Machine Learning · Cross-listed from quant-ph
A Quantum Roadmap for Softmax Attention: Exact Born-Rule Analogs for Softmax Attention on the Probability Simplex
Eric A. F. Reinhardt and Adam J. Hauser
The
attention mechanism forms the foundation of many modern AI models such as the
Transformer. In one subclass of problems where
attention is used, inputs and outputs are bound to the probability simplex so that all outputs sum to one. In this setting, softmax
attention admits an exact, component-by-component
quantum realization.
Attention scores are Hadamard-test statistics on block-encoded projections of amplitude-encoded inputs. The exponential softmax is the interior of a cosine-squared family generated by Born-rule measurement under an exact bijection, whose boundary expresses sparse
attention with exact zeros at finite parameter values. The softmax temperature is a repetition count where post-selected measurement rounds realize discretized inverse temperature exactly. Value aggregation is a deterministic column-loading channel that dilates the column-
stochastic value
matrix. The gated residual is the preparation angle of a single ancilla, with the additive identity at a mixing angle of $\pi$/2. Every learnable parameter is a rotation-gate angle. The composed layer is exact in the infinite-shot limit with one measure-and-reload step per
attention score; a fully-coherent variant is $\epsilon$-approximate via
quantum singular value transformation in the infinite depth limit. The algebraic core is machine-checked in Lean 4.
Read the full abstract ▾
Computation and Language
From Interpretability to Control: Insights from Six Years of the TrustNLP Workshop
Rahul Gupta, Abhinav Mohanty, Anaelia Ovalle et al. (+10 more)
The Workshop on Trustworthy Natural
Language Processing (TrustNLP), co-located with major ACL conferences since 2021, has grown from 8 proceedings papers to 41 over six editions, documenting a field-wide transition from post-hoc interpretability of static models to mechanistic understanding and proactive control of generative systems. We synthesize insights from all 144 proceedings papers, classifying them along six trust dimensions grounded in established frameworks (TrustLLM, DecodingTrust). We observe co-occurrences with capability emergence. The release of the first high-impact chat models activated all trust dimensions simultaneously, while subsequent model generations shifted focus toward truthfulness and safety
alignment. Analysis from the classification study reveals that truthfulness is the fastest-growing dimension (absent in 2021-2022, comprising 37% of papers by 2025-2026),
fairness remains the most consistent theme, and explainability exhibits a U-shaped trajectory; declining as post-hoc methods lost relevance but resurging in 2026 through mechanistic interpretability. A cross-venue comparison with ACL, NAACL, EACL, and EMNLP (~2K papers) in the same period shows that TrustNLP's topical distribution closely follows the field average. We identify four structural insights and conclude with actionable directions for the research community.
Read the full abstract ▾
All authors Rahul Gupta, Abhinav Mohanty, Anaelia Ovalle, Anil Ramakrishna, Anubrata Das, Apurv Verma, Jwala Dhamala, Ninareh Mehrabi, Tharindu Kumarage, Yada Pruksachatkun, Yang Trista Cao, Kai-Wei Chang, Aram Galstyan
Computer Vision
MultiModal Code-Switching: Interleaving Visual Objects into Language for Explicit Object-Level Alignment
Changhao Xiang, Shangyu Xing, Zhen Wu et al. (+2 more)
Existing
Multimodal Large
Language Models (MLLMs) predominantly rely on image-text pairs for modality
alignment pretraining, mapping global image representations to long textual descriptions. However, this image-level
alignment suffers from referential ambiguity: models struggle to infer the correspondences between multiple visual objects and textual entities from the global representation, leading to data inefficiency and suboptimal semantic grounding. To address this, we propose
MultiModal Code-Switching (MMCS), a novel pretraining paradigm that provides explicit object-level supervision. Inspired by the linguistic phenomenon of code-switching, MMCS interleaves
vision and
language by replacing textual entities with their corresponding visual objects, enforcing local
vision-
language grounding. We further develop a scalable data synthesis pipeline to generate a pretraining
dataset of 773K samples with accurate object-entity correspondences. Experiments show that MMCS is highly data-efficient: with only 50K samples, it matches or surpasses models trained on 600K image-text pairs. Furthermore, MMCS consistently improves visual grounding and perception capabilities across varying model scales.
