Peer Revenue Past citations are no guarantee of future returns

Papers

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

Computer Science

cs Watch Computer Science (sign in to use)

Thursday 13 August

Information Retrieval

STAR: Structured Tokenization and Target-Aware Interest Representation for PCVR Prediction

Post-click conversion rate (PCVR) prediction is a core ranking task in industrial recommender systems. Modern ranking models must jointly capture heterogeneous non-sequential features, multi-behavior user sequences, and target-item-aware user interests, while remaining robust to high-cardinality sparse features, missing values, and train-inference inconsistencies. In this paper, we present STAR (Structured Tokenization and Target-Aware Interest Representation), a practical framework for the KDD Cup 2026 Tencent UniRec Challenge. STAR combines structured feature tokenization with target-aware interest representation on top of a HyFormer-style multi-sequence backbone. It introduces high-cardinality signal recovery, explicit user-item interaction tokens, target-aware sequence decoding, and a weighted user-item contrastive auxiliary objective inspired by InfoNCE. We further align the training and inference pipelines by reconstructing feature remapping tables and structural hyperparameters from the saved training configuration. Experiments on the challenge dataset identify the components that most reliably improve ranking AUC, while LogLoss is reported as a calibration diagnostic. The main ablation study shows a large gain from temporal context, with smaller but useful contributions from contrastive alignment, target-aware interest encoding, and high-cardinality sequence feature recovery.
Read the full abstract

All authors Yimeng Xu, Ruihao Zhang, Yingqi Song, Ying Jiang, Lan Ma

Multiagent Systems

Reconcile Once, Write Anytime: A Trust-Tiered Librarian and a Multi-Agent Writer for Drift-Free, Point-in-Time Research

Long-form research reports generated by large language models drift, contradict themselves, and lose provenance: the same metric appears with different values, and rumor is quoted as confidently as an audited filing. We present a two-tier agentic system that separates a maintained, point-in-time knowledge library from report writing. A deterministic "librarian" ingests timestamped sources into a trust-tiered ontology, layering evidence cards, an authoritative metric ledger, and a claim graph into an always-current source of truth, not per-query RAG over raw chunks. A portable multi-agent "writer" runtime then composes a contradiction-free, evidence-grounded report at any knowledge cutoff T, reading only evidence with as_of <= T (no look-ahead); red-team verdicts flow back into the librarian. We evaluate on a self-collected, public corpus of 6,130 sources yielding 555,926 evidence cards (SEC EDGAR filings across 295 issuers and 11 sectors, U.S. Bureau of Labor Statistics releases, and Wikipedia). From the one library we compose four point-in-time reports on distinct theses and run eight reproducible experiments, whose headline metrics come from a deterministic quality-control gate, itself validated by defect-injection meta-evaluation at recall 1.0 and precision 1.0. A shared metric ledger removes 6,845 cross-section contradictions to zero. Tier-first selection is correct on 22/22 gold cases where a popularity-first baseline scores only 9/22; trust tiering leaks zero media-sourced numbers, and no government statistic displaces a company's own filing. A red-team refutation propagates back and self-corrects a later run with zero manual edits. Replay exhibits zero look-ahead violations across seven cutoffs while the library grows from 235,373 to 555,312 cards. Difficulty-tiered model routing exceeds the all-Opus quality ceiling while running 3.7x faster than serial.
Read the full abstract

All authors Xing Zhang, Yanwei Cui, Guanghui Wang, Peiyang He

Machine Learning

Learning the Mathematical Property for Designing Low Mutual Coherence Binary Sensing Matrices

In this research work, we are constructing the sensing matrix, which is essential for the success of the compressive sensing technique. We have chosen a learning-based technique for the construction of the sensing matrix. The novelty and uniqueness of the proposed technique is that it does not use any data set and also does not use a specific application. It uses the mathematical property/constraint for the construction of the sensing matrix for the perfect recovery of the signal. The perfect recovery of signals is an old and still very challenging problem in real-world applications. In late 2000, compressive sensing became a popular mathematical tool for the perfect recovery of sparse signals. The core of the compressive technique is the construction of the sensing matrix, which satisfies certain special properties such as restricted isometry property (RIP), null space property (NSP), and spark property (SP). All these properties are NP-hard problems and hence computationally challenging to solve. For all practical purposes, the construction of the sensing matrix needs to achieve low mutual coherence to achieve the perfect recovery of the signals. We have used a neural network for the construction of the sensing matrix, and this framework constructs a binary sensing matrix with low mutual coherence. The entries in the matrix are generated through a shared underlying rule. The proposed architecture is simple and does not use large-scale training data sets. Such uniqueness and novelty bring a drastic reduction in computational cost, and also, for the first time in literature, the use of a mathematical property for defining the loss function. In this proposed research work, the mutual coherence property has been used in the neural network framework. Such a neural network framework brings generality, robustness, and reduces storage requirements.
Read the full abstract

Computer Vision

DiCoR: Decoupled Referent Disambiguation and Contour Recalibration for Efficient Referring Remote Sensing Image Segmentation

