation and Language
Adapting Large Language Models via Multi-Task Learning for Mental Health Problem, Emotion, and Sentiment Detection
This study investigates sequential multi-task learning using Parameter-Efficient Fine-Tuning (PEFT), LoRA, and 4-bit quantization on Llama 3.1 8B Instruct for mental health problem detection, emotion detection, and sentiment analysis. Using 4,243 English social media posts, it compares two task sequences and demonstrates that task ordering significantly affects knowledge transfer and task interference. The MHP-first sequence achieves the best mental health problem detection (Micro-F1: 0.7624), while the Emotion-first sequence performs better in emotion (F1: 0.6052) and sentiment analysis (F1: 0.9459). A Joint-MTL baseline further highlights the challenges of simultaneous learning, including task interference and generative format collapse. Overall, the findings demonstrate the effectiveness of sequential PEFT for mental health NLP while showing that task order is critical to balancing performance across related tasks. The integration of LoRA and 4-bit quantization also reduces computational and memory requirements, supporting the adaptation of large language models in resource-constrained environments.
Read the full abstract
Information Theory
BP-based qLDPC Decoding under Long-range Correlated Noise: Trapping Configurations and Prior Mismatch
We study the robustness of two state-of-the-art BP-based qLDPC decoders, Relay-BP and beam search, designed for independent noise when applied to spatially correlated noise with unchanged single-mechanism error probabilities. We generate correlated errors with a hidden Gaussian Markov random field, which lets us control the strength and range of the correlation while keeping every marginal error probability exact. This setting complements correlation-aware decoding: it asks how existing decoders perform when their assumed joint error model is mismatched. Across multiple tested circuits and code instances that we simulated, we observe that correlations can change logical error rates by orders of magnitude, reduce the finite-size distance advantage, and alter the relative ranking of Relay-BP, beam search, and BP-OSD. On one of the two tested circuits, Relay-BP and beam search become worse than BP-OSD, and increasing relay memory or beam width does not compensate for the degradation. Configuration analysis indicates that correlation primarily increases the probability of trapping configurations that are known to challenge these decoders; at matched error weight, correlated errors still cause decoding failures several times more often than i.i.d. errors. Finally, supplying the decoders with the marginals conditioned on the hidden field recovers most of the lost performance in the tested settings identifying prior mismatch as a major contributor to performance degradation under correlation.
Read the full abstract
Atomic and Molecular Clusters
Spectroscopic method for determining the onset of clustering in supersonic gas jets
This article presents an original spectroscopic method for determining the parameters of the initial stage of clustering in supersonic jets. The method was tested on supersonic jets of argon and nitrogen, excited by electrons at a distance of 30 millimetres from the nozzle exit, where the influence of secondary processes on the intensity of the observed radiation is significantly suppressed. For argon, the continuum emission from neutral excited clusters was used as an indicator of cluster formation in the supersonic argon jet. In the case of nitrogen, for which no emissions attributable exclusively to clusters have been recorded, a study was carried out on the effect of clustering on the redistribution of the intensities of atomic lines and molecular bands in the Lyman Birge Hopfield system.
Read the full abstract
General Relativity and Quantum Cosmology
Variational Approach to the Mathisson-Papapetrou-Dixon Equations of Motion for Extended Objects in General Relativity
The exact motion of a finite size object in curved spacetime is captured by the Mathisson-Papapetrou-Dixon equations. They associate a unique worldline, momentum, and spin to that object and determine the evolution of such in terms of its fully independent multipole moments. We provide a thorough review of this theory and show how to build it directly via a manifestly gauge invariant variational principle well-suited for effective descriptions of relativistic compact objects. We discuss conditions under which the covariant spin supplementary condition can be enforced and identify the natural basis of curvature tensors conjugate to the object's multipole moments.
