# Research Papers — 2026-01 · bundle page 62 of 189 > Full-text markdown of papers #3051–3100 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-01/page/62 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/61.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/63.md) ### Implementing Tensor Logic: Unifying Datalog and Neural Reasoning via Tensor Contraction Authors: Swapn Shah, Wlodek Zadrozny Published: 2026-01-23 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.17188v1 The unification of symbolic reasoning and neural networks remains a central challenge in artificial intelligence. Symbolic systems offer reliability and interpretability but lack scalability, while neural networks provide learning capabilities but sacrifice transparency. Tensor Logic, proposed by Domingos, suggests that logical rules and Einstein summation are mathematically equivalent, offering a principled path toward unification. This paper provides empirical validation of this framework through three experiments. First, we demonstrate the equivalence between recursive Datalog rules and iterative tensor contractions by computing the transitive closure of a biblical genealogy graph containing 1,972 individuals and 1,727 parent-child relationships, converging in 74 iterations to discover 33,945 ancestor relationships. Second, we implement reasoning in embedding space by training a neural network with learnable transformation matrices, demonstrating successful zero-shot compositional inference on held-out queries. Third, we validate the Tensor Logic superposition construction on FB15k-237, a large-scale knowledge graph with 14,541 entities and 237 relations. Using Domingos's relation matrix formulation $R_r = E^\top A_r E$, we achieve MRR of 0.3068 on standard link prediction and MRR of 0.3346 on a compositional reasoning benchmark where direct edges are removed during training, demonstrating that matrix composition enables multi-hop inference without direct training examples. --- ### High-Rate Quantized Matrix Multiplication I Authors: Or Ordentlich, Yury Polyanskiy Published: 2026-01-23 Categories: cs.IT, cs.AI arXiv: https://arxiv.org/abs/2601.17187v2 This paper investigates the problem of quantized matrix multiplication (MatMul), which has become crucial for the efficient deployment of large language models (LLMs). We consider a Generic MatMul setting, where both matrices must be quantized (weight+activation quantization) without specific apriori (calibration) statistical information about the factors. We review the fundamental information-theoretic tradeoff between quantization rate and distortion (high-rate theory), and contrast those with the performance of popular quantization schemes (absmax INT and floating-point (FP)), for which we also derive accurate heuristic approximations. Part II of this paper studies the weight-only quantization setup where second-order statistics of the activation matrices are available at the encoder. --- ### LGDWT-GS: Local and Global Discrete Wavelet-Regularized 3D Gaussian Splatting for Sparse-View Scene Reconstruction Authors: Shima Salehi, Atharva Agashe, Andrew J. McFarland, Joshua Peeples Published: 2026-01-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17185v2 We propose a new method for few-shot 3D reconstruction that integrates global and local frequency regularization to stabilize geometry and preserve fine details under sparse-view conditions, addressing a key limitation of existing 3D Gaussian Splatting (3DGS) models. We also introduce a new multispectral greenhouse dataset containing four spectral bands captured from diverse plant species under controlled conditions. Alongside the dataset, we release an open-source benchmarking package that defines standardized few-shot reconstruction protocols for evaluating 3DGS-based methods. Experiments on our multispectral dataset, as well as standard benchmarks, demonstrate that the proposed method achieves sharper, more stable, and spectrally consistent reconstructions than existing baselines. The dataset and code for this work are publicly available --- ### FASTR: Reimagining FASTQ via Compact Image-inspired Representation Authors: Adrian Tkachenko, Sepehr Salem, Ayotomiwa Ezekiel Adeniyi, Zulal Bingol, Mohammed Nayeem Uddin, Akshat Prasanna, Alexander Zelikovsky, Serghei Mangul, Can Alkan, Mohammed Alser Published: 2026-01-23 Categories: q-bio.GN, cs.LG arXiv: https://arxiv.org/abs/2601.17184v1 Motivation: High-throughput sequencing (HTS) enables population-scale genomics but generates massive datasets, creating bottlenecks in storage, transfer, and analysis. FASTQ, the standard format for over two decades, stores one byte per base and one byte per quality score, leading to inefficient I/O, high storage costs, and redundancy. Existing compression tools can mitigate some issues, but often introduce costly decompression or complex dependency issues. Results: We introduce FASTR, a lossless, computation-native successor to FASTQ that encodes each nucleotide together with its base quality score into a single 8-bit value. FASTR reduces file size by at least 2x while remaining fully reversible and directly usable for downstream analyses. Applying general-purpose compression tools on FASTR consistently yields higher compression ratios, 2.47, 3.64, and 4.8x faster compression, and 2.34, 1.96, 1.75x faster decompression than on FASTQ across Illumina, HiFi, and ONT reads. FASTR is machine-learning-ready, allowing reads to be consumed directly as numerical vectors or image-like representations. We provide a highly parallel software ecosystem for FASTQ-FASTR conversion and show that FASTR integrates with existing tools, such as minimap2, with minimal interface changes and no performance overhead. By eliminating decompression costs and reducing data movement, FASTR lays the foundation for scalable genomics analyses and real-time sequencing workflows. Availability and Implementation: https://github.com/ALSER-Lab/FASTR --- ### Federated Proximal Optimization for Privacy-Preserving Heart Disease Prediction: A Controlled Simulation Study on Non-IID Clinical Data Authors: Farzam Asad, Junaid Saif Khan, Maria Tariq, Sundus Munir, Muhammad Adnan Khan Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17183v1 Healthcare institutions have access to valuable patient data that could be of great help in the development of improved diagnostic models, but privacy regulations like HIPAA and GDPR prevent hospitals from directly sharing data with one another. Federated Learning offers a way out to this problem by facilitating collaborative model training without having the raw patient data centralized. However, clinical datasets intrinsically have non-IID (non-independent and identically distributed) features brought about by demographic disparity and diversity in disease prevalence and institutional practices. This paper presents a comprehensive simulation research of Federated Proximal Optimization (FedProx) for Heart Disease prediction based on UCI Heart Disease dataset. We generate realistic non-IID data partitions by simulating four heterogeneous hospital clients from the Cleveland Clinic dataset (303 patients), by inducing statistical heterogeneity by demographic-based stratification. Our experimental results show that FedProx with proximal parameter mu=0.05 achieves 85.00% accuracy, which is better than both centralized learning (83.33%) and isolated local models (78.45% average) without revealing patient privacy. Through generous sheer ablation studies with statistical validation on 50 independent runs we demonstrate that proximal regularization is effective in curbing client drift in heterogeneous environments. This proof-of-concept research offers algorithmic insights and practical deployment guidelines for real-world federated healthcare systems, and thus, our results are directly transferable to hospital IT-administrators, implementing privacy-preserving collaborative learning. --- ### Systematicity between Forms and Meanings across Languages Supports Efficient Communication Authors: Doreen Osmelak, Yang Xu, Michael Hahn, Kate McCurdy Published: 2026-01-23 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17181v1 Languages vary widely in how meanings map to word forms. These mappings have been found to support efficient communication; however, this theory does not account for systematic relations within word forms. We examine how a restricted set of grammatical meanings (e.g. person, number) are expressed on verbs and pronouns across typologically diverse languages. Consistent with prior work, we find that verb and pronoun forms are shaped by competing communicative pressures for simplicity (minimizing the inventory of grammatical distinctions) and accuracy (enabling recovery of intended meanings). Crucially, our proposed model uses a novel measure of complexity (inverse of simplicity) based on the learnability of meaning-to-form mappings. This innovation captures fine-grained regularities in linguistic form, allowing better discrimination between attested and unattested systems, and establishes a new connection from efficient communication theory to systematicity in natural language. --- ### Conservative & Aggressive NaNs Accelerate U-Nets for Neuroimaging Authors: Inés Gonzalez-Pepe, Vinuyan Sivakolunthu, Jacob Fortin, Yohan Chatelain, Tristan Glatard Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17180v1 Deep learning models for neuroimaging increasingly rely on large architectures, making efficiency a persistent concern despite advances in hardware. Through an analysis of numerical uncertainty of convolutional neural networks (CNNs), we observe that many operations are applied to values dominated by numerical noise and have negligible influence on model outputs. In some models, up to two-thirds of convolution operations appear redundant. We introduce Conservative & Aggressive NaNs, two novel variants of max pooling and unpooling that identify numerically unstable voxels and replace them with NaNs, allowing subsequent layers to skip computations on irrelevant data. Both methods are implemented within PyTorch and require no architectural changes. We evaluate these approaches on four CNN models spanning neuroimaging and image classification tasks. For inputs containing at least 50% NaNs, we observe consistent runtime improvements; for data with more than two-thirds NaNs )common in several neuroimaging settings) we achieve an average inference speedup of 1.67x. Conservative NaNs reduces convolution operations by an average of 30% across models and datasets, with no measurable performance degradation, and can skip up to 64.64% of convolutions in specific layers. Aggressive NaNs can skip up to 69.30% of convolutions but may occasionally affect performance. Overall, these methods demonstrate that numerical uncertainty can be exploited to reduce redundant computation and improve inference efficiency in CNNs. --- ### TrojanGYM: A Detector-in-the-Loop LLM for Adaptive RTL Hardware Trojan Insertion Authors: Saideep Sreekumar, Zeng Wang, Akashdeep Saha, Weihua Xiao, Minghao Shao, Muhammad Shafique, Ozgur Sinanoglu, Ramesh Karri, Johann Knechtel Published: 2026-01-23 Categories: cs.CR, cs.AI, cs.AR arXiv: https://arxiv.org/abs/2601.17178v3 Hardware Trojans (HTs) remain a critical threat because learning-based detectors often overfit to narrow trigger/payload patterns and small, stylized benchmarks. We introduce TrojanGYM, an agentic, LLM-driven framework that automatically curates HT insertions to expose detector blind spots. Given high-level HT specifications, a suite of cooperating LLM agents (instantiated with GPT-4, LLaMA-3.3-70B, Gemini-2.5Pro, and Claude Opus 4.5) proposes and refines RTL modifications that realize diverse triggers and payloads without impacting functionality of both the HT and the design under attack. TrojanGYM implements an agentic loop co-designed with HT detectors, in which constraint-aware syntactic checking, testbench-based functional verification, and GNN-based HT detectors provide feedback that iteratively refines HT specifications and insertion strategies to better surface detector blind spots. We further propose Robust-GNN4TJ, a new implementation of GNN4TJ with improved graph extraction, training robustness, and prediction reliability, especially on LLM-generated HT designs. On the most challenging TrojanGYM-generated benchmarks, Robust-GNN4TJ raises HT detection rates from 0% to 60% relative to prior art. We instantiate TrojanGYM on SRAM, AES-128, UART, and RISC-V designs at RTL, and show that it systematically produces diverse, functionally correct HTs that reach up to 68.75% evasion rates against modern GNN-based detectors, revealing robustness gaps that are not apparent when these detectors are evaluated on existing TrustHub-style benchmarks. We release all codes and artifacts at https://github.com/DfX-NYUAD/TrojanGYM. --- ### Beyond Factual QA: Mentorship-Oriented Question Answering over Long-Form Multilingual Content Authors: Parth Bhalerao, Ruiwen Guan, Diola Dsouza, Oana Ignat Published: 2026-01-23 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.17173v3 Question answering systems are typically evaluated on factual correctness, yet many real-world applications-such as education and career guidance-require mentorship: responses that provide reflection and guidance. Existing QA benchmarks rarely capture this distinction, particularly in multilingual and long-form settings. We introduce MentorQA, the first multilingual dataset and evaluation framework for mentorship-focused question answering from long-form videos, comprising nearly 9,000 QA pairs from 180 hours of content across four languages. We define mentorship-focused evaluation dimensions that go beyond factual accuracy, capturing clarity, alignment, and learning value. Using MentorQA, we compare Single-Agent, Dual-Agent, RAG, and Multi-Agent QA architectures under controlled conditions. Multi-Agent pipelines consistently produce higher-quality mentorship responses, with especially strong gains for complex topics and lower-resource languages. We further analyze the reliability of automated LLM-based evaluation, observing substantial variation in alignment with human judgments. Overall, this work establishes mentorship-focused QA as a distinct research problem and provides a multilingual benchmark for studying agentic architectures and evaluation design in educational AI. The dataset and evaluation framework are released at https://github.com/AIM-SCU/MentorQA. --- ### Who Gets Which Message? Auditing Demographic Bias in LLM-Generated Targeted Text Authors: Tunazzina Islam Published: 2026-01-23 Categories: cs.CL, cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2601.17172v3 Large language models (LLMs) are increasingly capable of generating personalized, persuasive text at scale, raising new questions about bias and fairness in automated communication. This paper presents the first systematic analysis of how LLMs behave when tasked with demographic-conditioned targeted messaging. We introduce a controlled evaluation framework using three leading models: GPT-4o, Llama-3.3, and Mistral-Large-2.1, across two generation settings: Standalone Generation, which isolates intrinsic demographic effects, and Context-Rich Generation, which incorporates thematic and regional context to emulate realistic targeting. We evaluate generated messages along three dimensions: lexical content, language style, and persuasive framing. We instantiate this framework on climate communication and find consistent age- and gender-based asymmetries across models: male- and youth-targeted messages tend to emphasize more assertive and progressive framing, while female- and senior-targeted messages more often reflect warmth, care, and traditional themes. Contextual prompts systematically amplify these disparities, with persuasion scores being higher for male-targeted messages, while age-related differences vary across models. Our findings demonstrate how demographic stereotypes can surface and intensify in LLM-generated targeted communication, underscoring the need for bias-aware generation pipelines and transparent auditing frameworks that explicitly account for demographic conditioning in socially sensitive applications. --- ### A Unified Kantorovich Duality for Multimarginal Optimal Transport Authors: Yehya Cheryala, Mokhtar Z. Alaya, Salim Bouzebda Published: 2026-01-23 Categories: math.OC, stat.ML arXiv: https://arxiv.org/abs/2601.17171v1 Multimarginal optimal transport (MOT) has gained increasing attention in recent years, notably due to its relevance in machine learning and statistics, where one seeks to jointly compare and align multiple probability distributions. This paper presents a unified and complete Kantorovich duality theory for MOT problem on general Polish product spaces with bounded continuous cost function. For marginal compact spaces, the duality identity is derived through a convex-analytic reformulation, that identifies the dual problem as a Fenchel-Rockafellar conjugate. We obtain dual attainment and show that optimal potentials may always be chosen in the class of $c$-conjugate families, thereby extending classical two-marginal conjugacy principle into a genuinely multimarginal setting. In non-compact setting, where direct compactness arguments are unavailable, we recover duality via a truncation-tightness procedure based on weak compactness of multimarginal transference plans and boundedness of the cost. We prove that the dual value is preserved under restriction to compact subsets and that admissible dual families can be regularized into uniformly bounded $c$-conjugate potentials. The argument relies on a refined use of $c$-splitting sets and their equivalence with multimarginal $c$-cyclical monotonicity. We then obtain dual attainment and exact primal-dual equality for MOT on arbitrary Polish spaces, together with a canonical representation of optimal dual potentials by $c$-conjugacy. These results provide a structural foundation for further developments in probabilistic and statistical analysis of MOT, including stability, differentiability, and asymptotic theory under marginal perturbations. --- ### Interpreting Agentic Systems: Beyond Model Explanations to System-Level Accountability Authors: Judy Zhu, Dhari Gandhi, Himanshu Joshi, Ahmad Rezaie Mianroodi, Sedef Akinli Kocak, Dhanesh Ramachandran Published: 2026-01-23 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2601.17168v1 Agentic systems have transformed how Large Language Models (LLMs) can be leveraged to create autonomous systems with goal-directed behaviors, consisting of multi-step planning and the ability to interact with different environments. These systems differ fundamentally from traditional machine learning models, both in architecture and deployment, introducing unique AI safety challenges, including goal misalignment, compounding decision errors, and coordination risks among interacting agents, that necessitate embedding interpretability and explainability by design to ensure traceability and accountability across their autonomous behaviors. Current interpretability techniques, developed primarily for static models, show limitations when applied to agentic systems. The