Read the full abstract ▾
All authors Changhao Xiang, Shangyu Xing, Zhen Wu, Jianbing Zhang, Xinyu Dai
Software Engineering
Agentic Configuration Management (ACM): A Reference Configuration Model for Governed Agentic Systems
Audrey Quessada-Vial
Agentic systems are increasingly composed of heterogeneous
agents,
prompts, tools, models, skills, composite subsystems, policies, and execution workflows whose configurations evolve across frameworks and runtime environments. Existing LLMOps and AgentOps platforms support orchestration and observability but do not provide a common configuration-governance model for representing and governing these systems as coherent, versioned configurations. This paper introduces Agentic Configuration Management (ACM), a framework-independent governance and configuration reference model for heterogeneous agentic systems. ACM combines typed and independently versioned Agentic Configuration Items, immutable revisions and baselines, explicit configuration-runtime separation, lifecycle and assurance semantics, dependency-aware impact propagation, and runtime provenance. Heterogeneous native configurations are normalized through semantic projection into a canonical Configuration Graph on which common governance semantics operate. We provide a Python reference implementation with adapters for LangGraph, CrewAI, and the OpenAI
Agents SDK. The evaluation combines 27 governance scenarios with nine quantitative impact-propagation cases. For the evaluated configurations, the three frameworks yield governance-equivalent ACM representations and reproducible governance outcomes after projection. The impact semantics are formalized as monotone propagation over a finite
lattice, establishing convergence, termination, and uniqueness of the least fixed point above the initial impact valuation. These results provide evidence that common governance semantics can support reproducibility, auditability, dependency analysis, and interoperability across heterogeneous agentic execution abstractions within the evaluated scope.
Read the full abstract ▾
Information Retrieval
Role of Personality in Conversational Information Seeking
Abdisalam Abukar, Junchen Fu, Chengli Zhai et al. (+1 more)
Large
language models (LLMs) are increasingly used for information seeking, where users find, compare, and evaluate information through dialogue. In this role, the assistant does more than retrieve or generate content: it shapes how users articulate constraints, ask follow-up questions, verify claims, and decide when an answer is sufficient for action. Yet little is known about how user personality, assistant personality, and task context jointly influence these interactions. We examine personality as a controllable variable in conversational information seeking and study its effects on user behaviour and interaction quality. We conducted a controlled within-subject study in which assistant personality and task type were experimentally varied, while participant personality was measured using Big Five scores. Twenty-six participants each completed three information-seeking tasks under three assistant personality conditions: extraverted, conscientious, and neutral. Tasks covered exploratory travel planning, comparative smartphone shopping, and verification-sensitive health and diet information seeking. Data included conversation logs, behavioural traces, post-interaction questionnaires, an exit questionnaire, and Big Five measures. The assistant conditions were behaviourally distinct: the extraverted assistant produced longer turns, the conscientious assistant elicited higher user word share and more turns, and the neutral baseline fell between them. The strongest effect was a task-by-assistant interaction on trust and delegation, with preferred styles varying by task. No global winner emerged, but participants strongly preferred style choice or adaptation. These findings position assistant personality as a context-sensitive interactional design variable rather than a globally optimisable system property.
Read the full abstract ▾
All authors Abdisalam Abukar, Junchen Fu, Chengli Zhai, Joemon M. Jose
Data Structures and Algorithms
Matchings via Random Greedy Independent Set: A Simpler Algorithm and Analysis
Andrew McGregor
We show that a simple extension of the randomized greedy maximal independent set
algorithm yields a constant approximation for the maximum matching problem. The
algorithm is a simplification of an
algorithm used by Assadi et al. [JACM 2026] in the context of processing data streams in the dynamic setting where edges may be inserted and deleted. In contrast to the previous work, our analysis avoids consideration of fractional matchings and yields a significantly shorter and more direct proof of the approximation factor for the basic
algorithm.
Read the full abstract ▾
Machine Learning
Hierarchical Empirical-Bayes Naive Bayes: Minimax Smoothing and Calibration with AODE Extension
Nguyen Thai Anh, Truong Viet Vu, Tran Thien Thanh et al. (+2 more)
The Naive Bayes (NB) classifier remains a standard choice for categorical data, yet its widely used smoothing rules, such as Laplace, Lidstone, Krichevsky-Trofimov, and the $m$-estimate, all prescribe a fixed smoothing strength that ignores feature cardinality, sample size, and class imbalance, inducing a non-vanishing bias on modern high-cardinality tabular data. We propose hierarchical empirical-Bayes Naive Bayes (HEB-NB), in which each class-feature conditional probability is smoothed by a Dirichlet prior whose concentration is learned data-adaptively via Type-II maximum likelihood, enabling principled information sharing across classes while retaining closed-form inference. We further introduce HEB average one-dependence
estimators (HEB-AODE), showing that the adaptive smoothing transfers cleanly to structural relaxations of NB. Theoretically, we establish a non-asymptotic $\ell_1$ error bound for HEB-NB matching the empirical-distribution minimax rate plus a vanishing data-adaptive bias, together with a matching Laplace-tight lower bound that yields a finite-sample, risk-level strict separation from Laplace. We further derive a plug-in excess Bayes-risk bound via total-variation tensorization and a population top-1 expected calibration error (ECE) corollary. Empirically, across 31 UCI and OpenML
benchmarks, HEB-NB attains the best average Friedman rank on probabilistic metrics, with up to 22.1% log-loss reductions on high-cardinality
datasets and consistent improvements of HEB-AODE over vanilla AODE. Combining HEB-NB with mutual-information weighting reduces top-1 ECE by 41%-70%, demonstrating substantial gains in probabilistic accuracy and calibration.