Read the full abstract
Quantum Physics
Trainability-Simulability Tradeoff in Parameterized IQP-based Reinforcement Learning Policies
Parameterized instantaneous quantum polynomial-time (IQP) circuits generate distributions conjectured to be hard to sample classically. As reinforcement learning (RL) policies, their $2^n$ measurement outcomes are typically mapped to a smaller action space by a classical decoder. This aggregation can mitigate initialization concentration while making the policy classically simulable. We characterize this tradeoff through the decoder's Fourier–Pauli expansion. Under independent uniform initialization, we express policy variance as an exact sum over groups of nonconstant Pauli components with identical active-generator signatures. Each group contributes its squared aggregate coefficient times $2^{-r}$, where $r$ is the binary rank of its active generators. An inverse-polynomial aggregate coefficient with logarithmic rank prevents exponential policy concentration. Sparse Pauli expansions and contiguous action bins permit efficient classical additive estimation of action probabilities. We introduce a majority decoder combining protected low-weight components with exponentially large exact Fourier support, preserving initialization nonconcentration while excluding polynomial-size enumeration of the exact expansion. Numerical experiments corroborate the variance laws. In ten-qubit CartPole experiments, majority achieves higher returns than recursive parity and the tested local readouts, while higher-weight Pauli sectors contribute an increasing relative share of gradient magnitude during training. A complementary 21-qubit block-chain policy outperforms an extensive pool of fully connected classical neural networks in episode efficiency, even though its bounded-width structure permits efficient classical simulation. These results guide decoder design for quantum and quantum-inspired RL, distinguishing policy nonconcentration, learning performance, and classical simulability.
Read the full abstract
Machine Learning
Zeroth-Order Riemannian Optimization on Fixed-Rank Manifolds for LLM Fine-Tuning
Zeroth-order (ZO) optimization estimates descent directions from function values alone, which removes the activation storage required by backpropagation and makes it attractive for memory-efficient fine-tuning of large language models (LLMs). Existing low-rank ZO methods either parameterize the additive update through Euclidean factors with a non-unique representation, or restrict each finite-difference direction to be low-rank. In this work, we propose to constrain each additive update block to a fixed-rank matrix manifold while keeping the pretrained weights frozen, so the learned update remains rank-constrained throughout fine-tuning. On this geometry, we build a Riemannian ZO gradient estimator that projects a random matrix onto the tangent space, normalizes it to correct the scale mismatch induced by projection, and further propose Zeroth-order Riemannian Gradient Descent (ZO-RGD) and its smoothed variant, ZO-RGD-S. We show that the ambient finite difference matches its retracted counterpart, the normalized estimator recovers the Riemannian gradient in expectation, and ZO-RGD converges under trajectory-regularity conditions. Experiments on LLM fine-tuning tasks show that the proposed algorithms are competitive with strong forward-only baselines across model families and tasks, with gains in several settings.
Read the full abstract
Game Theory
Winning Reviews, Losing Discovery? The Co-Evolutionary Game of AI Research and Peer Review
Autonomous research systems can produce convincing manuscripts whose scientific value is difficult to verify, while automated reviewers can reward the appearance of completeness. We examine how this gap can become a self-reinforcing threat to the research community. In a two-population evolutionary game, authors choose between acceptance-oriented and exploratory research, and reviewers choose between proxy evaluation and careful assessment. Cheaper acceptance-oriented production can increase review pressure, making proxy assessment more attractive; proxy assessment can in turn strengthen the incentive to produce reviewer-compatible work. We derive conditions under which this feedback sustains a stable regime that rewards publication success while reducing the scientific value of accepted work. An analytical basin boundary explains why the same institutional rules can support either this regime or exploratory research with careful review. Numerical illustrations show how publication counts can conceal scientific loss, how faster adjustment can reinforce existing practices, and why partial reform may fail once the cycle is established. Careful audits and reviewer rewards can reverse the feedback, while temporary support can produce lasting recovery if it moves the community across the basin boundary. The model identifies a warning mechanism rather than forecasting a particular conference: a review system can remain operational and privately rewarding while becoming less effective at supporting discovery. Its institutional implication is to reward independently validated contributions and consequential checking, rather than accelerate paper production and scoring alone.