temporal dynamics, compounding decisions, and context-dependent behaviors of agentic systems demand new analytical approaches. This paper assesses the suitability and limitations of existing interpretability methods in the context of agentic systems, identifying gaps in their capacity to provide meaningful insight into agent decision-making. We propose future directions for developing interpretability techniques specifically designed for agentic systems, pinpointing where interpretability is required to embed oversight mechanisms across the agent lifecycle from goal formation, through environmental interaction, to outcome evaluation. These advances are essential to ensure the safe and accountable deployment of agentic AI systems. --- ### Benchmarking Deep Learning-Based Reconstruction Methods for Photoacoustic Computed Tomography with Clinically Relevant Synthetic Datasets Authors: Panpan Chen, Seonyeong Park, Gangwon Jeong, Refik Mert Cam, Umberto Villa, Mark A. Anastasio Published: 2026-01-23 Categories: physics.med-ph, cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.17165v1 Deep learning (DL)-based image reconstruction methods for photoacoustic computed tomography (PACT) have developed rapidly in recent years. However, most existing methods have not employed standardized datasets, and their evaluations rely on traditional image quality (IQ) metrics that may lack clinical relevance. The absence of a standardized framework for clinically meaningful IQ assessment hinders fair comparison and raises concerns about the reproducibility and reliability of reported advancements in PACT. A benchmarking framework is proposed that provides open-source, anatomically plausible synthetic datasets and evaluation strategies for DL-based acoustic inversion methods in PACT. The datasets each include over 11,000 two-dimensional (2D) stochastic breast objects with clinically relevant lesions and paired measurements at varying modeling complexity. The evaluation strategies incorporate both traditional and task-based IQ measures to assess fidelity and clinical utility. A preliminary benchmarking study is conducted to demonstrate the framework's utility by comparing DL-based and physics-based reconstruction methods. The benchmarking study demonstrated that the proposed framework enabled comprehensive, quantitative comparisons of reconstruction performance and revealed important limitations in certain DL-based methods. Although they performed well according to traditional IQ measures, they often failed to accurately recover lesions. This highlights the inadequacy of traditional metrics and motivates the need for task-based assessments. The proposed benchmarking framework enables systematic comparisons of DL-based acoustic inversion methods for 2D PACT. By integrating clinically relevant synthetic datasets with rigorous evaluation protocols, it enables reproducible, objective assessments and facilitates method development and system optimization in PACT. --- ### Information-Theoretic Causal Bounds under Unmeasured Confounding Authors: Yonghan Jung, Bogyeong Kang Published: 2026-01-23 Categories: stat.ML, cs.AI, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2601.17160v3 We develop a data-driven information-theoretic framework for sharp partial identification of causal effects under unmeasured confounding. Existing approaches often rely on restrictive assumptions, such as bounded or discrete outcomes; require external inputs (for example, instrumental variables, proxies, or user-specified sensitivity parameters); necessitate full structural causal model specifications; or focus solely on population-level averages while neglecting covariate-conditional effects. We overcome all four limitations simultaneously by establishing novel information-theoretic, data-driven divergence bounds. Our key theoretical contribution shows that the f-divergence between the observational distribution P(Y | A = a, X = x) and the interventional distribution P(Y | do(A = a), X = x) is upper bounded by a function of the propensity score alone. This result enables sharp partial identification of conditional causal effects directly from observational data, without requiring external sensitivity parameters, auxiliary variables, full structural specifications, or outcome boundedness assumptions. For practical implementation, we develop a semiparametric estimator satisfying Neyman orthogonality (Chernozhukov et al., 2018), which ensures root-n consistent inference even when nuisance functions are estimated via flexible machine learning methods. Simulation studies and real-world data applications, implemented in the GitHub repository (https://github.com/yonghanjung/Information-Theretic-Bounds), demonstrate that our framework provides tight and valid causal bounds across a wide range of data-generating processes. --- ### Interpretability of the Intent Detection Problem: A New Approach Authors: Eduardo Sanchez-Karhunen, Jose F. Quesada-Moreno, Miguel A. Gutiérrez-Naranjo Published: 2026-01-23 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.17156v1 Intent detection, a fundamental text classification task, aims to identify and label the semantics of user queries, playing a vital role in numerous business applications. Despite the dominance of deep learning techniques in this field, the internal mechanisms enabling Recurrent Neural Networks (RNNs) to solve intent detection tasks are poorly understood. In this work, we apply dynamical systems theory to analyze how RNN architectures address this problem, using both the balanced SNIPS and the imbalanced ATIS datasets. By interpreting sentences as trajectories in the hidden state space, we first show that on the balanced SNIPS dataset, the network learns an ideal solution: the state space, constrained to a low-dimensional manifold, is partitioned into distinct clusters corresponding to each intent. The application of this framework to the imbalanced ATIS dataset then reveals how this ideal geometric solution is distorted by class imbalance, causing the clusters for low-frequency intents to degrade. Our framework decouples geometric separation from readout alignment, providing a novel, mechanistic explanation for real world performance disparities. These findings provide new insights into RNN dynamics, offering a geometric interpretation of how dataset properties directly shape a network's computational solution. --- ### Dynamic Role Assignment for Multi-Agent Debate Authors: Miao Zhang, Junsik Kim, Siyuan Xiang, Jian Gao, Cheng Cao Published: 2026-01-23 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.17152v1 Multi-agent large language model (LLM) and vision-language model (VLM) debate systems employ specialized roles for complex problem-solving, yet model specializations are not leveraged to decide which model should fill which role. We propose dynamic role assignment, a framework that runs a Meta-Debate to select suitable agents before the actual debate. The meta-debate has two stages: (1) proposal, where candidates provide role-tailored arguments, and (2) peer review, where proposals are scored with data and role-specific criteria to choose the best agent for each position. We evaluate our method on LLM problem solving benchmarks. Applied on top of existing debate systems, our approach consistently outperforms uniform assignments (filling all roles with the same model) by up to 74.8% and random assignments (assigning models to roles without considering their suitability) by up to 29.7%, depending on the task and the specific assignment. This work establishes a new paradigm for multi-agent system design, shifting from static agent deployment to dynamic and capability-aware selection. --- ### Scaling medical imaging report generation with multimodal reinforcement learning Authors: Qianchu Liu, Sheng Zhang, Guanghui Qin, Yu Gu, Ying Jin, Sam Preston, Yanbo Xu, Sid Kiblawi, Wen-wai Yim, Timothy Ossowski, Tristan Naumann, Mu Wei et al. Published: 2026-01-23 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2601.17151v2 Frontier models have demonstrated remarkable capabilities in understanding and reasoning with natural-language text, but they still exhibit major competency gaps in multimodal understanding and reasoning especially in high-value verticals such as biomedicine. Medical imaging report generation is a prominent example. Supervised fine-tuning can substantially improve performance, but they are prone to overfitting to superficial boilerplate patterns. In this paper, we introduce Universal Report Generation (UniRG) as a general framework for medical imaging report generation. By leveraging reinforcement learning as a unifying mechanism to directly optimize for evaluation metrics designed for end applications, UniRG can significantly improve upon supervised fine-tuning and attain durable generalization across diverse institutions and clinical practices. We trained UniRG-CXR on publicly available chest X-ray (CXR) data and conducted a thorough evaluation in CXR report generation with rigorous evaluation scenarios. On the authoritative ReXrank benchmark, UniRG-CXR sets new overall SOTA, outperforming prior state of the art by a wide margin. We release our model at https://huggingface.co/microsoft/UniRG-CXR. --- ### Falsifying Discriminant Validity of Predictive Algorithms Authors: Amanda Coston Published: 2026-01-23 Categories: stat.ME, cs.CY, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.17146v2 Empirical investigations into unintended model behavior often show that the algorithm is predicting another outcome than what was intended. These exposés highlight the need to identify when