Referring remote sensing image segmentation (RRSIS) aims to delineate targets specified by natural language expressions in remote sensing imagery. Existing methods mainly follow joint fusion segmentation (JFS) or decoupled prompt segmentation (DPS). JFS is efficient but often suffers from limited accuracy because referent localization and mask delineation are optimized under a unified objective, whereas DPS separates localization from mask generation using spatial prompts and foundation segmenters at the cost of higher memory consumption and inference latency. To bridge this gap, we propose DiCoR, a decoupled referent disambiguation and contour recalibration framework built on an efficient JFS pipeline. DiCoR addresses two key challenges: distinguishing the correct referent from ambiguous candidates and refining coarse masks after localization. A disambiguation-aware localization guidance strategy ranks salient candidate regions with adaptive linguistic cues and injects the resulting localization prior into fused features. A lightweight contour recalibration module further predicts residual corrections to coarse logits under localized contour supervision, improving mask quality with limited computational overhead. Experiments on RefSegRS, RRSIS-D, and RISBench show that DiCoR achieves the best segmentation accuracy across all three benchmarks. On RefSegRS, it improves mIoU and gIoU by 4.90% and 2.16% over a competitive JFS method while running 2.4-6.5 times faster than the evaluated DPS method, demonstrating a favorable accuracy-efficiency trade-off. Code is available at https://github.com/zyGao1126/DiCoR.
Read the full abstract

All authors Ziyang Gao, Zhizhuo Jiang, Jingjing Chang, Yixin Yang, Yuwen Pan, Yong-Qiang Mao, Yu Liu, Hai-Bao Chen

Databases

Semantic Intelligence Against CSAM: The PreventCSA@EU Ontology Framework for Classification and Investigation

This work presents the PreventCSA@EU ontology, a semantically grounded framework designed to support the identification, classification, annotation, and analysis of online Child Sexual Abuse and Child Sexual Exploitation Material (CSAM/CSEM). The growing circulation and dissemination of CSAM/CSEM across digital environments, combined with inconsistencies in legal definitions and classification practices across jurisdictions, highlights the need for semantically interoperable frameworks capable of supporting cross-organizational cooperation and automated processing. The proposed ontology is developed through a systematic review and comparative analysis of existing CSA/CSE-related, metadata oriented, and investigative ontologies and taxonomies, with its primary design aimed at addressing the operational needs and domain-specific requirements of national LEA Directorates. It introduces a hierarchical semantic model built around core entities such as Media Object, Content, Person, Depiction, and Investigative Report, while enabling structured alignment with INHOPE UCS labels, Dublin Core-DMCI Metadata Terms, and Schema.org. The proposed framework emphasizes ontology-driven interoperability for structured annotation and analysis of CSA/CSE-related data, supporting consistent classification, child identification, and investigative processes for offender prosecution. The design aims extend existing classification approaches with additional conceptual structures for database conceptualization, process modeling, and ontology-driven data management. By integrating established classification standards with a novel hierarchical ontology, the proposed framework enhances cross-system compatibility, with particular relevance to emerging EU-level data infrastructures, including the envisaged EU Center database under the proposed Child Sexual Abuse Regulation (CSAR).
Read the full abstract

All authors Elias Tzortzakakis, Emmanouela Kokolaki, Evangelia Daskalaki, Paraskevi Fragopoulou

Cryptography and Security

Beyond Handcrafted Security: Towards Self-Evolving Defense for LLM Agents

The expanding operational capabilities of large language model (LLM) agents introduce sophisticated security threats. Runtime defenses have emerged as an effective approach to mitigating these risks by integrating security mechanisms into the agent execution loop. However, existing runtime defenses rely heavily on manually designed interventions and lack a principled framework for their construction and maintenance. In this work, we first develop a harness-level formulation of runtime defense that systematically characterizes how harness mechanisms enable defense construction and provides a unified view of existing runtime defense interventions from a harness perspective. Building on this formulation, we propose HARD (Harness-based Autonomous Runtime Defense Evolution), a self-evolving runtime defense framework that automatically identifies appropriate intervention strategies and iteratively improves defense artifacts based on observed failure traces. HARD transforms runtime defense development from manual engineering into an autonomous evolution process, and extensive experiments demonstrate that it improves security performance over existing handcrafted defenses while preserving benign task utility. Our findings highlight autonomous defense evolution as a promising new paradigm for securing deployed LLM agents, enabling agents to identify defense weaknesses and continuously improve their protection mechanisms.
Read the full abstract

All authors Jiajun Ruan, Peiyang Li, Yukun Chen, Fengting Li, Chao Feng

Computational Complexity

Fortune's Bounty: Taming Complexity by Trimming Trees — A Hands-On Problem-Solving Experience in Advanced Complexity Suitable for Introductory Students

This article provides an assignment designed to let undergraduate students who have completed an undergraduate CS1/CS2 sequence try to themselves, in groups, prove Fortune's Theorem. (Fortune's Theorem states that if the complement of the Boolean satisfiability problem polynomial-time reduces to a sparse set, then the Boolean satisfiability problem is polynomial-time computable. The assignment does not assume that students have previously seen the Boolean satisfiability problem, polynomial-time reductions, or sparse sets. Rather, it teaches those within the assignment. Note: Reworded into the technical vocabulary of complexity theory, Fortune's Theorem states that no sparse set is coNP-hard unless P=NP. Fortune's Theorem was a major advance in the understanding of the relationship between hardness and density.) We provide both the assignment handout (as the main body of this report plus Appendix A) and a solution to the assignment (as Appendix B, which would of course not be made available to the students until after they had handed in the assignment). The assignment handout, though the instructor can change this, is framed as having the students starting the assignment in teams in class for a whole class session, and then finishing it in those same teams as a take-home assignment, and handing it in before the next class session. We have found that student groups often succeed, partially or completely, in this challenge. This can mean a lot to the students: they see that they were able to make an advance that, when it was first obtained, appeared in what was arguably at the time the top journal venue for complexity theory research. This can give them confidence that they have substantial problem-solving skills (which basically means research skills) when they truly apply themselves to a given challenge.
Read the full abstract