Read the full abstract ▾
All authors Nguyen Thai Anh, Truong Viet Vu, Tran Thien Thanh, Vo Nguyen Quoc Bao, Ngo Hoang Tu
Information Theory · Cross-listed from quant-ph
Quantum Codes with Arbitrary Z-Rotation logical Gates and Applications to Fault-Tolerant Code Switching
Reza Dastbasteh, Ruben M. Otxoa, Pedro M. Crespo et al. (+1 more)
A technique for realizing a universal set of fault-tolerant
quantum operations is the code switching method, which leverages two
quantum codes with complementary sets of transversal gates. To date, the application of this technique has been largely limited to families of color codes supporting a logical $T$ gate. No analogous code switching protocols exist for many other prominent families, such as rotated surface codes, or for finer $Z$-rotation gates. In this work, we first utilize the doubling technique as a unified framework to construct a class of
quantum color codes encoding a single logical
qubit with an arbitrarily large minimum distance, enabling the transversal realization of arbitrary small logical $Z$-rotation gates. We investigate the structural properties of this code family, demonstrating that they improve upon the parameters of state-of-the-art triorthogonal codes, achieve lower
qubit overhead compared to certain known color codes, and admit single-shot decoding of $Z$-syndromes via meta-checks. Furthermore, we show that this framework extends beyond color codes; specifically, it enables the generation of $r$-orthogonal
quantum codes, $r \ge 2$, that inherit the local geometry of rotated surface codes. We then provide an overhead optimization protocol alongside several candidate codes tailored for realizing logical $Z$-rotation gates within rotated surface codes. Finally, we extend the fault-tolerant code switching protocol based on transversal CNOT gates to incorporate fault-tolerant realization of $Z$-rotation gates at any level of the Clifford hierarchy for geometries compatible with rotated surface codes. We present the first demonstration of fault-tolerant magic state preparation by means of code switching within a distance-three rotated surface code using a total footprint of only 45 physical
qubits, and evaluate its performance through a simulation.
Read the full abstract ▾
All authors Reza Dastbasteh, Ruben M. Otxoa, Pedro M. Crespo, Josu Etxezarreta Martinez
Computers and Society
Mediatised Participation: Citizen Journalism and the Decline in User-Generated Content in Online News Media
Simón Peña-Fernández, Ainara Larrondo-Ureta and Irati Agirreazkuenaga
The second generation of web tools shook the journalist profession approximately two decades ago with the proactive incorporation of audiences into the media. Citizen journalism and user-generated content arose as an object of interest due to the democratising value of participation attributed to them, with empowered citizens who could emulate the professional and institutional practises of journalists. However, difficulties soon came to the surface, and audience participation in news media began to be limited. Within this context, this article conducts a critical review of studies on audience participation in news media based on a systematic literature review. The results indicate that, in general, audiences showed low interest in the creation of informative content and that their participation has grown increasingly problematic. In addition, journalists are reticent as they defend their professional role above all else, while company strategies have prioritised making participation profitable. For this reason, the idea of citizen journalism that offers user-created content through the media appears to be a thing of the past, with many characteristics that could define it as a failed innovation. Therefore, the text concludes that audience participation in the media could be defined as mediatised participation.
Read the full abstract ▾
Computational Complexity
New Lower and Upper Bounds for the Grothendieck Constant
Rahul Saha, Alan Li, Anton Xue et al. (+4 more)
We establish new bounds on the Grothendieck constant $K_G$: \[ \frac{6\pi}{11} \le K_G \le \frac{\pi}{2\log(1+\sqrt2)} - 10^{-4}. \] Methodologically, our lower bound approach differs from previous works by establishing limitations on the asymptotically optimal Krivine schemes, rather than giving explicit constructions of gap instances. Our upper bound is obtained by proposing and analyzing the first asymptotic construction of rounding schemes, whereas previous works only consider low-dimensional schemes. Together, these bounds determine the previously unknown tenths digit of $K_G$ to be $7$. The bounds were discovered by a long-running collaborative effort of humans and a long-horizon AI research system that we engineered.
Read the full abstract ▾
All authors Rahul Saha, Alan Li, Anton Xue, Swarat Chaudhuri, Adam Klivans, Pravesh K Kothari, Raghu Meka
Machine Learning · Cross-listed from stat.ML
Conditional Independence Tests for Constraint-Based Causal Discovery: A Survey
Pavel Averin, Theodoros Moysiadis and Ioannis Katakis
Conditional Independence (CI) tests are the statistical engine of constraint-based causal discovery: in
algorithms such as PC (Peter-Clark) and FCI (Fast Causal Inference), skeleton
pruning and key orientations follow directly from CI decisions. This survey reviews CI testing with emphasis on assumptions, robustness, and scalability in high-dimensional and mixed-type settings common in biomedical domains. The survey organizes widely used CI methods into six families: partial-correlation, contingency-table, regression, nearest-neighbor, kernel, and machine-learning-based. Special emphasis is provided on the robustness layers that address the limitations of these families. For each family, the survey examines when CI decisions reflect the data-generating distribution and when they fail. By this, we link test-level properties, including power decay with conditioning set size and asymmetric type I/II error consequences, to graph-level errors in skeleton recovery and v-structure orientation. The survey also compares adoption across major R and Python libraries and summarizes open challenges, including mixed-type CI testing without discretization, small-sample error control, and strategies for improving scalability of CI-testing.
