Peer Revenue Play credits. Real papers.

Papers

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

All of arXiv

Tuesday 8 September

Machine Learning

SIM: Subspace Interaction-based Method for Token-Level Text Anomaly Detection

Token-level text anomaly detection, as an emerging trend of text anomaly detection, moves beyond coarse-grained document-level detection by localizing anomalous tokens within text. By providing fine-grained abnormality prediction, token-level text anomaly detection plays a critical role in various real-world applications, such as spam filtering and fake news detection. However, existing methods still rely on the global distance calculation for scoring, during which the local anomaly signals are severely diluted by numerous redundant normal feature dimensions. Moreover, pre-trained language models used in these methods inevitably smooth out surface anomalies, further limiting their effectiveness in token-level anomaly detection. To address these limitations, we propose a Subspace Interaction-based Method (SIM for short) for token-level text anomaly detection. To prevent local signal dilution, SIM adopts a subspace interaction-based anomaly detector, which decouples high-dimensional token embeddings into multiple low-dimensional ones, amplifying localized anomaly signals hidden within specific dimensions. To counteract the over-smoothing effect, we design a hard pseudo-anomaly generation module to construct pseudo-anomalous tokens, simulating the subtle anomalies obscured by semantic smoothing. Also, a probabilistic boundary loss is developed to standardize anomaly scores into statistical distances, effectively enforcing anomalous instances to deviate significantly from the normal distribution center. Extensive experiments on multiple benchmark datasets verify the effectiveness of SIM and demonstrate its remarkable efficiency, robustness, and interpretability. The source code is available at: https://github.com/yankehan/SIM-TAD.
Read the full abstract

All authors Kehan Yan, Yue Tan, Qingfeng Chen, Shiyuan Li, Yu Zheng, Yixin Liu

Networking

QoS-Aware RACH Preamble Slicing via Quota-Projected Branching Deep Reinforcement Learning

Quality-of-service (QoS)-aware random access requires adaptive allocation of a finite random access channel (RACH) preamble budget across heterogeneous traffic and access procedures. This paper proposes QP-BD3QN-RACH, a quota-projected branching deep reinforcement learning controller for mixed two-step (2RA) and four-step (4RA) contention-based random access. Four action branches correspond to the delay-sensitive and delay-tolerant 2RA/4RA preamble pools. A branching dueling Double DQN selects pool-specific multipliers, and deterministic quota projection converts them to nonnegative integer allocations that preserve the preamble budget. With five actions per branch, the controller represents 625 pre-projection branch-action tuples using 20 branch-action outputs. Evaluation covers five arrival loads, cross-method comparison under nominal seed 42, six-seed sensitivity of QP-BD3QN-RACH, and targeted ablations. Across the five-load grid, its mean direction-aligned differences relative to four comparators are positive: 5.74 to 8.21 percentage points for success/collision, 1.23 to 1.92 percentage points for fallback, 0.35 to 0.68 percentage points for blocking, and 0.128 to 0.456 decision intervals for successful-access delay. Load-wise results exhibit metric-dependent tradeoffs, particularly under intermediate and overload conditions.
Read the full abstract

All authors Jiulin Guo, Jiahan Xu, Jiashuo Zhang, Heng Yang, Yizhen Sun, Yutong Xie, Shanshan Li, Zhenyu Liu, Lei Zhang

General Economics

Self-Fulfilling Prophecies, Quasi Nonergodicity, and Wealth Inequality: A Comment

Bouchaud and Farmer (2023) argue that self-fulfilling beliefs generate a realistic wealth distribution. Their reported Gini coefficient, quantiles, and Pareto exponent are computed from dynasty-level wealth averaged over 250 dates, not from a cross section. Correcting the estimand strengthens their qualitative result: with $N=1{,}000{,}000$ dynasties, the terminal cross-sectional Gini is 0.883 rather than 0.7, and median wealth is 0.0052 rather than 0.39 of average wealth. The model therefore generates more inequality, but not the reported quantitative fit. Appendix F analyzes a different process and does not establish the claimed Pareto tail or exponent.
Read the full abstract

Materials Science

Pressure Evolution of Atomic Volume Systematics in Transition Metals

We investigated the evolution of the well-known parabolic dependence of atomic volume on atomic number in transition metals under extreme compression at pressures up to 400 GPa using density functional theory calculations. Our results reveal that the ambient-pressure parabolic trend transforms into a characteristic cubic-like behavior at high pressures. This evolution is attributed to the higher compressibility of bcc transition metals associated with comparatively large increases in the total energy. The present findings are discussed in relation to previous experimental observations and first-principles calculations.
Read the full abstract