Social and Information Networks

DTAMLP: Denoise Time-aware MLP for Session-based Recommendation

This paper reports two empirical findings on session-based recommendation (SBR), unified in a single model, DTAMLP. First, existing time-aware and GNN-based models (e.g., TiSASRec, SR-GNN) treat every click-time interval as equally informative, even though very short dwell times often reflect accidental clicks carrying little preference signal – a phenomenon we call sporadic noise. We show that a lightweight, plug-and-play weight fusion module, blending a model's attention weight with a threshold-capped time-interval weight, can be inserted into such models with almost no architectural change and yields a consistent accuracy gain; we view this as the most directly verifiable contribution of this work. Second, we revisit an under-explained observation from FMLP-Rec, where a learnable frequency-domain filter on item embeddings improves accuracy, and offer a possible explanation: time-domain behavior mixes several entangled psychological preferences, and a frequency-domain view may let a model separate and down-weight such preference noise more naturally – an interpretive conjecture rather than a proven mechanism. Building on both insights, DTAMLP, an all-MLP framework combining weight fusion and FFT-based filtering, is validated on Diginetica and RetailRocket. While this system-level design reflects the state of the field circa 2023 rather than a state-of-the-art claim, ablations confirm the two mechanisms contribute complementary, non-redundant improvements.
Read the full abstract

Machine Learning

Comment on "Modeling rapid language learning by distilling Bayesian priors into artificial neural networks"

McCoy & Griffiths (2025, henceforth M&G) suggest that a Bayesian prior can be distilled into Artificial Neural Networks (ANNs) through Model-Agnostic Meta-Learning (MAML, Finn et al., 2017). They support this empirically by showing that meta-trained networks demonstrate formal language learning abilities comparable to Yang & Piantadosi (2023)'s Bayesian learner, significantly outperforming standard ANNs. We point out that under the standard interpretation of a prior, M&G's procedure does not actually instill one; it merely initializes network weights favorably, leaving the objective function unchanged. We then consider a more permissive interpretation, where the system as a whole can be seen as implementing a Bayesian learner even without an explicit prior in the objective. We show that this interpretation faces nontrivial challenges. Finally, we assess how well MAML approximates the empirical results of Bayesian learning, showing that unlike genuine Bayesian learners, M&G's model overfits and generalizes poorly to unseen data.
Read the full abstract

All authors Orr Well, Idan Tarshish, Nur Lan, Roni Katzir

Machine Learning · Cross-listed from stat.ML

Online Inference for Quantile Distributional Reinforcement Learning

In this paper, we study online statistical inference for quantile-based distributional policy evaluation, which aims to characterize the distribution of discounted cumulative rewards under a given policy. To obtain a finite-dimensional representation of the return distribution, we consider the fixed point $\bm\eta_m$ of the quantile-projected distributional Bellman equation, parameterized by the quantile vector $\bm\theta_m$. We develop two online inference procedures for $\bm\theta_m$, covering synchronous and asynchronous sampling from a generative model as well as Markovian data. The first method uses the trajectory of quantile temporal-difference learning (QTD). We establish functional central limit theorems for the partial sums of QTD iterates and construct a recursively computable random-scaling statistic. The resulting confidence intervals have asymptotically exact coverage at deterministic horizons, with length of order $T^{-1/2}$ in probability. The second method constructs anytime-valid confidence intervals directly from samples. Exploiting the monotonicity of the quantile Bellman map, we develop a computationally explicit coordinate-shaving algorithm with finite-sample simultaneous coverage. Under positive limiting state-visit frequencies, the resulting confidence intervals have lengths of order $\sqrt{\log\log T/T}$, which we establish as classwise optimal through a matching lower bound. Together, these results provide complementary guarantees for online inference in quantile distributional reinforcement learning, combining asymptotically exact fixed-horizon inference with finite-sample anytime validity.
Read the full abstract

Computer Vision

Bias Mitigation in Face Recognition via Demographic-based Supervised Contrastive Learning

Face recognition systems have been shown to be biased toward certain demographic groups by exhibiting different error rates across gender, age, or ethnicity. Though the imbalance of the training data with respect to these demographics is one cause of this bias, training on artificially balanced groups does not completely mitigate the problem. For deployment, face recognition typically works at operating points allowing very low false match rates and, hence, on the tail of the non-match score distribution. While class balancing can improve the means of these distributions, the aim of our approach is to improve fairness by addressing the behavior in the tail. Particularly, we propose the Demographic-based Supervised Contrastive loss (DeSCon) for face recognition, which relies on a well-designed composition of training batches and demographic-aware pair selection. Our experimental evaluation on both demographically-labeled datasets and standard verification benchmarks shows that DeSCon can improve fairness beyond balancing training datasets while maintaining competitive verification performance. Source code is available upon request.
Read the full abstract

All authors Yu Linghu, Salman Mohammad, Xinyi Zhang, Manuel Günther

Software Engineering

Requirements-Augmented Generation for Trustworthy Acceptance Testing of LLM-Based Software