Read the full abstract ▾
Hardware Architecture
On the Sensitivity to Errors in Homomorphic Computing: Single Transient Bit-flip Client-side Error Characterization
Matías Mazzanti, Vattana Chan, Karthik Swaminathan et al. (+3 more)
Homomorphic Encryption (HE) enables computation on encrypted data without decryption and is a key primitive for
privacy-preserving computation in sensitive domains such as healthcare, finance, and government. Its security relies on noise injection, which introduces intrinsic error sensitivity and raises concerns about the fault tolerance of HE systems, as hardware- and software-induced faults can evade traditional detection mechanisms and lead to silent data corruption. In this work, we analyze the sensitivity of HE to bit-level faults, focusing on the CKKS (Cheon–Kim–Kim–Song) scheme widely used for approximate arithmetic in AI and machine learning workloads. We identify homomorphic multiplication as the most error-sensitive operation in practical HE pipelines and characterize how faults propagate and amplify through it, exposing a critical robustness vulnerability and motivating the need for more resilient HE deployments.
Read the full abstract ▾
All authors Matías Mazzanti, Vattana Chan, Karthik Swaminathan, Augusto Vega, Esteban Mocskos, Radha Venkatagiri
Machine Learning
DACRI: Decision-Aware Causal Intervention Ranking for Critical Supply Chains
Shiqi Huang, Jiani He, Dingyan Shang et al. (+4 more)
Detecting or attributing a supply-chain disruption is not the same as selecting the intervention that maximizes recoverable net value. We present CriticalSCM-Bench v1, a controlled synthetic
benchmark with causal ground truth, paired factual/counterfactual rollouts, and an explicit net-value objective. Relative to a full-information train-selected static
benchmark, LambdaMART improves median normalized net value by 5.7–16.2%, with paired statistical support on the
semiconductor and critical-material archetypes but not on digital infrastructure. On digital infrastructure, a domain-informed constant-buffer policy remains stronger, showing that greater model complexity is not uniformly justified. Across partial and delayed settings, LambdaMART retains 33–75% of full-clamp value. Stress tests further show that intervention fidelity, timing, cost, and held-out disruptions can alter policy ordering. Critical materials show the weakest out-of-distribution retention. Separately, a guarded explanation study over 540 generations preserves every fixed intervention decision after deterministic validation and template fallback, although exact wording remains unstable. Within this controlled setting, the results identify regimes in which adaptive ranking adds value and those in which simpler structural policies remain preferable.
Read the full abstract ▾
All authors Shiqi Huang, Jiani He, Dingyan Shang, Yihua Xu, Jize Li, Yan Lyu, Lakshmi Muraleedharan Nair
Cryptography and Security
Strategies to Avoid Illegal Data Access
Muhammad Mubeen, Arslan Bisharat and Giri Anandhi
For companies of all sizes, data security is a top priority. The chance of unauthorized data access increases as technology develops. To prevent unwanted access to their data, businesses must be proactive. This study examines technology solutions, personnel training, and policy enforcement as methods to prevent unauthorized data access. Data may be protected from illegal access using technological solutions like firewalls, intrusion detection systems, and encryption. Intrusion detection systems notify the administrator when suspicious behavior is found, while firewalls serve as a protective border between the internal network and the internet. Even if data is intercepted, encryption makes sure it is safe. Another effective method of avoiding unauthorized data access is employee education. Employees must be taught how to spot hazards like phishing emails and shady websites and react to them. Additionally, they should be taught the right way to utilize passwords and other security precautions. To secure data, organizations should create and implement policies. Policies should set out appropriate data and system use guidelines and provide repercussions for noncompliance. Policies should be evaluated regularly to ensure that they are current and useful. Businesses may prevent unwanted access to their data by installing technology solutions, training staff, and enforcing regulations. Organizations may reduce data breach risk and maintain regulatory compliance by taking these precautions. ...
Read the full abstract ▾
Distributed Computing
Scheduling Mixed RL Rollouts Beyond Prefix Locality
Zetao Hong, Song Yuan, Yuanhao Ding et al. (+4 more)
Modern reinforcement learning (RL) post-training pipelines for large
language models (LLMs) increasingly combine rollout workloads across multiple domains and feedback paradigms. Prefix-aware routing improves inference efficiency through cache reuse and load balancing, but it does not control how heterogeneous rollout sessions compete for KV-cache capacity. When reinforcement learning with verifiable rewards (RLVR), reinforcement learning from human feedback (RLHF), and agentic rollouts share an asynchronous inference service, their distinct sequence structures, interaction patterns, and KV-residency times create substantially different serving demands. Rollout scheduling must account for this heterogeneity without distorting the workload mixture specified by the trainer. We present MISA-T, a routing-layer admission policy for mixed rollout serving. MISA-T combines adaptive session admission, workload-aware KV-capacity allocation, and residency-time-aware KV accounting. In rollout-only ablations on Step3.7 and Qwen3.6-35B-A3B, MISA-T improves rollout throughput over a sweep-tuned cache-aware vLLM Router by 53.3% and 43.6%, respectively, while maintaining high prefix-cache hit rates. In a matched 50-iteration Step3.7 experiment, it increases rollout throughput by 35.6% and reduces mean iteration time by 22.8%, while keeping the consumed workload mixture close to the trainer target and achieving comparable task scores.