Artificial Intelligence

Qiushi Engine on AstaBench E2E-Bench-Hard

This report analyzes Qiushi Engine v0.8 across all 40 test tasks in AstaBench E2E-Bench-Hard, a benchmark that requires autonomous agents to carry a research question through experimental design, code implementation, actual execution, result analysis, and report delivery. Qiushi Engine is model-configurable; this evaluation selected DeepSeek deepseek-v4pro-preview as the model backend. The official AstaBench leaderboard records a score of 0.816 and an average benchmark cost of USD 15.209 per task, while the full-precision local recomputation is $81.59 \pm 1.87$. Four tasks satisfied every rubric item, yielding a full-task completion rate of 4/40 = 10% – 7 percentage points above, and about 3.3 times, the approximately 3% best rate reported for AstaBench's official agents. Across 507 required rubric items, 416 were satisfied (82.1%). Official scoring archives and 40 Meta-Trace records show sustained production and verification of reports, code, and experimental artifacts; the principal gaps lie in repeated runs, external dependencies, specified metrics, and ablation studies. The report explains the benchmark, system workflow, aggregate results, representative cases, and limits of interpretation.
Read the full abstract

All authors Wenhao Li, Shuxing Yang, Fujia Chen, Jincheng Mi, Yuang Pan, Rui Zhao, Zichen Li, Junyao Wu, Shenzhan Hong, Yaqi Li, Yize Wang, Kaihao Zhu, Taowen Deng, Junjie Yang, Hongsheng Chen, Yihao Yang

Probability

Geometric influence on the limiting behavior of diffusion processes in one-sided Brownian environments on disconnected fractal sets

We investigate diffusion processes on disconnected fractal sets in one-sided Brownian environments. On the real line, it is established that the process exhibits either diffusion or trapping, each occurring with probability $1/2$. In this study, we demonstrate that for fractal sets, this behavior is governed by the geometric parameters $r$ (the reciprocal of the similitude ratio) and $N$ (the number of contraction mappings), which define the fractal structure. Two distinct regimes emerge: a diffusive regime on the environment-free side and a localization regime on the side influenced by the environment. The transition between these regimes is determined by whether the random environment first hits the threshold $\log r$ or $-\log N$. Consequently, the probability of diffusion versus trapping is explicitly characterized by the Hausdorff dimension $d_f = \log N / \log r$. This result demonstrates that fractal geometry scales the limiting distributions and dictates the stochastic regime.
Read the full abstract

Combinatorics

Maximizing $K_r + I_r$ in graphs with fixed edge density

For every integer $r\ge4$, and $\rho \in [0,1]$, we asymptotically determine the maximum proportion of $r$-element sets of vertices that induce either a clique or an independent set in a large graph with density $\rho$. This generalizes a result of Olpp for $r=3$. After the initial idea for the main proof was found by the authors, various AI models were used to streamline the argument and perform the calculations necessary for completion of the proof.
Read the full abstract

All authors József Balogh, Andrzej Grzesik, Bernard Lidický, Theodore Molla, Dhruv Mubayi, Jan Volec

Analysis of PDEs

Fefferman–Stein type inequalities via area and maximal functions for Schrödinger operators with applications

In this paper, we establish a Fefferman–Stein inequality in terms of area function and non-tangential maximal function associated with the Schrödinger operator $\mathcal{L} = -\Delta + V$ on stratified Lie groups $\mathcal G$, where $\Delta$ denotes the sub-Laplacian on $\mathcal G$ and $V$ is a nonnegative locally integrable function. As an application, we extend this inequality to the tensor product $\mathcal G_1 \times \mathcal G_2$ of two stratified Lie groups and develop atomic decompositions associated with the Schrödinger operator for functions in the Orlicz space $L\log^{+}L(\mathcal G_1 \times \mathcal G_2).$ Using these atomic decompositions, we further prove weak-type endpoint estimates for the area integral operator and the Riesz transforms associated with the Schrödinger operator on $L\log^{+}L(\mathcal G_1 \times \mathcal G_2),$ thereby extending the celebrated result of R.\,Fefferman and E.M.\,Stein \cite{FSt1982} to the setting of singular integrals with non-smooth kernels.
Read the full abstract