LLM-based software (LBS) integrates large language models as core components to deliver flexible, personalised responses. Unlike traditional software with deterministic outputs, LBSs exhibit context-dependent, stochastic behaviour that renders classical acceptance testing and test oracles insufficient: the same query may require fundamentally different responses depending on user personas and software context. This gap creates an urgent need for automated acceptance testing frameworks that autonomously interpret user instructions, while reliably inferring user intentions in a changing environment. In this paper, we present an automated acceptance testing framework for LBS with calibrated verdict reliability via two technical contributions. First, we introduce Requirements-Augmented Generation (REAG), which interprets user intentions by retrieving relevant software requirements, domain knowledge, and personas via adaptive RAG and self-reasoning to generate context-aware test oracles. Second, recognising that oracle generation may retrieve irrelevant constraints, misinterpret intent, or hallucinate requirements, we introduce a confidence-calibrated cascade judgment. This method quantifies verdict reliability via simulated expert agreement – accepting high-confidence verdicts, escalating ambiguous cases, or abstaining when uncertain – with empirical reliability guarantees backed by conformal risk control. An industrial case study on a production nutrition advisory application demonstrates that REAG achieves a 3.91/5 oracle quality score, reaching qualified or marginal oracle quality in 82% of cases. The confidence-calibrated cascade achieves 98.8% accuracy, improves oracle quality from 3.91 to 4.30 by filtering unqualified outputs, and delivers a 31.7% cost-efficiency improvement over single-judge baselines, validating industrial viability
Read the full abstract

All authors Fanyu Wang, Chetan Arora, Zhenping Xie, Yonghui Liu, Kla Tantithamthavorn, Aldeida Aleti, Siwei Jiang

Machine Learning

Understanding Backdoor Vulnerabilities in Vertical Federated Learning: The Gap Between Research and Practice

Vertical Federated Learning (VFL) enables organizations holding complementary features of shared entities to collaborate and train models. In this setting, the initiator can withhold information about the learning task, while other contributors participate without exposing their local datasets, creating an asymmetric information structure aligned with growing privacy demands. However, this asymmetry is a double-edged sword. Among various threats, backdoor attacks are particularly concerning because VFL not only enables malicious contributors to poison the model during training, but also allows them to activate the backdoor at inference time to manipulate predictions. Although prior work has reported near-perfect attack success rates and proposed effective defenses, we find that most findings fail to hold under realistic conditions, exposing a fundamental gap between research and practice. In this paper, we present a systematic, practice-oriented study of backdoor vulnerabilities in VFL, revealing this gap in both methodological design and evaluation practices. We show that existing approaches overlook key practical constraints and therefore rely on unrealistic prior knowledge. Furthermore, these limitations have remained hidden due to poorly designed evaluation practices in the literature. To bridge this gap, we redefine threat models under realistic constraints, propose practical backdoor workflows, and introduce BVBench, a backdoor-centric benchmark that enables fair, practical, and comprehensive evaluation, preloaded with state-of-the-art baselines. BVBench provides strong evidence of the fragility of the current understanding of VFL backdoor risks and establishes a foundation for steering research toward uncovering practical vulnerabilities and developing more meaningful defenses.
Read the full abstract

All authors Ziqi Zhao, Jialin Lu, Junjie Shan, Junyuan Zhang, Shuya Yang, Ka-Ho Chow

Artificial Intelligence

Moose: Latent concept learning with reasoning-shortcut awareness in $\mathcal{EL}^{++}$

The OWL 2 EL profile is used in some of the largest production ontologies, including the Gene Ontology and SNOMED CT. Existing neuro-symbolic (NeSy) learning methods accept propositional theories or Datalog, and reasoning-shortcut (RS) awareness has not been investigated in ontology settings. We present Moose, a method that compiles an $\mathcal{EL}^{++}$ TBox and finite ABox to a Sentential Decision Diagram (SDD). The SDD acts as a differentiable weighted-model-counting layer, and we add closure clauses outside the $\mathcal{EL}^{++}$ profile on declared exhaustive families to overcome the limited expressivity of $\mathcal{EL}^{++}$ under partial supervision. We show termination, soundness, completeness, and polynomial intermediate sizes, and validate the proofs in Lean. We then define the first formal partial-supervision latent-concept-learning task over an OWL EL ontology, i.e., learning per-individual classifiers for latent concepts from observed ABox literals, and evaluate Moose on MNIST-with-ontology and Pizzaïolo. Moose improves over propositional-NeSy, fuzzy-logic, and ontology embedding baselines, and presents the first reasoning-shortcut analysis in an OWL EL setting.
Read the full abstract

All authors Olga Mashkova, Asaad Mohammedsaleh, Fernando Zhapa-Camacho, Robert Hoehndorf