algorithms predict unintended quantities - ideally before deploying them into consequential settings. We propose a falsification framework that provides a principled statistical test for discriminant validity: the requirement that an algorithm predict intended outcomes better than impermissible ones. Drawing on falsification practices from causal inference, econometrics, and psychometrics, our framework compares calibrated prediction losses across outcomes to assess whether the algorithm exhibits discriminant validity with respect to a specified impermissible proxy. In settings where the target outcome is difficult to observe, multiple permissible proxy outcomes may be available; our framework accommodates both this setting and the case with a single permissible proxy. Throughout we use nonparametric hypothesis testing methods that make minimal assumptions on the data-generating process. We illustrate the method in an admissions setting, where the framework establishes discriminant validity with respect to gender but fails to establish discriminant validity with respect to race. This demonstrates how falsification can serve as an early validity check. We also provide analysis in a criminal justice setting, where we highlight the limitations of our framework and emphasize the need for complementary approaches to assess other aspects of construct validity and external validity. --- ### Communication-Avoiding Linear Algebraic Kernel K-Means on GPUs Authors: Julian Bellavita, Matthew Rubino, Nakul Iyer, Andrew Chang, Aditya Devarakonda, Flavio Vella, Giulia Guidi Published: 2026-01-23 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2601.17136v2 Clustering is an important tool in data analysis, with K-means being popular for its simplicity and versatility. However, it cannot handle non-linearly separable clusters. Kernel K-means addresses this limitation but requires a large kernel matrix, making it computationally and memory intensive. Prior work has accelerated Kernel K-means by formulating it using sparse linear algebra primitives and implementing it on a single GPU. However, that approach cannot run on datasets with more than approximately 80,000 samples due to limited GPU memory. In this work, we address this issue by presenting a suite of distributed-memory parallel algorithms for large-scale Kernel K-means clustering on multi-GPU systems. Our approach maps the most computationally expensive components of Kernel K-means onto communication-efficient distributed linear algebra primitives uniquely tailored for Kernel K-means, enabling highly scalable implementations that efficiently cluster million-scale datasets. Central to our work is the design of partitioning schemes that enable communication-efficient composition of the linear algebra primitives that appear in Kernel K-means. Our 1.5D algorithm consistently achieves the highest performance, enabling Kernel K-means to scale to data one to two orders of magnitude larger than previously practical. On 256 GPUs, it achieves a geometric mean weak scaling efficiency of $79.7\%$ and a geometric mean strong scaling speedup of $4.2\times$. Compared to our 1D algorithm, the 1.5D approach achieves up to a $3.6\times$ speedup on 256 GPUs and reduces clustering time from over an hour to under two seconds relative to a single-GPU sliding window implementation. Our results show that distributed algorithms designed with application-specific linear algebraic formulations can achieve substantial performance improvement. --- ### ConceptACT: Episode-Level Concepts for Sample-Efficient Robotic Imitation Learning Authors: Jakob Karalus, Friedhelm Schwenker Published: 2026-01-23 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2601.17135v1 Imitation learning enables robots to acquire complex manipulation skills from human demonstrations, but current methods rely solely on low-level sensorimotor data while ignoring the rich semantic knowledge humans naturally possess about tasks. We present ConceptACT, an extension of Action Chunking with Transformers that leverages episode-level semantic concept annotations during training to improve learning efficiency. Unlike language-conditioned approaches that require semantic input at deployment, ConceptACT uses human-provided concepts (object properties, spatial relationships, task constraints) exclusively during demonstration collection, adding minimal annotation burden. We integrate concepts using a modified transformer architecture in which the final encoder layer implements concept-aware cross-attention, supervised to align with human annotations. Through experiments on two robotic manipulation tasks with logical constraints, we demonstrate that ConceptACT converges faster and achieves superior sample efficiency compared to standard ACT. Crucially, we show that architectural integration through attention mechanisms significantly outperforms naive auxiliary prediction losses or language-conditioned models. These results demonstrate that properly integrated semantic supervision provides powerful inductive biases for more efficient robot learning. --- ### Learning to Collaborate: An Orchestrated-Decentralized Framework for Peer-to-Peer LLM Federation Authors: Inderjeet Singh, Eleonore Vissol-Gaudin, Andikan Otung, Motoyoshi Sekiya Published: 2026-01-23 Categories: cs.LG, cs.AI, cs.CR, cs.DC, cs.MA arXiv: https://arxiv.org/abs/2601.17133v1 Fine-tuning Large Language Models (LLMs) for specialized domains is constrained by a fundamental challenge: the need for diverse, cross-organizational data conflicts with the principles of data privacy and sovereignty. While Federated Learning (FL) provides a framework for collaboration without raw data exchange, its classic centralized form introduces a single point of failure and remains vulnerable to model inversion attacks. Decentralized FL (DFL) mitigates this risk by removing the central aggregator but typically relies on inefficient, random peer-to-peer (P2P) pairings, forming a collaboration graph that is blind to agent heterogeneity and risks negative transfer. This paper introduces KNEXA-FL, a novel framework for orchestrated decentralization that resolves this trade-off. KNEXA-FL employs a non-aggregating Central Profiler/Matchmaker (CPM) that formulates P2P collaboration as a contextual bandit problem, using a LinUCB algorithm on abstract agent profiles to learn an optimal matchmaking policy. It orchestrates direct knowledge exchange between heterogeneous, PEFT-based LLM agents via secure distillation, without ever accessing the models themselves. Our comprehensive experiments on a challenging code generation task show that KNEXA-FL yields substantial gains, improving Pass@1 by approx. 50% relative to random P2P collaboration. Critically, our orchestrated approach demonstrates stable convergence, in stark contrast to a powerful centralized distillation baseline which suffers from catastrophic performance collapse. Our work establishes adaptive, learning-based orchestration as a foundational principle for building robust and effective decentralized AI ecosystems. --- ### From Emotion to Expression: Theoretical Foundations and Resources for Fear Speech Authors: Vigneshwaran Shankaran, Gabriella Lapesa, Claudia Wagner Published: 2026-01-23 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.17132v1 Few forces rival fear in their ability to mobilize societies, distort communication, and reshape collective behavior. In computational linguistics, fear is primarily studied as an emotion, but not as a distinct form of speech. Fear speech content is widespread and growing, and often outperforms hate-speech content in reach and engagement because it appears "civiler" and evades moderation. Yet the computational study of fear speech remains fragmented and under-resourced. This can be understood by recognizing that fear speech is a phenomenon shaped by contributions from multiple disciplines. In this paper, we bridge cross-disciplinary perspectives by comparing theories of fear from Psychology, Political science, Communication science, and Linguistics. Building on this, we review existing definitions. We follow up with a survey of datasets from related research areas and propose a taxonomy that consolidates different dimensions of fear for studying fear speech. By reviewing current datasets and defining core concepts, our work offers both theoretical and practical guidance for creating datasets and advancing fear speech research. --- ### Impact of Graph Structure on Membership-Inference Risk for Graph Neural Networks Authors: Megha Khosla Published: 2026-01-23 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2601.17130v2 Graph neural networks (GNNs) are widely used for tasks such as node classification and link prediction, but their use in sensitive settings raises concerns about training-data leakage. Prior work on privacy leakage in GNNs largely borrows assumptions from non-graph domains, overlooking the role of graph structure. We argue for a graph-specific analysis of privacy risk and study how graph structure affects node-level membership inference. We formalize membership inference (MI) over node-neighborhood tuples and investigate two important dimensions: (i) training-graph construction and (ii) inference-time edge access. We compare snowball sampling, a structure-aware procedure, with uniform random node sampling for constructing training graphs. Our experiments show that snowball sampling often hurts generalization relative to random sampling due to its coverage bias. In contrast, allowing access to inter-train-test edges at inference improves test accuracy, reduces the