All authors Ji Li, Wanjun Li, Liang Song, Lixin Yan

Optics

Integrated yoctosecond-precision timing detector

Precise timing detection is essential for exploring ultrafast phenomena in fields ranging from free-electron lasers to ultra-high-power laser facilities. However, achieving simultaneous high resolution and large dynamic range remains a fundamental challenge, and state of the art systems are often constrained by their physical size, power requirements, and limited scalability. Here we introduce an integrated dual electro optic sub cycle timing detector (DEST) that overcomes these limitations. In a proof of principle measurement, the device resolves timing jitter as small as 11 yoctoseconds (ys, $10^{-24}$ s) at 1 MHz–equivalent to the transit time of light across two protons–while maintaining an unambiguous measurement range of 6.15 ps and a dynamic range exceeding 155 dB. The core detection unit is miniaturized to chip scale dimensions of 18 mm x 2 mm x 1 mm on a thin film lithium niobate platform, ensuring inherent stability and immunity to environmental disturbances. Moreover, the architecture naturally lends itself to massive parallelization through array integration, with the potential to push timing precision to the sub 10 rontosecond (rs, $10^{-27}$ s) level within a 1 $m^2$ footprint. This combination of extreme sensitivity, wide dynamic range, compact size, and scalability opens new avenues for detecting previously inaccessible weak signals, including those from gravitational waves, quantum vacuum fluctuations, and beyond.
Read the full abstract

All authors Jie Yang, Tong Wang, Yulin Shen, Dehui Pan, Zhichao Chen, Ming-Yang Zheng, Bin Wang, Shaobo Fang, Yi Zhang, Ke Zhang, Jiahui Yao, Minghua Chen, Guorong Wu, Xueming Yang, Ming Xin

Formal Languages and Automata

An Elementary Proof of the $\widetilde O(n^{1/3})$ Bound for Separating Words

For two distinct binary words of length $n$, the separating words problem asks for a small deterministic finite automaton that accepts exactly one of them. Chase proved a $\widetilde O(n^{1/3})$ upper bound using a complex-analytic estimate for sparse polynomials. We replace that estimate by a finite-difference argument and a second-order real recurrence cutoff. The resulting elementary proof gives an explicit bound of $O(n^{1/3}(\log n)^{7/3})$ states.
Read the full abstract

Combinatorics

A Note on Sphere Packing Bounds for Tuple Lattice Sieving

A finite set of unit vectors is $k$-irreducible if every signed sum of between two and $k$ distinct elements has norm greater than one. Let $\mathcal{R}_k$ be the maximal asymptotic rate of such sets, and let $\kappa(\alpha)$ be the maximal asymptotic rate of spherical codes with pairwise inner products at most $\alpha$. For $k \ge 2$ we show: \begin{align} \mathcal{R}_k \le \min_{1 \le r \le \lfloor k/2 \rfloor} \frac{1}{r} \, \kappa\!\left(1 - \frac{1}{2r}\right) \, . \end{align} Combining this with standard sphere packing bounds, for large $k$ we obtain an almost-tight asymptotic comparison with the known lower bounds: \begin{align} \left(\tfrac{1}{2}-o(1)\right) \, \frac{\log_2 k}{k} \le \mathcal{R}_k \le (1 + o(1)) \, \frac{\log_2 k}{k} \, . \end{align}
Read the full abstract

Artificial Intelligence

Do Dynamic Routers Need Memory? HeRo: History-Aware Routing for Efficient LLM Inference

Dynamic layer routing reduces the inference cost of Large Language Models (LLMs) by learning to skip layers for individual tokens. Existing methods, however, treat each routing decision as a local operation conditioned solely on the current hidden state which is a formulation that overlooks the sequential, path-dependent nature of routing across depth: earlier decisions shape the representations seen by downstream routers, and the layer-usage objective couples all decisions jointly. We propose History-Aware Routing (HeRo), a dynamic routing framework that resolves this mismatch by introducing a router memory mechanism to maintain an explicit routing state across model depth. The memory is constructed via linear attention, incrementally aggregating preceding routing scores and their induced residual updates into a compact history representation. At each routed layer, the router conditions jointly on this accumulated state and the current hidden representation to select the executed branch. Instantiated for token-wise FFN routing, HeRo trains only lightweight routers and adapters on a frozen backbone, requiring no modification to pretrained parameters. Across Llama 3.1-8B, Llama 2-7B, and Llama 2-13B, HeRo consistently achieves the highest aggregate performance retention among ten baselines. On Llama 3.1-8B, it bypasses 26.87% of model parameters while achieving 100.24% of dense model performance across seven benchmarks, and retains 97.01% while bypassing 38.82% of model parameters under a tighter computation budget. Ablation studies confirm that removing routing history consistently degrades performance, most notably on multistep reasoning and code generation, validating that explicit routing memory enables more accurate and adaptive dynamic routing than solely conditioning on hidden state.
Read the full abstract

Artificial Intelligence

Bridging the Semantic-Utility Gap in Multimodal RAG via Generator-in-the-Loop Alignment