Computer Vision

A Deep RL based Framework for Targeted White Matter Tractography

Fiber tractography's ability to reconstruct the brain's structural pathways, has made it a crucial component of modern neuroimaging, enabling detailed, non-invasive mapping of structural connectivity and supporting a wide range of neurological research and clinical applications. However, despite its importance, tractography remains a challenging task due to the inherent complexity of white matter structure and its susceptibility to false positives, which can lead to the misrepresentation of critical pathways. To overcome these limitations, in this thesis, we propose a hybrid framework that integrates reinforcement learning with supervised learning for refining RL policies, specifically tailored for tract-specific tractography. Notably, our framework does not rely on ground-truth fibers for training. Moreover, the tract-specific formulation bypasses the need for an explicit segmentation process, simplifying the overall pipeline. Our work includes two main contributions, each building upon the previous. First, we introduce a hybrid approach that combines reinforcement learning with supervised learning (specifically, GPT-based policy learning) to refine policies in a tract-specific context. Second, we propose a scalable framework for data-driven multi-policy fusion, which leverages the complementary strengths of multiple RL policies to improve tractography performance and robustness. We demonstrate the effectiveness of our framework through extensive validation on benchmark public datasets including TractoInferno, HCP, and ISMRM-2015, highlighting its ability to generalize across data sources and accurately reconstruct brain white matter tracts. We believe that these contributions represent significant advancements in the field of tractography, improving robustness, reliability, and accuracy while reducing dependence on ground-truth annotations.
Read the full abstract

Machine Learning

The Objective Is the Bottleneck: Latent World Models Encode What Their Planners Cannot Use

Latent world models are judged by how well they predict, so when planning fails at long horizons the natural reading is that the predictor degrades. On a reproduction of LeWorldModel on TwoRoom we show the binding constraint is the planner's objective instead. The predictor is not the limit: its imagined state seventy-five environment steps ahead is still only 0.189 as wrong as assuming the world froze, while the planner never imagines beyond twenty-five. The objective is. Cross-entropy-method planning minimises squared latent distance, which tracks true distance at r = 0.426, saturates by about eighty arena units and decreases beyond a hundred and twenty, so moving away from the goal can lower the cost. The information is present throughout: a ridge probe recovers position from the frozen embedding at R^2 0.9922. The pathology is the method's, not one reimplementation's. It is present in the authors' released weights, and across four checkpoints long-horizon success rank-orders exactly with metric quality and inversely with prediction accuracy. Replacing only the objective, with nothing retrained and no GPU, lifts goals reached at offset 100 from 26.0% to 98.0%, equals the 98.0% at offset 25, and reaches 92.0% under a third of the budget: planning stops depending on the horizon. The best cost is not the most accurate. A head learned from frame separation alone predicts spatial distance worse than a position probe (r = 0.819 against 0.9897) yet plans better, charging 24% more to cross the environment's dividing wall where squared latent distance charges 4% less. It has learned reachability, not proximity.
Read the full abstract

Emerging Technologies

Exponential multi-graylevel computational-weighted dithering for high-quality binarized Fourier single-pixel imaging

Binarized Fourier single-pixel imaging (FSI) takes full advantage of the high modulation speed of digital micromirror devices by applying Floyd-Steinberg spatial dithering to binarize grayscale Fourier patterns. However, the use of the spatial dithering introduces substantial quantization errors, leading to decreasing imaging quality. Here, we propose a binarization method for grayscale Fourier patterns based on exponential multi-graylevel computational-weighted dithering, aimed at reducing quantization errors and then enhancing the imaging quality of binarized FSI. The proposed method quantizes Fourier patterns into $2^{R}$ values $\{0, 1/(R-1), 2/(R-1)... 1\}$ and then decomposes them into binarized patterns. Both simulation and experimental results demonstrate that the method significantly reduces quantization errors in Fourier coefficients acquisition and improves imaging quality. The mean absolute percentage error of Fourier coefficients decreases from $194\%$ to $28\%$ and the structural similarity of reconstructed images ($256\times256$ pixels) improves from 0.430 to 0.971, a $126\%$ enhancement compared to the conventional method. Lateral resolution of this proposed method almost approaches the theoretical lateral resolution limit calculated by Rayleigh Criterion.
Read the full abstract

All authors Qigao Zhu, Haojia Jiang, Guan Wang, Lianhao Zhang, Hanlei Gong, Huaxia Deng, Xinglong Gong

Machine Learning

I-SDPO: Instance-Level Adaptive Self-Distillation Policy Optimization

Group Relative Policy Optimization (GRPO) learns from reward differences within a rollout group, but receives no useful relative signal when every sampled response is incorrect. Privileged self-distillation can fill this gap with dense token supervision, yet applying it throughout training creates a different failure mode: the teacher is a biased, low-variance surrogate for the reward objective, so persistent imitation can oppose reward-improving updates after the policy becomes capable of producing successful trajectories. We introduce I-SDPO (Instance-Level Adaptive Self-Distillation Policy Optimization), which treats teacher reliance as capability-dependent. I-SDPO makes one routing decision per input instance and shares it across that instance's rollout group: all-incorrect groups use a privileged self-distillation objective, whereas any-success groups remain intact for GRPO. This design uses imitation only where group-relative rewards are uninformative. A local analysis characterizes when teacher and reward directions align and shows that a non-vanishing biased distillation weight induces an optimization bias floor. The routing rule automatically reduces the expected distillation rate as success probability rises, withdrawing teacher influence without a hand-designed schedule. On SciKnowEval, I-SDPO obtains the best result in all four scientific domains and improves average mean@16 accuracy from 56.67% with GRPO to 70.31%, with a maximum domain gain of 18.24 points.
Read the full abstract