train-test gap, while also having a strong and setting-dependent effect on membership advantage. These results show that graph structure directly shapes privacy risk. We further show that the generalization gap, measured as the performance difference between training and test nodes, is an incomplete proxy for membership inference risk: membership advantage can rise or fall independently of changes in this gap, with inference-time edge access often playing a crucial role. Theoretically, we show that for node-level tasks, standard privacy-auditing results based on membership inference do not directly carry over to inductive graph settings, because training and test nodes are structurally dependent rather than interchangeable. We release the code and data at https://github.com/PriXAI/GraphStructurePrivacyAnalysis-public. --- ### EveNet: A Foundation Model for Particle Collision Data Analysis Authors: Ting-Hsiang Hsu, Bai-Hong Zhou, Qibin Liu, Yue Xu, Shu Li, George Wei-Shu Hou, Benjamin Nachman, Shih-Chieh Hsu, Vinicius Mikuni, Yuan-Tang Chou, Yulei Zhang Published: 2026-01-23 Categories: hep-ex, cs.LG, hep-ph arXiv: https://arxiv.org/abs/2601.17126v1 While deep learning is transforming data analysis in high-energy physics, computational challenges limit its potential. We address these challenges in the context of collider physics by introducing EveNet, an event-level foundation model pretrained on 500 million simulated collision events using a hybrid objective of self-supervised learning and physics-informed supervision. By leveraging a shared particle-cloud representation, EveNet outperforms state-of-the-art baselines across diverse tasks, including searches for heavy resonances and exotic Higgs decays, and demonstrates exceptional data efficiency in low-statistics regimes. Crucially, we validate the transferability of the model to experimental data by rediscovering the $Υ$ meson in CMS Open Data and show its capacity for precision physics through the robust extraction of quantum correlation observables stable against systematic uncertainties. These results indicate that EveNet can successfully encode the fundamental physical structure of particle interactions, which offers a unified and resource-efficient framework to accelerate discovery at current and future colliders. --- ### iFSQ: Improving FSQ for Image Generation with 1 Line of Code Authors: Bin Lin, Zongjian Li, Yuwei Niu, Kaixiong Gong, Yunyang Ge, Yunlong Lin, Mingzhe Zheng, JianWei Zhang, Miles Yang, Zhao Zhong, Liefeng Bo, Li Yuan Published: 2026-01-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17124v2 The field of image generation is currently bifurcated into autoregressive (AR) models operating on discrete tokens and diffusion models utilizing continuous latents. This divide, rooted in the distinction between VQ-VAEs and VAEs, hinders unified modeling and fair benchmarking. Finite Scalar Quantization (FSQ) offers a theoretical bridge, yet vanilla FSQ suffers from a critical flaw: its equal-interval quantization can cause activation collapse. This mismatch forces a trade-off between reconstruction fidelity and information efficiency. In this work, we resolve this dilemma by simply replacing the activation function in original FSQ with a distribution-matching mapping to enforce a uniform prior. Termed iFSQ, this simple strategy requires just one line of code yet mathematically guarantees both optimal bin utilization and reconstruction precision. Leveraging iFSQ as a controlled benchmark, we uncover two key insights: (1) The optimal equilibrium between discrete and continuous representations lies at approximately 4 bits per dimension. (2) Under identical reconstruction constraints, AR models exhibit rapid initial convergence, whereas diffusion models achieve a superior performance ceiling, suggesting that strict sequential ordering may limit the upper bounds of generation quality. Finally, we extend our analysis by adapting Representation Alignment (REPA) to AR models, yielding LlamaGen-REPA. Codes is available at https://github.com/Tencent-Hunyuan/iFSQ --- ### Acoustic Field Video for Multimodal Scene Understanding Authors: Daehwa Kim, Chris Harrison Published: 2026-01-23 Categories: cs.HC, cs.CV, cs.RO arXiv: https://arxiv.org/abs/2601.17123v1 We introduce and explore a new multimodal input representation for vision-language models: acoustic field video. Unlike conventional video (RGB with stereo/mono audio), our video stream provides a spatially grounded visualization of sound intensity across a scene, offering a new and powerful dimension of perceptual understanding. Our real-time pipeline uses low-cost beamforming microphone arrays that are already common in smart speakers and increasingly present in robotics and XR headsets, yet this sensing capability remains unutilized for scene understanding. To assess the value of spatial acoustic information, we constructed an evaluation set of 402 question-answer scenes, comparing a state-of-the-art VLM given conventional video with and without paired acoustic field video. Results show a clear and consistent improvement when incorporating spatial acoustic data; the VLM we test improves from 38.3% correct to 67.4%. Our findings highlight that many everyday scene understanding tasks remain underconstrained when relying solely on visual and audio input, and that acoustic field data provides a promising and practical direction for multimodal reasoning. A video demo is available at https://daehwakim.com/seeingsound --- ### Low-Rank Tensor Approximation of Weights in Large Language Models via Cosine Lanczos Bidiagonalization Authors: A. El Ichi, K. Jbilou Published: 2026-01-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.17112v1 Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language tasks but suffer from extremely large memory footprints and computational costs. In this paper, we introduce a tensor compression framework based on the cproduct for computing low rank approximation In the first part of our approach, we leverage the algebraic structure of the cproduct to represent weight tensors such as those in embedding layers, attention projections, and feed forward networks in a transform domain where frontal slices can be jointly approximated by low rank tensor factors. This enables computationally efficient compression that exploits multidimensional correlations beyond traditional SVD methods. --- ### Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts Authors: Xuan-Phi Nguyen, Shrey Pandit, Austin Xu, Caiming Xiong, Shafiq Joty Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17111v1 Mixture-of-Experts (MoE) models are typically pre-trained with explicit load-balancing constraints to ensure statistically balanced expert routing. Despite this, we observe that even well-trained MoE models exhibit significantly imbalanced routing. This behavior is arguably natural-and even desirable - as imbalanced routing allows models to concentrate domain-specific knowledge within a subset of experts. Expert parallelism (EP) is designed to scale MoE models by distributing experts across multiple devices, but with a less-discussed assumption of balanced routing. Under extreme imbalance, EP can funnel a disproportionate number of tokens to a small number of experts, leading to compute- and memory-bound failures on overloaded devices during post-training or inference, where explicit load balancing is often inapplicable. We propose Least-Loaded Expert Parallelism (LLEP), a novel EP algorithm that dynamically reroutes excess tokens and associated expert parameters from overloaded devices to underutilized ones. This ensures that all devices complete their workloads within the minimum collective latency while respecting memory constraints. Across different model scales, LLEP achieves up to 5x speedup and 4x reduction in peak memory usage compared to standard EP. This enables faster and higher-throughput post-training and inference, with ~1.9x faster for gpt-oss-120b. We support our method with extensive theoretical analysis and comprehensive empirical evaluations, including ablation studies. These results illuminate key trade-offs and enable a principled framework for hardware-specific hyper-parameter tuning to achieve optimal performance. --- ### Forecasting Energy Consumption using Recurrent Neural Networks: A Comparative Analysis Authors: Abhishek Maity, Viraj Tukarul Published: 2026-01-23 Categories: cs.CY, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.17110v1 Accurate short-term energy consumption forecasting is essential for efficient power grid management, resource allocation, and market stability. Traditional time-series models often fail to capture the complex, non-linear dependencies and external factors affecting energy demand. In this study, we propose a forecasting approach based on Recurrent Neural Networks (RNNs) and their advanced variant, Long Short-Term Memory (LSTM) networks. Our methodology integrates historical energy consumption data with external variables, including temperature, humidity, and time-based features. The LSTM model is trained and evaluated on a publicly available dataset, and its performance is compared against a conventional feed-forward neural network baseline. Experimental results show that the LSTM model substantially outperforms the baseline, achieving lower Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). These findings demonstrate the effectiveness of deep learning models in providing reliable and precise short-term energy forecasts for real-world applications. --- ### Authority Signals in AI Cited Health Sources: A Framework for Evaluating Source