Vision-language models (VLMs) augmented with retrieval-augmented generation (RAG) benefit from access to external evidence. However, standard retrievers and rerankers optimize for semantic similarity rather than answer utility, creating a preference gap: documents that appear relevant may not help the generator produce a correct answer. Motivated by this, we propose a two-stage generator-in-the-loop alignment framework that closes this gap without human document-level relevance annotations. Our framework consists of two stages: in Stage 1, a VLM generates a hypothetical text passage from the image-query pair, which is used as the retrieval query for dense text search, bridging the image-to-text modality gap. In Stage 2, a cross-encoder reranker adapted with low-rank adaptation (LoRA) is fine-tuned using answer-supervised preference pairs mined from the frozen VLM: given the dataset answer label, a candidate document is labeled positive if the VLM produces the correct answer when given that document as context, and negative otherwise. This generator-guided signal is compatible with multiple alignment loss functions, including contrastive (triplet) loss, pairwise direct preference optimization (DPO), and supervised fine-tuning (SFT), and supports periodic re-mining to refresh preference pairs as the reranker improves. Experiments on VQA-X and A-OKVQA with Qwen3.5-2B and Qwen3-VL-4B-Instruct show that our proposed framework consistently outperforms rank-order, random, and REPLUG-style likelihood baselines under various alignment losses and pool size settings, suggesting that answer-level generator feedback is an effective supervision signal for preference alignment.
Read the full abstract

All authors Zhan-Lun Chang, Dong-Jun Han, Seyyedali Hosseinalipour, Mung Chiang, Christopher G. Brinton

Signal Processing

BeamFocusNet: Beamforming-based Explicit Spatial Signal Focusing for Robust DoA Estimation under Low-SNR and Single-Snapshot Conditions

DoA estimation plays a crucial role in signal processing. Inspired by beamforming, recent works employ neural networks to estimate filters for filtering received signals to achieve DoA estimation. These methods typically generate filters by implicitly focusing signals and suppressing noise. However, this couples the two objectives, making training difficult to balance and resulting in poor robustness, especially under low SNR and limited snapshots. To address this issue, we propose the BeamFocusNet method, which generates filters through explicit signal focusing, introducing a new paradigm for neural network-based filter generation. Extensive experiments under various challenging conditions demonstrate the superiority and robustness of the proposed method in DoA estimation. Code is available at https://github.com/colaudiolab/BeamFocusNet.
Read the full abstract

All authors Xuyao Deng, Qisheng Xu, Shuo Liu, Yong Dou, Wen Zhang, Kele Xu

Artificial Intelligence

Does Deeper Reasoning Compromise Alignment? Revealing and Mitigating of Alignment Collapse in Large Reasoning Models

The emergence of Chain-of-Thought (CoT) has established a robust foundation for Large Reasoning Models (LRMs). While deep reasoning is widely believed to enhance safety alignment, the stability of alignment mechanisms under extended reasoning remains underexplored. This paper challenges the prevailing view by revealing a critical vulnerability: Deep Reasoning May Induce Alignment Collapse. To rigorously quantify this phenomenon, we propose the Alignment Loss Rate (ALR) metric. Our experiments demonstrate that as reasoning depth increases, ALR rises significantly, indicating a severe degradation in model robustness against external perturbations. Capitalizing on this instability, a novel jailbreaking paradigm, Reasoning Trap (RT), is proposed. RT induces the model into extended reasoning to amplify the impact of adversarial attacks, leading to a sharp decline in safety capabilities. To elucidate the mechanism behind this collapse, we identify Attention Dilution as the root cause, arising from the competition for attention between the extended reasoning process and the original input. To mitigate this, Reasoning Residual Alignment (RRA), a lightweight defense strategy that dynamically re-emphasizes the input via residual connections integrated with the reasoning process.
Read the full abstract

All authors Yu-Hang Wu, Yu-Jie Xiong, Henghua Zhang, Bairui Zhang, Jia-Chen Zhang, Shaohua Li

Information Retrieval

Cassette: Case-to-Case Structural Distillation for Efficient Legal Case Retrieval