Read the full abstract
Number Theory
On the lower bound for the $L^1$ norm of Liouville exponential sums. II
Let $\lambda$ denote the Liouville function, and put \[ I_\lambda(X)=\int_0^1\left|\sum_{n\le X}\lambda(n)e^{2\pi i n\alpha}\right|\,d\alpha. \] In this paper, we prove that \[ \limsup_{X \to \infty} \frac{\log I_{\lambda}(X)}{\log X} \ge \frac{1}{3}. \]
Read the full abstract
Machine Learning
Generating Diverse and Verified Synthetic Clinical Notes with Labels by Construction
Under strict privacy regulations in the clinical domain, de-identification is necessary to enable the use of clinical data such as for natural language processing. However, we currently lack sophisticated ways to tell which de-identification models actually are suitable. Specifically, we need benchmarks that are challenging, diverse, and clinically plausible, all while ensuring label correctness. To this end, we present SPARCLING, an LLM-based pipeline to generate synthetic discharge letters with character-exact PHI span annotations. Our pipeline combines structured attribute sampling, template generation, and red teaming to obtain diverse PHI-labeled discharge letters. In our experiments, we demonstrate this superior content diversity over existing public datasets. Further, we show SPARCLING's downstream utility as a benchmark to uncover failure modes of de-identifiers in the German clinical domain. Finally, our experiments show that training on our synthetic data can yield more capable de-identifiers than training on the scarce public data alone. Our code is available at https://github.com/Matt00n/phi_discharge_letter_generator.
Read the full abstract
Rings and Algebras
Linear structure graphs associated with commuting graphs of Lie algebras
The commuting graph (CG) of a finite-dimensional Lie algebra over a field of characteristic zero forgets the linear structure, and a non-trivial center creates redundant vertices. We introduce linear structure graphs (LSGs), whose vertices are the nonzero vectors of a vector space and whose edge set is stable under linear operations. Each vertex generates a finite ascending sequence of subspaces, which defines its LSG-height; this height governs connectivity. We characterize LSGs of height at most 1 and show that the CG of a Lie algebra with trivial center has height at most 1 if and only if the algebra is a commutative transitive Lie algebra (CTLA). The reduced commuting graph (RCG) restricts the CG to a complement of the center; it is well defined up to LSG-isomorphism, preserves the connectivity, diameter and number of components of the CG, and is compatible with direct sums. Finally, we characterize almost abelian Lie algebras (AALAs) by their RCGs.
Read the full abstract
Machine Learning (Statistics)
Trustworthy Recursive Self-Improvement with Statistical Guarantees
Recursive self-improvement offers a promising path toward AI agents that improve through experience by refining their own harnesses without retraining their underlying large language models (LLMs). Making this self-improvement process reliable, however, requires more than generating new updates: agents must decide which changes are worth keeping. A greedy gate that accepts updates based solely on higher validation scores can mistake an observed gain for genuine progress and lead to performance degeneration, whereas overly conservative statistical gates may reject beneficial updates and hinder performance improvement. We introduce Trustworthy Recursive Updates with Statistical Testing (TRUST), a framework that adapts an online multiple hypothesis testing procedure to control the rate of false evolution across rounds, with an LLM analyzing historical context to help determine how much evidence is required to update the incumbent. Under interpretable assumptions, we prove that TRUST limits the proportion of accepted updates that result in unacceptable performance loss. We further derive a lower bound on the final performance gain. Across the benchmarks HotpotQA, DROP, and APPS with the models DeepSeek-V4-Flash and Qwen-3.8-Flash, TRUST reduces the average ratio of accepted inferior updates from the greedy gate's 45.59% to 2.78%, while achieving higher final scores than the state-of-the-art statistical gate baselines for recursive self-improvement. These findings demonstrate that TRUST strikes a balance between statistical reliability and performance gain in recursive self-improvement.