All authors Yubo Zhang, Xinhong Ma, Zezhong Tan, Ziqiang Dong

Computational Complexity

PSPACE-Completeness of Multi-Agent Path Finding for Large Agents

Multi-Agent Path Finding for Large Agents (LA-MAPF) is a geometric variant of MAPF in which agents are modeled as disks and conflicts are determined by physical overlap in the underlying Euclidean workspace. The goal of LA-MAPF is to decide whether there exists a sequence of conflict-free transitions from a start configuration to a goal configuration. Agafonov and Yakovlev proved that LA-MAPF is NP-hard. In this paper, we strengthen their result by proving that LA-MAPF is PSPACE-complete via a polynomial-time reduction from Restricted Sliding Tokens.
Read the full abstract

Computation and Language

Unifying Depth and Width Pruning for LLMs via Binary Knapsack Optimization

Structured pruning is a promising approach for compressing large language models (LLMs), yet existing methods rely heavily on greedy heuristics that produce myopic decisions, and often fail to precisely meet target compression budgets. We present SNIPER, a two-stage structured pruning framework that solves a knapsack optimization over coarse-granularity components to yield conditionally optimal parameter allocations with respect to fixed importance estimates, followed by a fine-grained pruning stage to meet strict budget constraints. We introduce the Compression Ratio Adherence Factor (CRAFT) to quantify budget fidelity, showing that while existing pruners deviate from target compression ratios by up to 33%, SNIPER achieves near-exact adherence with a CRAFT score of 0.98. Evaluations across four diverse architectures over a set of 18 tasks spanning five domains demonstrate SNIPER's consistent improvements in average performance retention and task-level stability over six state-of-the-art pruners. Across all pruning configurations, SNIPER achieves an excellent mean rank of 1.25, indicating its robust cross-architectural generalizability and excellent reliability.
Read the full abstract

All authors Palaash Goel, Ayan Sengupta, Akshay Nambi, Tanmoy Chakraborty

Sound

VoxAudio: Vocalized Audio Synthesis via Multi-Reward Autoregressive Flow Matching

Vocalized audio synthesis, the task of generating audio in which intelligible speech is embedded within an environmental soundscape, underpins applications such as podcast production and video dubbing. Existing Text-to-Audio (T2A) systems either reduce quoted speech to unintelligible vocal murmur or delegate it to a separate TTS model with post-hoc mixing, which forfeits control over when speech occurs and how it interacts with the scene. We present VoxAudio, a causal autoregressive flow matching model that addresses this problem from three complementary aspects. At the architecture level, chunk-wise causal factorization with independent per-chunk noise levels lets audio be emitted through sliding-window streaming inference with KV caching at variable target durations; to enable inference at arbitrary chunk granularities, we further pretrain the model with randomized chunk boundaries. At the preference level, multi-reward Negative-aware FineTuning (NFT) jointly optimizes semantic fidelity, linguistic accuracy, aesthetic quality, and temporal grounding At the data level, to supply the missing supervision for vocal content, we build VoxCorpus, a large-scale corpus whose captions quote the verbatim transcript of embedded speech with time intervals, and VoxBench, an interval-annotated benchmark with a temporal-grounding metric. Experiments on four benchmarks spanning general audio, speech, and unified vocalized audio validate the effectiveness and efficiency of VoxAudio. Our code and demos are available at https://voxaudio.github.io.
Read the full abstract

All authors Wenxiang Guo, Changhao Pan, Ziyue Jiang, Zhou Zhao, Fei Wu

Discrete Mathematics · Cross-listed from math.CO

A relaxation of the Bermond-Thomassen conjecture

The well-known Bermond-Thomassen conjecture states that every digraph of minimum out-degree at least $2k-1$ contains $k$ vertex-disjoint directed cycles. Despite being posed in 1981, this conjecture remains unresolved for all $k \ge 4$. We prove a relaxation of this conjecture: every digraph $D$ of minimum out-degree at least $2k-1$ contains $k$ vertex-disjoint cycles, each of which either is directed or can be made directed by reversing one of its arcs. This bound is sharp and answers a question raised by Cames van Batenburg during the online workshop "Entropy Compression and Related Methods" in $2021$.
Read the full abstract

All authors Stéphane Bessy, Matthijs Muis, Jean-Sébastien Sereni, Raphael Steiner, Sebastian Wiederrecht

Distributed Computing

Efficient Randomized LL/SC that Preserves History Independence

We study the fundamental problem of implementing $m$ linearizable LL/SC objects with constant expected step complexity in a system of $n$ processes, using bounded base objects commonly available in hardware. Assuming that each process may have at most $\tau$ outstanding LL() operations, the best known deterministic algorithm requires $\Omega(n^2\tau + m)$ base objects (CAS objects and registers) [Blelloch and Wei, DISC 2020]. Previously, no comparable randomized algorithm was known. By employing randomization and FADD objects in addition to CAS objects, we obtain a space bound of $O(n\tau+m)$ against the weak adaptive adversary. For $m=O(1)$ this matches a lower bound for algorithms using CAS objects and registers [Aghazadeh and Woelfel, PODC 2015]. In addition, our object can be employed by quiescently history-independent (QHI) algorithms: Whenever no operation on the object is pending and no process has an outstanding LL() operation, its internal memory state is uniquely determined by the values of the $m$ LL/SC objects. An important application is a recent QHI dynamic hashing algorithm, which uses $\Theta(m)$ hardware LL/SC objects to maintain a hash table of size $m$ [Attiya, Bender, Farach-Colton, Oshman, and Schiller, STOC 2025]. But LL/SC is not available in hardware, and prior to our work no wait-free or efficient lock-free software implementation of LL/SC with similar properties was known. Our work demonstrates that one can actually implement the hashing algorithm on available hardware, without an asymptotic increase in step and space complexity, under the reasonable assumption that $m=\Omega(n)$.
Read the full abstract