Legal case retrieval (LCR) is an essential tool for not only assisting legal practitioners to efficiently retrieve precedents but also enabling ordinary individuals to find valuable legal case information without relying on expensive professional legal services. Our previous work CaseLink demonstrated the effectiveness of using case to case graph structures to improve retrieval accuracy. However, its high computational cost during inference on large-scale legal databases limits its practical use in real-world settings. The main inefficiency comes from constructing test time graphs and computing pairwise term frequency similarities of cases. This process has O(n^2) complexity for n legal cases, making the runtime prohibitive as the number of candidates grows. For example, the retrieval time for one query on a database (COLIEE2022) with 1,563 candidate cases is more than 500 milliseconds, while the runtime would increase drastically to more than 3,500 seconds for a database (LeCaRDv2) with 55,192 candidate cases. To further enhance the retrieval performance while achieving a significant speed-up, in this extension paper, Cassette framework is proposed with a distillation strategy involving ranking objective and eigen-matching objective for an effective transfer of knowledge from a powerful and well-trained heavy teacher retriever to a lightweight and efficient hybrid student dual encoder. Specifically, the student query encoder is implemented as a multilayer perceptron model designed for fast online processing, whereas the student candidate encoder adopts a GNN architecture, suitable for an offline manner within the case database. Extensive experiments are conducted on three benchmark datasets, and the results verify the effectiveness of the ranking distillation while achieving high efficiency. The code has been released on https://github.com/yanran-tang/Cassette/.
Read the full abstract

All authors Yanran Tang, Ruihong Qiu, Hongzhi Yin, Xue Li, Zi Huang

Databases

UnespDataLens-RM: A Reference Model for Analytical Data Engineering with Governance, Quality, Provenance, and Reproducibility

The growing reliance on data in analytical processes and evidence-based decision-making has reinforced the importance of Data Engineering in building pipelines capable of integrating, transforming, validating, and delivering data from heterogeneous sources. However, the reliability of analytical assets depends not only on data processing capabilities but also on mechanisms for governance, quality assurance, provenance, traceability, versioning, and reproducibility throughout their lifecycle. These responsibilities are commonly addressed by different models, frameworks, and operational practices, resulting in methodological fragmentation across the analytical data lifecycle. To address this gap, this article proposes UnespDataLens-RM, a technology-independent reference model that integrates technical-operational processes and cross-cutting capabilities within a unified structure for Analytical Data Engineering. The model aims to support the specification, organization, and evolution of analytical pipelines by incorporating governance, quality, provenance, traceability, and reproducibility from the design stage. Developed following the Design Science Research approach, UnespDataLens-RM comprises eight technical-operational modules, eight cross-cutting modules, complementary dimensions, and a formalized set of artifacts, metrics, and validation criteria. The resulting specification offers a conceptual and methodological framework for future instantiations and empirical evaluations of analytical pipelines designed to be more governable, documented, traceable, auditable, and reproducible.
Read the full abstract

Computation and Language

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness

Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native models developed to explore this path through agentic post-training. Our system combines a heterogeneous model pool with intelligent routing, recording the predicted capability demand, selected service tier, and subsequent interaction for each user turn. These records are converted into training examples that preserve interleaved reasoning, tool calls, and harness context, and are admitted through structural validation, six-dimensional semantic evaluation, and subscene-level labeling. Routing signals organize supervised fine-tuning into a three-stage curriculum and extend to routing-guided on-policy distillation, where a teacher supervises student-generated responses under the same progression. Capability-guided allocation then converts evaluation feedback into the next training mixture, closing an evaluation-selection-update loop in which what the system learns to do shapes what it learns from next. Across eleven benchmarks covering harness-based agents, tool use, coding, and instruction following, post-training raises the macro-average from 58.94 to 64.87 at 4B and from 65.60 to 69.04 at 9B, substantially narrowing the aggregate gap between the post-trained 4B model and the 9B base model. We further adapt NeoHorse-1-4B into NeoHorse-Jev-4B, a prefill-only model that returns structured decisions and probabilities for agent workflows. NeoHorse-1 provides an initial prototype of this feedback-driven process and a path toward harness-mediated RSI across successive iterations.
Read the full abstract

All authors NeoHorse Team, Guoliang Cao, Guohao Dai, Tianyu Guo, Kai Han, Hailin Hu, Zihan Jiang, Xiang Kuang, Boxun Li, Yulong Li, Zehua Pei, Yuchuan Tian, Jiamin Wang, Yu Wang, Yunhe Wang, Yihong Wu, Haiyang Xu, Shuo Zhang, Hang Zhou, Siyang Cheng, Jiayu Fan, Wei He, Qingrui Jiao, Hongguang Li, Zhiyuan Li, Runke Liu, Xi Liu, Xinchen Liu, Sinno Jialin Pan, Yi Ren, Liuyang Song, Chenyu Wang, Bei Yu, Quanlu Zhang, Xiangyu Zhang, Mengyu Zheng, Yingjie Zong