Credibility in ChatGPT Responses Authors: Erin Jacques, Erela Datuowei, Vincent Jones, Corey Basch, Celeta Vanderpool, Nkechi Udeozo, Griselda Chapa Published: 2026-01-23 Categories: cs.DL, cs.AI arXiv: https://arxiv.org/abs/2601.17109v1 Health information seeking has fundamentally changed since the onset of Large Language Models (LLM), with nearly one third of ChatGPT's 800 million users asking health questions weekly. Understanding the sources of those AI generated responses is vital, as health organizations and providers are also investing in digital strategies to organically improve their ranking, reach and visibility in LLM systems like ChatGPT. As AI search optimization strategies are gaining maturity, this study introduces an Authority Signals Framework, organized in four domains that reflect key components to health information seeking, starting with "Who wrote it?" (Author Credentials), followed by "Who published it?" (Institutional Affiliation), "How was it vetted?" (Quality Assurance), and "How does AI find it?" (Digital Authority). This descriptive cross-sectional study randomly selected 100 questions from HealthSearchQA which contains 3,173 consumer health questions curated by Google Research from publicly available search engine suggestions. Those questions were entered into ChatGPT 5.2 Pro to record and code the cited sources through the lens of the Authority Signals Framework's four domains. Descriptive statistics were calculated for all cited sources (n=615), and cross tabulations were conducted to examine distinction among organization types. Over 75% of the sources cited in ChatGPT's health generated responses were from established institutional sources, such as Mayo Clinic, Cleveland Clinic, Wikipedia, National Health Service, PubMed with the remaining citations sourced from alternative health information sources that lacked established institutional backing. --- ### Hybrid Mamba-Attention Neural Architecture for Channel Estimation Authors: Dianxin Luan, Chengsi Liang, Jie Huang, Zheng Lin, Kaitao Meng, John Thompson, Cheng-Xiang Wang, Ozgur Akan Published: 2026-01-23 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2601.17108v3 This paper proposes a hybrid Mamba-attention neural architecture to achieve improved channel estimation for orthogonal frequency-division multiplexing (OFDM) waveforms, particularly for configurations with a large number of subcarriers. By integrating a customized Mamba module, the proposed framework handles large-scale subcarrier channel estimation efficiently while capturing long-distance dependencies among these subcarriers effectively. Unlike the conventional Mamba structure, this paper implements a bidirectional selective scan to enable information propagation from both directions, because channel gains at different subcarriers are inherently non-causal. In addition, by integrating Mamba to reduce the reliance on quadratic-complexity self-attention, the proposed solution achieves lower space complexity than fully transformer architectures. Simulation results based on the 3GPP TS 36.101 channel demonstrate that compared to other baseline neural networks, the proposed method achieves superior channel estimation performance with fewer tunable parameters and exhibits good generalization across previously unseen channels. --- ### StealthMark: Harmless and Stealthy Ownership Verification for Medical Segmentation via Uncertainty-Guided Backdoors Authors: Qinkai Yu, Chong Zhang, Gaojie Jin, Tianjin Huang, Wei Zhou, Wenhui Li, Xiaobo Jin, Bo Huang, Yitian Zhao, Guang Yang, Gregory Y. H. Lip, Yalin Zheng et al. Published: 2026-01-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17107v1 Annotating medical data for training AI models is often costly and limited due to the shortage of specialists with relevant clinical expertise. This challenge is further compounded by privacy and ethical concerns associated with sensitive patient information. As a result, well-trained medical segmentation models on private datasets constitute valuable intellectual property requiring robust protection mechanisms. Existing model protection techniques primarily focus on classification and generative tasks, while segmentation models-crucial to medical image analysis-remain largely underexplored. In this paper, we propose a novel, stealthy, and harmless method, StealthMark, for verifying the ownership of medical segmentation models under black-box conditions. Our approach subtly modulates model uncertainty without altering the final segmentation outputs, thereby preserving the model's performance. To enable ownership verification, we incorporate model-agnostic explanation methods, e.g. LIME, to extract feature attributions from the model outputs. Under specific triggering conditions, these explanations reveal a distinct and verifiable watermark. We further design the watermark as a QR code to facilitate robust and recognizable ownership claims. We conducted extensive experiments across four medical imaging datasets and five mainstream segmentation models. The results demonstrate the effectiveness, stealthiness, and harmlessness of our method on the original model's segmentation performance. For example, when applied to the SAM model, StealthMark consistently achieved ASR above 95% across various datasets while maintaining less than a 1% drop in Dice and AUC scores, significantly outperforming backdoor-based watermarking methods and highlighting its strong potential for practical deployment. Our implementation code is made available at: https://github.com/Qinkaiyu/StealthMark. --- ### Performance uncertainty in medical image analysis: a large-scale investigation of confidence intervals Authors: Pascaline André, Charles Heitz, Evangelia Christodoulou, Annika Reinke, Carole H. Sudre, Michela Antonelli, Patrick Godau, M. Jorge Cardoso, Antoine Gilson, Sophie Tezenas du Montcel, Gaël Varoquaux, Lena Maier-Hein et al. Published: 2026-01-23 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.17103v2 Performance uncertainty quantification is essential for reliable validation and eventual clinical translation of medical imaging artificial intelligence (AI). Confidence intervals (CIs) play a central role in this process by indicating how precise a reported performance estimate is. Yet, due to the limited amount of work examining CI behavior in medical imaging, the community remains largely unaware of how many diverse CI methods exist and how they behave in specific settings. The purpose of this study is to close this gap. To this end, we conducted a large-scale empirical analysis across a total of 24 segmentation and classification tasks, using 19 trained models per task group, a broad spectrum of commonly used performance metrics, multiple aggregation strategies, and several widely adopted CI methods. Reliability (coverage) and precision (width) of each CI method were estimated across all settings to characterize their dependence on study characteristics. Our analysis revealed five principal findings: 1) the sample size required for reliable CIs varies from a few dozens to several thousands of cases depending on study parameters; 2) CI behavior is strongly affected by the choice of performance metric; 3) aggregation strategy substantially influences the reliability of CIs, e.g. they require more observations for macro than for micro; 4) the machine learning problem (segmentation versus classification) modulates these effects; 5) different CI methods are not equally reliable and precise depending on the use case. Finally, we derived practical implications of this study in the form of a decision tree which shall prove useful to the community. This paves the way for future consensus guidelines on reporting performance uncertainty. --- ### Beyond Instrumental and Substitutive Paradigms: Introducing Machine Culture as an Emergent Phenomenon in Large Language Models Authors: Yueqing Hu, Xinyang Peng, Yukun Zhao, Lin Qiu, Ka-lai Hung, Kaiping Peng Published: 2026-01-23 Categories: cs.CY, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.17096v1 Recent scholarship typically characterizes Large Language Models (LLMs) through either an \textit{Instrumental Paradigm} (viewing models as reflections of their developers' culture) or a \textit{Substitutive Paradigm} (viewing models as bilingual proxies that switch cultural frames based on language). This study challenges these anthropomorphic frameworks by proposing \textbf{Machine Culture} as an emergent, distinct phenomenon. We employed a 2 (Model Origin: US vs. China) $\times$ 2 (Prompt Language: English vs. Chinese) factorial design across eight multimodal tasks, uniquely incorporating image generation and interpretation to extend analysis beyond textual boundaries. Results revealed inconsistencies with both dominant paradigms: Model origin did not predict cultural alignment, with US models frequently exhibiting ``holistic'' traits typically associated with East Asian data. Similarly, prompt language did not trigger stable cultural frame-switching; instead, we observed \textbf{Cultural Reversal}, where English prompts paradoxically elicited higher contextual attention than Chinese prompts. Crucially, we identified a novel phenomenon termed \textbf{Service Persona Camouflage}: Reinforcement Learning from Human Feedback (RLHF) collapsed cultural variance in affective tasks into a hyper-positive, zero-variance ``helpful assistant'' persona. We conclude that LLMs do not simulate human culture but exhibit an emergent Machine Culture -- a probabilistic phenomenon shaped by \textit{superposition} in high-dimensional space and \textit{mode collapse} from