Read the full abstract
Plasma Physics
Exact analytic anisotropic three-dimensional equilibria with nested magnetic surfaces, confined drift orbits and without spatial symmetry
Exact toroidal fluid and kinetic equilibria with nested magnetic surfaces and no continuous symmetry are of great interest in laboratory, astrophysical and space plasmas. Their existence and stability have been open in physics and mathematics for decades. We construct such anisotropic-pressure equilibria by applying the Clemente transformation to exact vacuum fields. They connect smoothly to these fields on a fixed regular domain and preserve a regular magnetic axis, perfectly nested surfaces and spatially varying rotational transform. To our knowledge, the sheared configurations are the first exact anisotropic equilibria combining these properties with positive pressures and a positive firehose margin. The fixed-boundary Chew–Goldberger–Low (CGL) quadratic energy splits into a nonnegative vacuum-field term and a local pressure term, nonnegative for p_perp/p_parallel <= (3+\sqrt{21})/2 ~ 3.79. For p_parallel >= p_perp each configuration minimizes the finite-amplitude CGL energy over smooth boundary-fixed ideal material deformations. For the current-free configurations, physical guiding-center distributions produce these pressures within the same bound, with no free energy for magnetic-moment-conserving electrostatic perturbations. Trapped particles inherit the vacuum field's second adiabatic invariant at a rescaled magnetic moment, so local closed action contours carry over. The equilibrium code DESC reproduces them from the exact solution. Kinetic stability for surface-dependent rescalings and against general nonlinear electromagnetic perturbations remains open.
Read the full abstract
Machine Learning
Proximal Preference Optimization: A Unified Framework for Stable and Reward-Free Alignment
We introduce a proximal preference optimization framework that unifies the implicit reward formulation of Direct Preference Optimization (DPO) with the trust-region stability of Proximal Policy Optimization (PPO). Our framework yields two complementary algorithms. P$^2$O (Proximal Preference Optimization) introduces a clipped importance ratio directly into the DPO preference margin, producing a reward-model-free algorithm that supports safe dataset reuse, bounded policy divergence, and monotonic preference improvement guarantees. PKTO (Proximal Kahneman-Tversky Optimization) extends P$^2$O by replacing the DPO margin with KTO's prospect-theoretic utility margins, adding a self-regulating adaptive baseline that tightens the effective trust region as the policy drifts, and an asymmetric weighting that handles imbalanced or asymmetric preference datasets. We provide a unified theoretical analysis, including Kullback Leibler (KL) trust-region bounds, a policy-improvement lower bound, and convergence guarantees under multi-epoch proximal updates. Empirically, P$^2$O and PKTO match or exceed DPO and KTO in alignment quality while maintaining better stability under repeated dataset reuse across multiple LLMs on three preference benchmarks. The code base is available here.
Read the full abstract
Analysis of PDEs
A sufficient condition for the Liouville property of solutions to the Euler-Lagrange equations of variational integrals with nonstandard growth in terms of the ellipticity modulus near the origin
We consider entire solutions $u$: $\mathbb{R}^{n}\rightarrow\mathbb{R}$ of the Euler-Lagrange equation $\operatorname{div}(\nabla f(\nabla u))=0$ with convex density $f$: $\mathbb{R}^{n}\rightarrow\mathbb{R}$ of nonstandard growth and nonuniform ellipticity. It turns out that the Liouville property can be characterized through the behavior of $D^{2}f(p)$ for points $p\in\mathbb{R}^{n}$ close to the origin: if, for example, $D^{2}f(p)$ behaves as $|p|^{\rho-2}\mathbf{1}$, $\mathbf{1}=\text{unit matrix}$, with exponent $\rho\ge2$ and for $|p|\le1$ then the validity of $n\le\rho+2$ together with the boundedness of $u$ implies its constancy.