Computational Engineering

Spinal Coupling in Frontal and Transversal Plane During Gait - A Segmental and Time-Dependent Analysis of the Thoracic and Lumbar Spine

Introduction: Coupling between lateral deviation and axial rotation is a known feature of spinal mechanics, yet its behavior at the individual vertebral level during gait, as well as the association with sagittal posture remains poorly understood. Methods: This study analyzed spinal kinematics in a diverse cohort (n=642) using a non-invasive rasterstereography system with an instrumented treadmill, quantifying time-dependent coupling of rotation and lateral deviation for each vertebra from T3 to L4 during walking as well as the influence of sagittal posture on this coupling. Coupling behavior was analyzed with absolute phase lag, normalized signed area and tilt angle derived from the Fourier series. Results: Results revealed cranio-caudal patterns for all three metrics with differences between almost all adjacent vertebrae and different turning points, i.e., where the coupling behavior changed (from increase to decrease and vice versa). Generalized, as well as pooled static sagittal posture significantly modulated these patterns, while individual deviations from static sagittal posture influenced the metrics. Conclusion: These findings provide the first dynamic, vertebra-level characterization of spinal coupling during gait. Furthermore, they offer a foundation for an understanding of spinal biomechanics and the influence of static sagittal posture, potentially relevant to the diagnosis and treatment of spinal disorders.
Read the full abstract

All authors Jonas Dully, Carlo Dindorf, Henriette Rönsch, Dennis Perchthaler, Jürgen Konradi, Ulrich Betz, Michael Fröhlich

Machine Learning

CardioState-JEPA: Delay-Aware Cross-Modal Learning of a Shared Cardiac Representation

Electrocardiography (ECG), photoplethysmography (PPG), and phonocardiography (PCG) provide complementary views of the same cardiac cycle, yet existing cardiac foundation models are trained for a single sensing modality, leaving the shared physiology across sensors unexploited. We introduce CardioState-JEPA, a cardiac foundation model to learn a single shared representation jointly across ECG, PPG, and PCG, built on a physiology-aware joint-embedding predictive architecture. The model maps heterogeneous waveforms into a common token space, processes them with a single shared Transformer encoder, and learns by predicting masked latent cardiac states, placing the pretraining target on shared physiology rather than sensor-specific waveform appearance. To handle the temporal offsets between electrical, mechanical, and hemodynamic events, cross-modal prediction uses a learned delay aligner that matches signals at the corresponding cardiac time. Because synchronized multi-sensor recordings are scarce, CardioState-JEPA first learns within-modality structure from abundant unimodal data and then uses paired data to align modalities in latent cardiac time. Evaluated as a frozen encoder across 25 downstream tasks spanning ECG, PPG, and PCG, our encoder improves average PPG classification by 8.2 AUROC points, PCG murmur detection by 18.8 AUROC points, and ECG classification by 15.5 AUROC points over the best self-supervised signal baseline and matches or exceeds cardiac models trained with privileged clinical text or supervised labels on several ECG benchmarks. These results establish that heterogeneous cardiac signals can mutually supervise a single foundation model of cardiac physiology.
Read the full abstract

All authors Hamza Shafiq, Hung Manh Pham, Bin Zhu, Pan Zhou, Jun Hu, Aaqib Saeed

Machine Learning

Diagnosing JEPA World Models with Action-Conditioned Predictive Consistency

Joint-embedding predictive architectures (JEPAs) learn world models that predict in a compact latent space rather than in pixels, reducing the pressure to model nuisance appearance. Yet this provides no guarantee against visual perturbations: they can still alter the encoded representation and affect subsequent action-conditioned predictions. Bisimulation captures this requirement precisely: two observations should be treated as the same state only when their action-conditioned consequences agree. Guided by this criterion, we introduce Action-Conditioned Predictive Consistency (ACPC), a diagnostic that measures how far a clean history and a visually perturbed view of it diverge after being rolled forward under the same action sequence. We prove that this divergence bounds the perturbation-induced change in multi-step prediction error and planner cost. Building on pairwise ACPC, we define two complementary measures: the Invariance Radius (IR) summarizes clean-perturbed rollout spread, while the Separation Rate (SR) checks whether different states remain distinguishable after rollout. Experiments on four visual control tasks show that pairwise ACPC predicts perturbation-induced prediction and cost changes. On LeWM, the IR-SR screen transfers across tasks, and the joint diagnostic remains informative under blur and resize. PLDM exhibits similar diagnostic trends under a different architecture.
Read the full abstract

All authors Guo An, Zijing Wu, Honghua Dong, Yuhao Yan, Zixuan Gui, Haochong Chen, Shanzhao Ruan, Xiang Wang, Yurong Ling, Qi Tian