safety alignment. --- ### LoD Sketch Extraction from Architectural Models Using Generative AI: Dataset Construction for Multi-Level Architectural Design Generation Authors: Xusheng Du, Athiwat Kongkaeo, Ye Zhang, Haoran Xie Published: 2026-01-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.17095v1 For architectural design, representation across multiple Levels of Details (LoD) is essential for achieving a smooth transition from conceptual massing to detailed modeling. However, traditional LoD modeling processes rely on manual operations that are time-consuming, labor-intensive, and prone to geometric inconsistencies. While the rapid advancement of generative artificial intelligence (AI) has opened new possibilities for generating multi-level architectural models from sketch inputs, its application remains limited by the lack of high-quality paired LoD training data. To address this issue, we propose an automatic LoD sketch extraction framework using generative AI models, which progressively simplifies high-detail architectural models to automatically generate geometrically consistent and hierarchically coherent multi-LoD representations. The proposed framework integrates computer vision techniques with generative AI methods to establish a progressive extraction pipeline that transitions from detailed representations to volumetric abstractions. Experimental results demonstrate that the method maintains strong geometric consistency across LoD levels, achieving SSIM values of 0.7319 and 0.7532 for the transitions from LoD3 to LoD2 and from LoD2 to LoD1, respectively, with corresponding normalized Hausdorff distances of 25.1% and 61.0% of the image diagonal, reflecting controlled geometric deviation during abstraction. These results verify that the proposed framework effectively preserves global structure while achieving progressive semantic simplification across different LoD levels, providing reliable data and technical support for AI-driven multi-level architectural generation and hierarchical modeling. --- ### The Mouth is Not the Brain: Bridging Energy-Based World Models and Language Generation Authors: Junichiro Niimi Published: 2026-01-23 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.17094v2 Large Language Models (LLMs) generate fluent text, yet whether they truly understand the world or merely produce plausible texts about it remains contested. We propose an architectural principle, the mouth is not the brain, that explicitly separates world models from language models. Our architecture comprises three components: a DBM that captures domain structure as an energy-based world model, an adapter that projects latent belief states into embedding space, and a frozen GPT-2 that provides linguistic competence without domain knowledge. We instantiate this framework in the consumer review domain using Amazon smartphone reviews. Experiments demonstrate that (1) world model conditioning achieves lower cross-entropy loss and higher semantic similarity than architectural baselines including direct projection and full fine-tuning, while qualitative analysis reveals that soft prompt conditioning resolves a trade-off that prompt-based approaches cannot: simple prompts lack expressiveness while detailed prompts cause output collapse in small LLMs; (2) the DBM's energy function distinguishes coherent from incoherent market configurations, assigning higher energy to implausible brand-price combinations; and (3) interventions on specific attributes propagate causally to generated text with intervened outputs exhibiting distributions statistically consistent with naturally occurring samples sharing the target configuration. These findings suggest that even small-scale language models can achieve consistent, controllable generation when connected to an appropriate world model, providing empirical support for separating linguistic competence from world understanding. --- ### The Triangle of Similarity: A Multi-Faceted Framework for Comparing Neural Network Representations Authors: Olha Sirikova, Alvin Chan Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17093v1 Comparing neural network representations is essential for understanding and validating models in scientific applications. Existing methods, however, often provide a limited view. We propose the Triangle of Similarity, a framework that combines three complementary perspectives: static representational similarity (CKA/Procrustes), functional similarity (Linear Mode Connectivity or Predictive Similarity), and sparsity similarity (robustness under pruning). Analyzing a range of CNNs, Vision Transformers, and Vision-Language Models using both in-distribution (ImageNetV2) and out-of-distribution (CIFAR-10) testbeds, our initial findings suggest that: (1) architectural family is a primary determinant of representational similarity, forming distinct clusters; (2) CKA self-similarity and task accuracy are strongly correlated during pruning, though accuracy often degrades more sharply; and (3) for some model pairs, pruning appears to regularize representations, exposing a shared computational core. This framework offers a more holistic approach for assessing whether models have converged on similar internal mechanisms, providing a useful tool for model selection and analysis in scientific research. --- ### CUROCKET: Optimizing ROCKET for GPU Authors: Ole Stüven, Keno Moenck, Thorsten Schüppstuhl Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17091v1 ROCKET (RandOm Convolutional KErnel Transform) is a feature extraction algorithm created for Time Series Classification (TSC), published in 2019. It applies convolution with randomly generated kernels on a time series, producing features that can be used to train a linear classifier or regressor like Ridge. At the time of publication, ROCKET was on par with the best state-of-the-art algorithms for TSC in terms of accuracy while being significantly less computationally expensive, making ROCKET a compelling algorithm for TSC. This also led to several subsequent versions, further improving accuracy and computational efficiency. The currently available ROCKET implementations are mostly bound to execution on CPU. However, convolution is a task that can be highly parallelized and is therefore suited to be executed on GPU, which speeds up the computation significantly. A key difficulty arises from the inhomogeneous kernels ROCKET uses, making standard methods for applying convolution on GPU inefficient. In this work, we propose an algorithm that is able to efficiently perform ROCKET on GPU and achieves up to 11 times higher computational efficiency per watt than ROCKET on CPU. The code for CUROCKET is available in this repository https://github.com/oleeven/CUROCKET on github. --- ### SFO: Learning PDE Operators via Spectral Filtering Authors: Noam Koren, Rafael Moschopoulos, Kira Radinsky, Elad Hazan Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17090v3 Partial differential equations (PDEs) govern complex systems, yet neural operators often struggle to efficiently capture the long-range, nonlocal interactions inherent in their solution maps. We introduce Spectral Filtering Operator (SFO), a neural operator that parameterizes integral kernels using the Universal Spectral Basis (USB), a fixed, global orthonormal basis derived from the eigenmodes of the Hilbert matrix in spectral filtering theory. Motivated by our theoretical finding that the discrete Green's functions of shift-invariant PDE discretizations exhibit spatial Linear Dynamical System (LDS) structure, we prove that these kernels admit compact approximations in the USB. By learning only the spectral coefficients of rapidly decaying eigenvalues, SFO achieves a highly efficient representation. Across six benchmarks, including reaction-diffusion, fluid dynamics, and 3D electromagnetics, SFO achieves state-of-the-art accuracy, reducing error by up to 40% relative to strong baselines while using substantially fewer parameters. --- ### GRASP: Guided Region-Aware Sparse Prompting for Adapting MLLMs to Remote Sensing Authors: Qigan Sun, Chaoning Zhang, Jianwei Zhang, Xudong Wang, Jiehui Xie, Pengcheng Zheng, Haoyu Wang, Sungyoung Lee, Chi-lok Andy Tai, Yang Yang, Heng Tao Shen Published: 2026-01-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17089v1 In recent years, Multimodal Large Language Models (MLLMs) have made significant progress in visual question answering tasks. However, directly applying existing fine-tuning methods to remote sensing (RS) images often leads to issues such as overfitting on background noise or neglecting target details. This is primarily due to the large-scale variations, sparse target distributions, and complex regional semantic features inherent in RS images. These challenges limit the effectiveness of MLLMs in RS tasks. To address these challenges, we propose a parameter-efficient fine-tuning (PEFT) strategy called Guided Region-Aware Sparse Prompting (GRASP). GRASP introduces spatially structured soft prompts associated with spatial blocks extracted from a frozen visual token grid. Through a question-guided sparse fusion mechanism, GRASP dynamically aggregates task-specific context into a compact global prompt, enabling the model to focus on relevant regions while filtering out background noise. Extensive experiments on multiple RSVQA benchmarks show that GRASP achieves competitive performance compared to existing fine-tuning and prompt-based methods while maintaining high parameter efficiency. --- ### GlassesGB: Controllable 2D GAN-Based Eyewear Personalization for 3D Gaussian Blendshapes Head Avatars Authors: Rui-Yang