Read the full abstract ▾
All authors Zetao Hong, Song Yuan, Yuanhao Ding, Yibo Zhu, Daxin Jiang, Zhibin Wang, Chen Tian
Information Theory · Cross-listed from quant-ph
Optimal von Neumann Entropy Estimation
Ziyun Chen, Minbo Gao and Qisheng Wang
We show that the von Neumann
entropy of an unknown $d$-dimensional
quantum state $\rho$ can be estimated to within additive error $\varepsilon$ using \[ \Theta\!\left(\frac{d^2}{\varepsilon\log^2(d)(1+\varepsilon\log^2(d))} + \frac{\log^2(d)}{\varepsilon^2}\right) \] samples of $\rho$, matching the lower bound due to Wang (2026). Our
estimator is based on a high-degree approximation
polynomial and the main technical challenge is bounding its variance despite the large
polynomial degree. We overcome this using a
quantum Hoeffding decomposition, which preserves cancellations between different
polynomial terms and reduces the analysis to derivative bounds for the approximation
polynomial.
Read the full abstract ▾
Computer Vision
CausalSplat: Towards Comprehensive Hierarchical Reasoning in 3D Gaussian Splatting
Jiayu Ding, Meilu Song, Yun Chen et al. (+2 more)
While 3D Gaussian Splatting (3DGS) has advanced open vocabulary scene understanding, existing methods remain confined to explicit queries. They struggle to interpret implicit intents, complex spatial constraints, and commonsense
reasoning required for practical
embodied interactions. To address this gap, we introduce the task of
reasoning 3D Gaussian
segmentation and construct two
benchmarks, Causal-LERF and Causal-ScanNet. These
benchmarks systematically evaluate commonsense, spatial, affordance, and counterfactual
reasoning. Evaluations reveal that current state of the art methods perform poorly on these
reasoning challenges. Therefore, we propose CausalSplat, a framework that integrates
vision-
language models with 3D scene graphs to disentangle explicit structural perception from implicit logical inference. Extensive experiments demonstrate that CausalSplat achieves state of the art performance on our
reasoning benchmarks while showing strong generalizability on standard referring and open vocabulary 3D
segmentation tasks. Project Page: https://jiayuding031020.github.io/CausalSplat
Read the full abstract ▾
All authors Jiayu Ding, Meilu Song, Yun Chen, Wei Gao, Ge Li
Computer Vision
PRMU: A Corpus-Free Benchmark for Person-Centric Knowledge Unlearning in Multimodal Large Language Models
Huafeng Chen, Yueming Lyu, Ziyuan Chen et al. (+4 more)
Multimodal large
language models (MLLMs) have demonstrated remarkable capabilities in storing and recalling rich person-related knowledge, raising increasing concerns about reliable knowledge removal. However, existing machine unlearning approaches for MLLMs typically assume access to original forget and retain corpora, which are often unavailable in realistic deletion scenarios. To address this limitation, we introduce PRMU, a
benchmark for evaluating corpus-free
multimodal unlearning under realistic person-centric deletion requests. PRMU focuses on naturally acquired person-related knowledge and evaluates whether models can remove target knowledge while preserving related knowledge through diverse textual and visual probes, including
adversarial evaluation and fine-grained locality analysis. To facilitate research in this setting, we further introduce Similarity-Gated Projection Editing (SGPE), a lightweight corpus-free unlearning baseline with knowledge displacement, protected parameter-space editing, and locality-aware
multimodal control. Extensive experiments on representative MLLMs reveal that existing unlearning methods often suffer from unfavorable forgetting-locality trade-offs, with significant locality degradation under aggressive forgetting settings, and remain vulnerable to
multimodal knowledge reactivation. Meanwhile, SGPE provides a competitive trade-off between target forgetting, locality preservation, and general
multimodal utility. We hope PRMU can facilitate future research toward realistic and scalable
multimodal machine unlearning. Code and
dataset will be released at https://github.com/2231122/PRMU.
Read the full abstract ▾
All authors Huafeng Chen, Yueming Lyu, Ziyuan Chen, Wenda Tan, Chenyang Si, Liucheng Guo, Caifeng Shan
Hardware Architecture
When and Where Faults Matter: A Study of Transient Errors in CKKS Multiplication
Vattana Chan, Matías Mazzanti, Karthik Swaminathan et al. (+3 more)
Homomorphic Encryption (HE) is a
privacy-preserving encryption paradigm that enables computation directly on encrypted data without requiring decryption. In this paper, we study errors in fully homomorphic encryption (FHE) computations, with a particular focus on server-side homomorphic multiplication in the unoptimized CKKS (Cheon–Kim–Kim–Song) scheme. We show that both the timing and the location of errors in the ciphertext components \(c_0\) and \(c_1\) have a significant impact on the correctness of the final FHE output.