Computer Vision

WSPolypNet: Weakly Supervised Polyp Localization in Colonoscopy Videos

Because dense frame-level annotation of colonoscopy videos is costly, we propose WSPolypNet, a weakly supervised framework for polyp localization using only video-level labels. WSPolypNet employs a 3D convolutional neural network trained with video-level supervision to generate class activation maps (CAMs), which identify candidate polyp regions without requiring frame-level spatial annotations. The CAM-derived localization cues are further enhanced using a multi-view strategy and provided to MedSAM2 as point prompts. MedSAM2 then propagates segmentation masks across the video, refining the coarse localization cues according to polyp boundaries. WSPolypNet achieved CorLoc scores of 47.80%, 43.68%, and 35.01% at IoU thresholds of 0.3, 0.5, and 0.7, respectively, compared with 36.87%, 33.72%, and 27.94% in the single-view setting. For small polyps, the multi-view strategy improved CorLoc@0.5 from 16.01% to 30.97%. The framework also achieved a recall of 94.51%. These results demonstrate the potential of weakly supervised spatiotemporal learning to substantially reduce spatial annotation requirements for polyp localization in colonoscopy videos.
Read the full abstract

All authors Giseong Hwang, Minjae Jo, Yeonghyeon Park, Kyeonghun Kim, Seoyeon Han, Donghoon Han, Haneul Kim, Yului Jeong, Insung Hwang, Pa Hong, Ken Ying-Kai Liao, Nam-Joon Kim

Optics

Topology of Photonic Time Crystals under Complex Refractive Index Modulation

Photonic time crystals, constructed by periodic modulation of electromagnetic parameters in time, hold unique topological properties. Here, we explore permittivity and conductivity co-modulated PTCs (PCM PTCs) by dynamically tuning permittivity and conductivity simultaneously of the spatially-uniform materials, and discover their distinctive non-Hermitian effects. We find that such PCM PTCs hold a global shift of the Floquet spectrum along the imaginary quasifrequency axis, which controls field amplification and decay in both momentum bands and band gaps. By further considering the temporal interface consisting of two different PCM PTCs, we can unveil the relationship between the emergence of topological edge states and momentum band inversion by analyzing the eigenstates at exceptional points. Moreover, the characteristic of the topology of PCM PTCs can also be captured from the complex Dirac mass. Our work studies the topology of PTCs in complex modulation regime, offering new opportunities for manipulating topological edge states in the time domain.
Read the full abstract

All authors Huili Lin, Zhaohui Dong, Xianfeng Chen, Luqi Yuan

Artificial Intelligence

Less Is Personal: Learning Minimal Sufficient User Profiles for Personalized Language Models

Retrieval-augmented personalization enables large language models to produce more accurate and preference-aligned outputs using relevant records retrieved from user histories. Personalized language models typically prepend a fixed number of retrieved user records, even when additional history is redundant, harmful, or unrelated to a user's distinctive behavior. We study minimal sufficient personalization: constructing the least costly ordered profile for each input while preserving the utility achievable from a retrieved candidate pool. We introduce ENOUGH, a method that iteratively appends behavioral records or emits STOP to construct profiles with adaptive lengths. Offline, bounded counterfactual search evaluates profile prefixes by jointly considering downstream gains, user specificity, and token costs. The resulting long-horizon targets are distilled into a multi-head value controller with explicit ranking and stopping supervision. At inference, the controller selects and orders records through lightweight decisions, and the frozen generator is invoked once after stopping. Extensive experiments on six personalized tasks demonstrate that ENOUGH consistently outperforms strong heuristic and retrieval-augmented baselines in both effectiveness and efficiency, achieving minimal sufficient profiles that preserve personalization utility while reducing unnecessary context costs.
Read the full abstract

All authors Minghang Liu, Qiang Qiu, Yuanzhuo Wang, Huawei Shen, Xueqi Cheng

Number Theory

Relative $(\varphi, \Gamma)$-modules and $p$-adic differential equations

Let $X$ be an affinoid rigid analytic space over a $p$-adic field, equipped with a suitable étale map to a unit polydisk. We provide a formalism of imperfect relative period rings over $X$, which lie inside the corresponding perfect relative period rings constructed by Kedlaya–Liu. Then we establish a relative version of the Fontaine–Cherbonnier–Colmez equivalence between $p$-adic local systems on $X$ and étale $(\varphi, \Gamma)$-modules over such imperfect relative period rings, generalizing previous works of Andreatta–Brinon and others. Using this equivalence, we construct $p$-adic differential equations attached to de Rham local systems, carrying both geometric and arithmetic differential operators. This generalizes the work of Berger to the relative geometric setting. Along the way, we study a relative Fontaine–Sen theory on the decompletion of $\Gamma$-modules over the relative $\mathbf{B}_{\mathrm{dR}}^+$-period rings.
Read the full abstract