Numerical Analysis · Cross-listed from math-ph

Transparent Boundary Conditions for the Heat Equation on Metric Graphs

We study transparent boundary conditions (TBCs) for the time-dependent heat equation in a branching, quasi-one-dimensional domain modeled as a metric star graph. By combining the classical concept of TBCs with the theory of partial differential equations (PDEs) on networks, we derive an exact vertex condition that ensures unobstructed thermal flow across junctions. Specifically,we derive a sum rule for the diffusion coefficients that eliminates thermal backflow at the vertex.We demonstrate the validity of this analytical model numerically using a Crank-Nicolson finite-difference method, which confirms smooth, unobstructed heat propagation through the network.These results provide a practical mathematical framework for tunable control and optimization of thermal diffusion in low-dimensional structures. In particular, combining the well-known concept of the TBCs and theory for PDE on metric graphs, we propose a mathematical model providing a control tool for thermal diffusion in networks.
Read the full abstract

Artificial Intelligence · Cross-listed from quant-ph

AutoQuREO: A Framework for Automated Quantum Resource Estimation and Optimization

As quantum computing progresses from proof-of-principle demonstrations toward practical utility, a significant impediment is the need to augment algorithmic feasibility with system-level optimization across heterogeneous hardware and software stacks. Quantum resource estimation (QRE) plays a central role in this transition, yet existing approaches remain largely compilation-heavy or domain-knowledge-guided symbolic annotations, and tightly coupled to long-term fault-tolerant assumptions, limiting their topical applicability. In this work, we introduce AutoQuREO, an Automated framework for full-stack Quantum Resource Estimation and Optimization. AutoQuREO is built around four core novelties: (i) a flexible, user-defined abstraction of the quantum computing stack; (ii) a modular library of reusable stack components enabling rapid full-stack prototyping; (iii) surrogate modeling of layer-wise resources via algorithmic profiling and neuro-symbolic learning; and (iv) integrated multi-objective optimization that embeds QRE directly into deployment pipelines. Together, these design choices enable AutoQuREO to serve as a digital twin for quantum computing stacks, supporting the tractable exploration of complex design spaces. We demonstrate the capabilities of AutoQuREO through representative co-design case studies, including early-fault-tolerant quantum algorithms, small error correction codes, gate decomposition and variational training of parametric quantum circuits. These examples illustrate how AutoQuREO enables systematic discovery of unexploited resource trade-offs that are computationally intractable or abstruse using existing QRE tools. AutoQuREO is positioned as a general-purpose platform for advancing quantum technology readiness.
Read the full abstract

All authors Harshkumar Oza, Aritra Sarkar, Syed Naqi Abbas, Rahul Bhowmick, Aryan Prakash, Prateek P Kulkarni, Krishna Kumar Sabapathy

Artificial Intelligence

Decomposition of Evidence, Contradiction, and Fragility in Perturbation Responses

Perturbation methods explain model decisions by measuring prediction changes under altered inputs, but response magnitude tells us only how much a model reacts, not what that reaction means. The same magnitude can support the final factual-counterfactual difference, oppose it, or arise strongly along the perturbation path yet vanish at the endpoint. We therefore track how the contrast develops as paired inputs are progressively revealed, using the final contrast to interpret the trajectory. We introduce DECAF (Decomposition of Evidence, Contradiction, And Fragility), which routes aligned, opposed, and endpoint-null responses into evidence E, contradiction C, and fragility F. The decomposition preserves ordinary magnitude exactly, Abs = E + C + F, and is unique under endpoint-relative axioms. Across controlled vision and tabular settings, the three components track independently measured behavior. In a 72-model ImageNet-9 audit, we compare cases with nearly identical response magnitude but different independently measured behaviors. The largest DECAF component agrees with an observed behavior in 96.4% of cases, compared with 35.0% for magnitude alone. Changing only the reveal path increases total response by nearly 80%, yet evidence barely changes while fragility grows by more than 4x. On FunnyBirds and ImageNet-1k, short forward-only DECAF trajectories outperform the tested general-purpose attribution baselines. On a 1B-scale DINOv2 model, a short trajectory matches a strong gradient-based baseline with 4.75x lower wall time and 2.36x lower peak memory.
Read the full abstract

Hardware Architecture

Dryas: A Reprogrammable Engine for High-Speed Interconnect Tracing and Analysis

The proliferation of heterogeneous components in modern computing systems has been accompanied by new higher bandwidth and lower latency interconnects. These interfaces and protocols are enormously complex and the process of developing, debugging, and analyzing FPGA-based implementations requires significant engineering work. Moreover, once a functional implementation is completed, optimization of the controller and associated software requires processing potentially hundreds of gigabytes of trace data. In this paper, we present Dryas, an open source tool for analyzing such an interconnect. We developed our tool, using minimal hardware resources, alongside an FPGA implementation of a very high speed, low latency (30~GiB/s, 200~ns) interconnect. With our run-time reprogrammable overlay engine we can inspect this interconnect to find rare, complex, or transient events even at full operation. This filtering engine is based on non-deterministic finite automata (NFAs), efficiently implemented using state transition elements (STEs), allowing us to trace events at a cache-line granularity. Moreover we can change the filters in less than a second, without reprogramming the FPGA or interfering with the running application. This data enables not only debugging the implementation of the interconnect itself, but analyzing the behavior of accelerated applications. We examine the mathematical basis for using NFAs and describe their implementation on a real coherent CPU-FPGA research platform. We then evaluate the scalability of Dryas for various size NFAs, followed by two different use cases: debugging FPGA implementation of the interconnect and analyzing cache behavior.
Read the full abstract

All authors Manuel Bröchin, Tom Kuchler, Michael Giardino, David Cock, Timothy Roscoe