Read the full abstract ▾
All authors Vattana Chan, Matías Mazzanti, Karthik Swaminathan, Augusto Vega, Esteban Mocskos, Radha Venkatagiri
Computation and Language
The Illusion of Cross-Lingual Safety in Low-Resource Languages
Abigail Oppong, P Sam Sahil, Tadesse Destaw Belay et al. (+12 more)
Safety
alignment in large
language models (LLMs) is largely developed in English, assuming these safeguards generalize across
multilingual settings. However, this assumption remains underexplored and exposes a vulnerability in low-resource
languages. We investigate cross-lingual safety transfer in four African
languages, Twi, Hausa, Amharic, and Swahili, using LoDNA, a new safety
dataset that pairs literal translations with culturally localized
prompts. To move beyond generation-based evaluation, we propose a latent
geometric framework that probes hidden-state refusal representations in LLMs. Our experimental results show that cross-lingual safety transfer is severely limited; harmful
prompts retain less than 10% of the English refusal signal across most
language-model pairs. Literal and localized
prompts are semantically aligned (cosine 0.95-0.996) but drift across layers, suggesting models encode the concepts without routing them to safety mechanisms. These findings demonstrate that current
multilingual safety
alignment is superficial, providing strong evidence against the assumption of a universal,
language-agnostic harm
manifold within the specific low-resource
languages studied. Warning: This paper contains example data that may be offensive or harmful.
Read the full abstract ▾
All authors Abigail Oppong, P Sam Sahil, Tadesse Destaw Belay, Maryam Ibrahim Mukhtar, Esmael Ahmed Abdu, Tassallah Abdullahi, Jessica Oparebea, Saminu Mohammad Aliyu, Idris Abdulmumin, Abubakar Juma Chilala, Nicholaus Dismas Ladislaus, Alfred Malengo Kondoro, Lemofouet Valdini Douglace, Shamsuddeen Hassan Muhammad, Seid Muhie Yimam
Machine Learning
A Recommendation System Approach for Interference-Robust Sensor Subset Selection
Kaan Buyukkalayci, Kyle Pak, Merve Karakas et al. (+1 more)
This paper develops a method for sensor-subset selection for tracking. Prior work showed that low-cost acoustic Received Signal Strength Indicator (RSSI) measurements can be used to recommend subsets of sensor nodes whose expensive sensing modalities, such as cameras, can achieve high tracking accuracy. While efficient, RSSI-based approaches are challenged by acoustic interference. We propose a recommendation-system-inspired framework that instead leverages frequency-band acoustic features and a Two-Tower Multi-Layer Perceptron (MLP) architecture to efficiently score candidate sensor subsets. Experimental results on outdoor vehicle-tracking deployments show that the proposed method can improve accuracy by around 20% over the RSSI baseline while maintaining the low computational overhead required for real-time selective sensing.
Read the full abstract ▾
All authors Kaan Buyukkalayci, Kyle Pak, Merve Karakas, Christina Fragouli
Computer Vision
SAR2Agri: Learning SAR Intensity Representations for Agricultural Monitoring
Moti Rattan Gupta and Anupam Sobti
Agricultural monitoring faces unique challenges, arising from the landscape's complex temporal, phenological, and
climate dynamics, yet monitoring them is critical for ensuring food security. Synthetic Aperture Radar (SAR) satellites offer all-weather day-night imaging capability supporting key monitoring tasks including crop type mapping, yield prediction and phenological event detection. Existing
multimodal remote sensing foundation models including TerraMind and CopernicusFM learn SAR representations by grounding them in optical imagery using joint encoding and contrastive learning techniques, while SAR-specific foundation models such as SAR-JEPA, SARMAE, and SAR-W-MixMAE primarily focus on target detection, flood mapping, and land cover classification applications. Recent work has introduced phenology inspired temporal pretext tasks with optical imagery which has shown strong performance on agricultural downstream tasks. In this work, we propose the first self-supervised learning pipeline focused on using only SAR intensity imagery for agricultural applications. We improve the temporal pretext tasks through masking and curriculum learning to enhance the pretraining pipeline's ability to capture phenological features from SAR. On the SICKLE
benchmark, our final model achieves 84.9% IoU on crop type mapping, outperforming optical baselines (by 15.3 pt) and existing SAR baselines (by 2.2 pt), demonstrating the effectiveness of our proposed pipeline for pretraining SAR intensity encoders for agricultural monitoring.