Statistical Methodology

Multicollinearity-agnostic feature screening for non-Euclidean responses: a factor adjusted approach

In high-dimensional settings, multicollinearity is a pervasive issue that can substantially impair the performance of feature screening methods based on marginal Fréchet regression. Feature screening for non-Euclidean responses becomes unreliable when ultrahigh-dimensional predictors suffer from multicollinearity, because feature-specific signals may be masked by shared latent factors. To mitigate this effect, we propose a Factor adjusted Fréchet sure independence screening procedure. The method first recovers latent common factors from the predictors and then evaluates each feature by the incremental Fréchet coefficient of determination contributed by its idiosyncratic component beyond the common factors. Under regularity conditions, we establish uniform approximation rates for the feasible screening utilities and prove the sure screening and sure ranking properties. Extensive numerical experiments provide compelling empirical support for the validity and effectiveness of our approach, particularly in scenarios with highly correlated covariates. We further illustrate the practical performance of our method through two representative non-Euclidean datasets: the ADNI dataset and the mortality dataset, both with distribution-valued responses.
Read the full abstract

All authors Moshu Xu, Leheng Cai, Yanmei Shi, Xu Guo, Qirui Hu

Data Structures and Algorithms

Beyond Cut Balance: Spectral Sparsification of the Nonlinear Directed Laplacian

Digraphs with constant cut balance admit nearly linear directed cut sparsifiers. This condition requires the total arc weights in the two directions of every cut to be within a constant factor of each other. We ask whether this condition also permits nearly linear spectral sparsification with respect to the energy of the nonlinear directed Laplacian. For a weighted digraph $G=(V,E,w)$, let \[ Q_G^+(x)=\sum_{(u,v)\in E}w_{uv}(x_u-x_v)_+^2, \qquad (t)_+:=\max\{t,0\}. \] This energy agrees with the outgoing-cut function on binary vectors. A spectral sparsifier is a nonnegatively reweighted subgraph that preserves $Q_G^+(x)$ within a factor of $1\pm\varepsilon$ simultaneously for all $x\in\mathbb R^V$. We show that cut balance alone does not yield nearly linear spectral sparsifiers: for constant error, the worst-case support size for simple unweighted Eulerian digraphs is $\widetilde\Theta(n^{3/2})$, although Eulerian digraphs are perfectly cut-balanced and admit nearly linear directed cut sparsifiers. In contrast, we prove that every $n$-vertex tournament has a spectral sparsifier with $\widetilde O(n/\varepsilon^3)$ arcs, without any assumption on its cut balance. This includes the transitive tournament, whose cut balance is unbounded. Thus perfect balance does not guarantee nearly linear spectral sparsification, while unbounded imbalance does not preclude it. Finally, we use convex duality to show that preserving $Q_G^+$ also preserves, for every feasible demand vector, the optimum quadratic cost of a nonnegative flow. Hence the guarantee contains information beyond directed cut values.
Read the full abstract

Algebraic Geometry

Hilbert schemes of low-degree rational curves on a prime Fano threefold of degree $22$

In this paper, we give a complete description of the Hilbert schemes of rational curves up to degree $6$ on a prime Fano threefold $X$ of degree $22$. One of the key ingredients in the geometry of these Hilbert schemes is the geometry of bisecant conics associated with rational curves on $X$. As applications, we describe additional geometric features of these moduli spaces and derive Donaldson–Thomas (DT) type invariants.
Read the full abstract

Artificial Intelligence

A Theory of Reliable Self-Evolution for Agent Harnesses

In harness self-evolution, agents modify their own prompts, code, tools, and orchestration while keeping the underlying language model fixed. Recent work has shown that agents can improve themselves in response to task failures and achieve substantial performance gains. However, gains on failed tasks do not automatically ensure that performance on previously successful tasks is preserved, raising concerns about reliable adoption. In this work, we provide a theoretically grounded condition under which a self-evolved harness can be reliably adopted. We then propose a validation rule to make the evolved system satisfy the reliable adoption condition with theoretical guarantees. Consequently, the system can achieve progressive improvement through evolution. This leads to a natural question: Does reliable self-evolution have a performance ceiling, and which factors govern this ceiling? Our theoretical results show that the performance ceiling is determined by the costs of verification and evaluation. On the other hand, self-evolution may stall in practice. In this regard, we show that experimental evidence on agent performance shifts can be used to identify the sources of stagnation. Our work thus establishes a theoretical framework for understanding and advancing reliable harness self-evolution.
Read the full abstract

All authors Qianshu Cai, Yonggang Zhang, Jun Nie, Maohao Ran, Huajiang Zheng, Jun Song, Xinmei Tian, Yike Guo, Wei Xue

Artificial Intelligence

OntologyBench: Can Dense Retrieval Satisfy Structured Biomedical Constraints?

We introduce OntologyBench, a tiered biomedical retrieval benchmark comprising 471,854 training and 125,744 evaluation query-document relevance pairs across concept grounding, relational retrieval, and compositional phenotype-based retrieval. Although these tasks can be tractable using ontology-aware reference methods, across task tiers, embedding performance is generally lower on relational and compositional tasks than on concept-grounding tasks. Fine-tuning on ontology-derived supervision improves performance on several relational and compositional tasks, whereas the evaluated reranking and LLM-based candidate-scoring methods provide little or no end-to-end improvement. Errors frequently reflect diseases matching only subsets of the phenotype evidence. These findings indicate that the evaluated embedding and reranking configurations do not reliably recover the compatibility encoded by the selected ontology relations and phenotype combinations and motivate retrieval systems that better integrate learned representations with structured biomedical knowledge.
Read the full abstract

Artificial Intelligence

Key Path Identification for Resolving Knowledge Conflicts via SAE-based Steering

Sparse autoencoder (SAE)-based steering has been widely used to address knowledge conflicts by guiding LLMs to be more faithful to the contextual knowledge. Existing methods usually perform mass steering, which modifies a large batch of SAE features identified via correlation-based methods. However, due to the inaccurate correlation and the neglected feature interactions, mass steering methods fail to precisely identify the features that play the key roles in steering and introduce a large number of redundant ones, which add noise and weaken the steering effects. Our empirical studies reveal that steering only a small subset of the identified features can achieve comparable or even better performance. Motivated by this finding, we propose Key Path Identification (KPI), a novel method that identifies key steering features characterized by strong causal dependencies with both upstream and downstream features. From these features, KPI constructs key paths and steers through less feature modifications. In this way, KPI advances SAE-based steering from quantity-driven to quality-focused, offering a perspective for more precise and interpretable model editing. Experiments in RAG tasks with knowledge conflicts show that our method improves the accuracy by 18% on average compared to the best baseline of mass steering, effectively filtering redundant features, alleviating side effects and demonstrating the core role of key paths in steering.
Read the full abstract

All authors Wenbo Zhang, Zhongxiang Sun, Zhiguang Han, Jun Xu

Statistical Computation

Optimal Slice-Adaptive Tuning of Hybrid Slice Sampling

Slice sampling is a Markov chain Monte Carlo algorithm that draws its next state uniformly from a "slice"—a super-level set of the target density function—at each iteration, thereby providing automatic local adaptivity to the scale of the target. In practice the exact slice is not known, so general-purpose implementations use an approximate slice that is grown from a starting interval of length $w>0$, with a computational cost that depends on $w$. This work presents an analysis of the average per-iteration number of target density evaluations, as a function of $w$, of hybrid slice sampling with various slice-finding schemes for targets with contiguous slices. The paper uses the results of the analysis to develop automated, slice-adaptive tuning schemes along with suboptimality bounds and asymptotic convergence guarantees. Simulations demonstrate that the tuning schemes reliably yield near-optimal slice-adaptive tuning with essentially no dependence on the initial setting of $w$.
Read the full abstract

Computation and Language

EviSI: An Evidence-Based Evaluation Agent for Simultaneous Interpreting

Low-latency simultaneous speech-to-speech translation must keep pace with ongoing speech while preserving key information. To meet these demands, systems use segmentation, reformulation and condensation to reorganize and rephrase information. However, metrics developed for text translation, including BLEU and COMET, may not consistently distinguish faithful adaptations from semantic errors. We propose EviSI, a large language model evaluation agent combining Multidimensional Quality Metrics (MQM) with criteria developed with professional interpreters. Shared source evidence guides assessment across four dimensions: Anchor, Event, Logic and Fluency. Verified errors are deduplicated before deterministic scoring. On human-rated English to Chinese and Chinese to English data, EviSI recovers the aggregate English to Chinese human system ranking. Mean within-dataset Kendall correlations for system rankings reach 0.707 and 0.467, respectively, exceeding evaluated BLEU and COMET baselines. A multilingual extension to five directions without human ratings retains the dimensions and scoring rule, showing positive system ranking correlations with COMET throughout.
Read the full abstract

All authors Ben Yan, Zongyao Li, Xiaoyu Chen, Daimeng Wei, Weidong Liu, Huan Zhao, Chong Li, Yaode Wang, Yuzhe Shang