Read the full abstract ▾
Systems and Control · Cross-listed from math.OC
Certificate-based Synthesis of Coordinated Droop Control for Heterogeneous Radial Distribution Networks
Georgios Pantazis and Michelle Chong
Voltage certificates for droop-controlled radial distribution networks are often constructed from worst-case quantities. In heterogeneous radial networks, this approach can hide where voltage deviations are most likely. Furthermore, such certificates become increasingly conservative as the network sensitivities accumulate. Leveraging the structure of the linearized DistFlow model and slope-restricted droop controllers, we derive tighter deterministic voltage certificates that retain heterogeneous network, disturbance and inverter characteristics of each bus. The certificates reveal the buses and local limitations that dominate certified voltage performance. Worst-case bounds are recovered as a special case. Although tighter, the heterogeneous certificates can still deteriorate downstream due to the network structure. To address this, we leverage our certificates as design variables for coordinated voltage control. Specifically, we develop a virtual droop architecture with coordination and affine feedforward compensation to reshape the effective voltage sensitivity, and formulate a linear program that jointly synthesizes the controller and minimizes its heterogeneous voltage certificates under operational, communication and inverter placement constraints. The resulting controller guarantees the voltage and inverter bounds for all admissible disturbances. Evaluation on two network
benchmarks, a five-customer residential feeder and a 26-customer rural network comprising four feeders, demonstrates improved certificates and voltage behaviour under inverter limits.
Read the full abstract ▾
Computation and Language
Attention-Path Fragility as an Uncertainty Signal in Large Language Models
Minsoo Kim, Sungyoung Ji, Kisung Moon et al. (+1 more)
We propose that a model's uncertainty about a
token is reflected not only in the breadth of its output distribution but also in whether a confident prediction is fragile under perturbation of its
attention pathways. We instantiate this as ASMI (
Attention-Subnetwork Mutual Information), a training-free
estimator that masks
attention heads and measures the BALD mutual information among the resulting subnetworks, with a semantic-agreement kernel to discount surface-form disagreement. The signal is not a restatement of output confidence: on grounded QA an out-of-fold test shows it adds error-predictive information beyond single-pass confidence and
entropy, concentrated in confident-but-fragile predictions, where acting on it roughly halves the retained error of a confidence filter. The distinctness is regime-graded, so ASMI predicts its own domain of applicability, strong where answers are routed through provided context and bounded by design where they are recalled from parametric knowledge. Sem-ASMI reads the signal from a single greedy response, without the
stochastic generations the strongest baselines require, and ties or beats Semantic
Entropy on ten of the twelve grounded
benchmark-backbone settings. Across the same twelve settings, the best ASMI variant, typically the adaptive one reusing the ten samples already drawn for the baselines, ties or leads the strongest baseline in eight, significantly in three under a paired test. On parametric QA all variants revert to or below the zero-cost MSP baseline, exactly as predicted, and the estimates are near-deterministic across reruns. A head-level analysis shows that what tracks this boundary is not the presence of head-level fragility but whether that fragility couples to errors.
Read the full abstract ▾
All authors Minsoo Kim, Sungyoung Ji, Kisung Moon, Ilyong Yoon
Artificial Intelligence
sLTN: Structural Logic Tensor Networks
Davide Rinaldi and Luciano Serafini
Logic Tensor Networks (LTN) provide a neurosymbolic framework in which first-order logic is interpreted through tensor operations, enabling logical constraints to be integrated with differentiable learning. However, the original formulation of LTN is primarily suited to data represented as flat collections of individuals, and does not explicitly capture structural organization such as temporal order, sequential position, or graph connectivity. We introduce sLTN, an extension of LTN that makes structural dimensions first-class elements of the
language. Structural dimensions represent named tensor axes associated with domain-specific organization, such as time steps, sequence positions, or graph nodes. They can be quantified explicitly, related through structural relations, and used to express temporal, sequential, and relational constraints directly at the logical level. We formalize the syntax and fuzzy tensor semantics of sLTN and show that, in the absence of structural dimensions, the framework recovers the original LTN semantics as a special case. We further describe a PyTorch implementation based on a declarative signature, formula parsing, and tensorial interpretation. The framework is illustrated on representative temporal and sequential
reasoning examples. This paper serves as a companion to the sltn library, available at https://github.com/logictensornetworks/sltn.
Read the full abstract ▾
Computer Vision
Is There Really a Camouflaged Object? Towards Realistic Camouflaged Object Detection
Huafeng Chen, Yueming Lyu, Chenyang Si et al. (+3 more)
Camouflaged object detection (COD) aims to segment objects that are visually concealed in their surroundings and has attracted increasing
attention in recent years. However, most existing COD methods are developed under a closed-world assumption, where each input image is assumed to contain a camouflaged object. This assumption ignores realistic scenarios with pure backgrounds or non-camouflaged objects, causing existing models to produce severe false positives when deployed in open-world environments. To address this limitation, we propose OPC16K, a large-scale
benchmark for realistic COD. OPC16K contains 16,245 images from 14 sources and is carefully organized into camouflaged-object images, pure background images, and non-camouflaged-object images, enabling comprehensive evaluation of both
segmentation quality and negative-sample rejection. Based on this
benchmark, we further propose OPCNet, a presence-aware camouflage network that reformulates COD from a pure
segmentation task into a joint problem of object localization and camouflage existence
reasoning. Specifically, OPCNet introduces hierarchical existence
reasoning to distinguish CO, BG, and NOCOD scenarios, similarity-aware camouflage relation modeling to capture foreground-background camouflage cues, and existence-aware feature refinement to regulate
segmentation features with existence predictions. Extensive experiments on OPC16K demonstrate that OPCNet achieves superior performance under the proposed realistic COD evaluation protocol, significantly reducing false positives on negative samples while maintaining accurate camouflaged-object
segmentation. Code and
dataset will be released at https://github.com/2231122/OPCOD.
Read the full abstract ▾
All authors Huafeng Chen, Yueming Lyu, Chenyang Si, Wende Tan, Liucheng Guo, Caifeng Shan
Computer Vision
AlbumentationsX: One Augmentation Pipeline for Images and Related Annotations
Vladimir Iglovikov
Augmentation can corrupt a training example when an image and its annotations receive different random changes. A crop must use the same coordinates for the image, mask, boxes, keypoints, stereo views,
video frames, or volume. Code paths that choose these values separately can silently misalign the data. AlbumentationsX keeps the transform list, probabilities, annotation settings, and random seed in one Compose object. Each call chooses random values once and applies them to every supported part of the training example. The library keeps each object's mask, box, and label together and lets projects add their own transforms. It can also save the pipeline definition, show what happened in one call, and run that call again. The examples place Compose after files have been decoded into arrays and before PyTorch groups examples into a batch. AlbumentationsX executes the declared transforms. Practitioners still decide whether a flip, crop, color change, or other operation preserves the correct label for their task.
Read the full abstract ▾
Numerical Analysis
How different is rational approximation from piecewise polynomial approximation?
Daan Huybrechs
The first aim of this paper is to show that there is merit to the question posed in the title. Indeed, for certain function classes, approximation by rational functions and by piecewise
polynomials are surprisingly similar. Between these two,
polynomials are more widely studied and more widely used. One setting in which both approaches are equally well understood, and equally practical in use, is that of approximating univariate functions with point singularities. In this context we can fully address the question. We review classical literature on the topic which shows that both approaches do indeed achieve similar convergence rates. However, rational approximations come with significantly smaller constants. Owing to recent advances in practical rational approximation, we can augment the discussion with a comparison of modern numerical techniques that achieve the optimal rates. We end by showing numerically that the difference becomes even more pronounced in several variables.
Read the full abstract ▾
Hardware Architecture
You Only Charge Once 2.0 : A End-to-End Analog Computing-in-Memory Architecture with Reconfigurable Switched Capacitors
Zihao Xuan, Yewen Li, Jia Chen et al. (+3 more)
Analog Computing-in-
Memory (ACiM) accelerates deep neural networks by keeping weights inside
memory arrays and executing dot products in the analog domain. However, modern ACiM accelerators are often limited by the "ADC wall": analog-to-digital converters consume a large fraction of energy and area, while bit-sliced execution repeatedly invokes these converters. Existing designs reduce this cost with low-resolution readout or time multiplexing, but they either lose output fidelity or introduce serialization overhead. Charge-CIM addresses this bottleneck by using switched-capacitor charge redistribution as a unified computing and conversion substrate. The same capacitor fabric performs input conversion, analog MAC, weighted shift-and-add, and readout quantization, reducing both standalone converter overhead and intermediate ADC invocations. A differential readout path further combines paired partial sums during ADC quantization, providing a highly compact and energy-efficient solution for array integration. With dataflow architecture support, we evaluated Charge-CIM on a suite of DNN
benchmarks, from CNNs to
Transformer models, and experimental results show that Charge-CIM reduces ADC energy by 91.7% under our evaluation setup and improves energy efficiency by 2.7x and throughput by 2.0x compared to the state-of-the-art charge-domain CIM accelerator.
Read the full abstract ▾
All authors Zihao Xuan, Yewen Li, Jia Chen, Wei Xuan, Xiao Huo, Fengbin Tu
Machine Learning
Two-stage Odd Residual Flows for Mean-Preserving Probabilistic Time Series Forecasting
Kiran Madhusudhanan, Christian Klötergens, Lars Schmidt-Thieme et al. (+1 more)
Probabilistic
forecasting plays an essential role in risk-sensitive decision-making, particularly in long-horizon settings. However, existing approaches often face a fundamental trade-off between distributional flexibility and accurate mean prediction. Traditional parametric methods, such as Mean Variance Estimation (MVE), can suffer from degraded point accuracy when trained under joint Negative Log-Likelihood (NLL) objectives, while modern-flexible generative models, including Normalizing Flows and
Diffusion Models, typically rely on costly Monte Carlo sampling and may yield suboptimal mean estimates. To address this limitation, we propose Two-stage Odd Residual Flows (TORF), a framework that decouples mean
forecasting from uncertainty estimation. In the first stage, a pre-trained deterministic model is used to produce an accurate mean prediction. In the second stage, a Restricted Normalizing Flow, with strictly odd functions learns flexible residual distributions around the point forecast, guaranteeing mean preservation from the first stage without sampling. Experiments show that TORF achieves state-of-the-art deterministic accuracy (NMAE) while providing strong density estimation performance (CRPS) on short and long-horizon
forecasting.
Read the full abstract ▾
All authors Kiran Madhusudhanan, Christian Klötergens, Lars Schmidt-Thieme, Vijaya Krishna Yalavarthi