Ju, Jen-Shiun Chiang Published: 2026-01-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17088v1 Virtual try-on systems allow users to interactively try different products within VR scenarios. However, most existing VTON methods operate only on predefined eyewear templates and lack support for fine-grained, user-driven customization. While GlassesGAN enables personalized 2D eyewear design, its capability remains limited to 2D image generation. Motivated by the success of 3D Gaussian Blendshapes in head reconstruction, we integrate these two techniques and propose GlassesGB, a framework that supports customizable eyewear generation for 3D head avatars. GlassesGB effectively bridges 2D generative customization with 3D head avatar rendering, addressing the challenge in achieving personalized eyewear design for VR applications. The implementation code is available at https://ruiyangju.github.io/GlassesGB. --- ### Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluations Authors: Preethi Seshadri, Samuel Cahyawijaya, Ayomide Odumakinde, Sameer Singh, Seraphina Goldfarb-Tarrant Published: 2026-01-23 Categories: cs.HC, cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2601.17087v2 Agentic benchmarks increasingly rely on LLM-simulated users to scalably evaluate agent performance, yet the robustness, validity, and fairness of this approach remain unexamined. Through a user study with participants across the United States, India, Kenya, and Nigeria, we investigate whether LLM-simulated users serve as reliable proxies for real human users in evaluating agents on τ-Bench retail tasks. We find that user simulation lacks robustness, with agent success rates varying up to 9 percentage points across different user LLMs. Furthermore, evaluations using simulated users exhibit systematic miscalibration, underestimating agent performance on challenging tasks and overestimating it on moderately difficult ones. African American Vernacular English (AAVE) speakers experience consistently worse success rates and calibration errors than Standard American English (SAE) speakers, with disparities compounding significantly with age. We also find simulated users to be a differentially effective proxy for different populations, performing worst for AAVE and Indian English speakers. Additionally, simulated users introduce conversational artifacts and surface different failure patterns than human users. These findings demonstrate that current evaluation practices risk misrepresenting agent capabilities across diverse user populations and may obscure real-world deployment challenges. --- ### SonoEdit: Null-Space Constrained Knowledge Editing for Pronunciation Correction in LLM-Based TTS Authors: Ayush Pratap Singh, Harshit Singh, Nityanand Mathur, Akshat Mandloi, Sudarshan Kamath Published: 2026-01-23 Categories: cs.SD, cs.AI, eess.AS arXiv: https://arxiv.org/abs/2601.17086v1 Neural text-to-speech (TTS) systems systematically mispronounce low-resource proper nouns, particularly non-English names, brands, and geographic locations, due to their underrepresentation in predominantly English training corpora. Existing solutions typically rely on expensive multilingual data collection, supervised finetuning, or manual phonetic annotation, which limits the deployment of TTS systems in linguistically diverse settings. We introduce SonoEdit, a model editing technique that surgically corrects pronunciation errors in pre-trained TTS models without retraining. Instead of costly finetuning or explicit phoneme injection, we propose a parsimonious alternative based on Null-Space Pronunciation Editing, which performs a single-shot parameter update to modify the pronunciation of specific words while provably preserving all other model behavior. We first adapt Acoustic Causal Tracing to identify the Transformer layers responsible for text-to-pronunciation mapping. We then apply Null-Space Constrained Editing to compute a closed-form weight update that corrects the target pronunciation while remaining mathematically orthogonal to the subspace governing general speech generation. This constrained update steers the model's acoustic output toward a desired pronunciation exemplar while guaranteeing zero first-order change on a preserved speech corpus. --- ### Recovering Performance in Speech Emotion Recognition from Discrete Tokens via Multi-Layer Fusion and Paralinguistic Feature Integration Authors: Esther Sun, Abinay Reddy Naini, Carlos Busso Published: 2026-01-23 Categories: eess.AS, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2601.17085v1 Discrete speech tokens offer significant advantages for storage and language model integration, but their application in speech emotion recognition (SER) is limited by paralinguistic information loss during quantization. This paper presents a comprehensive investigation of discrete tokens for SER. Using a fine-tuned WavLM-Large model, we systematically quantify performance degradation across different layer configurations and k-means quantization granularities. To recover the information loss, we propose two key strategies: (1) attention-based multi-layer fusion to recapture complementary information from different layers, and (2) integration of openSMILE features to explicitly reintroduce paralinguistic cues. We also compare mainstream neural codec tokenizers (SpeechTokenizer, DAC, EnCodec) and analyze their behaviors when fused with acoustic features. Our findings demonstrate that through multi-layer fusion and acoustic feature integration, discrete tokens can close the performance gap with continuous representations in SER tasks. --- ### ChemNavigator: Agentic AI Discovery of Design Rules for Organic Photocatalysts Authors: Iman Peivaste, Ahmed Makradi, Salim Belouettar Published: 2026-01-23 Categories: physics.chem-ph, cs.AI, physics.comp-ph arXiv: https://arxiv.org/abs/2601.17084v1 The discovery of high-performance organic photocatalysts for hydrogen evolution remains limited by the vastness of chemical space and the reliance on human intuition for molecular design. Here we present ChemNavigator, an agentic AI system that autonomously derives structure-property relationships through hypothesis-driven exploration of organic photocatalyst candidates. The system integrates large language model reasoning with density functional tight binding calculations in a multi-agent architecture that mirrors the scientific method: formulating hypotheses, designing experiments, executing calculations, and validating findings through rigorous statistical analysis. Through iterative discovery cycles encompassing 200 molecules, ChemNavigator autonomously identified six statistically significant design rules governing frontier orbital energies, including the effects of ether linkages, carbonyl groups, extended conjugation, cyano groups, halogen substituents, and amine groups. Importantly, these rules correspond to established principles of organic electronic structure (resonance donation, inductive withdrawal, $π$-delocalization), demonstrating that the system can independently derive chemical knowledge without explicit programming. Notably, autonomous agentic reasoning extracted these six validated rules from a molecular library where previous ML approaches identified only carbonyl effects. Furthermore, the quantified effect sizes provide a prioritized ranking for synthetic chemists, while feature interaction analysis revealed diminishing returns when combining strategies, challenging additive assumptions in molecular design. This work demonstrates that agentic AI systems can autonomously derive interpretable, chemically grounded design principles, establishing a framework for AI-assisted materials discovery that complements rather than replaces chemical intuition. --- ### Do VLMs Have a Moral Backbone? A Study on the Fragile Morality of Vision-Language Models Authors: Zhining Liu, Tianyi Wang, Xiao Lin, Penghao Ouyang, Gaotang Li, Ze Yang, Hui Liu, Sumit Keswani, Vishwa Pardeshi, Huijun Zhao, Wei Fan, Hanghang Tong Published: 2026-01-23 Categories: cs.CY, cs.AI, cs.CL, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.17082v1 Despite substantial efforts toward improving the moral alignment of Vision-Language Models (VLMs), it remains unclear whether their ethical judgments are stable in realistic settings. This work studies moral robustness in VLMs, defined as the ability to preserve moral judgments under textual and visual perturbations that do not alter the underlying moral context. We systematically probe VLMs with a diverse set of model-agnostic multimodal perturbations and find that their moral stances are highly fragile, frequently flipping under simple manipulations. Our analysis reveals systematic vulnerabilities across perturbation types, moral domains, and model scales, including a sycophancy trade-off where stronger instruction-following models are more susceptible to persuasion. We further show that lightweight inference-time interventions can partially restore moral stability. These results demonstrate that moral alignment alone is insufficient and that moral robustness is a necessary criterion for the responsible deployment of VLMs. --- ### PC-MCL: Patient-Consistent Multi-Cycle Learning with multi-label bias correction for respiratory sound classification Authors: Seung Gyu Jeong, Seong-Eun Kim Published: 2026-01-23 Categories: eess.AS, cs.AI, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2601.17080v1 Automated respiratory sound classification supports the diagnosis of pulmonary diseases. However, many deep models still rely on cycle-level analysis and suffer from patient-specific overfitting. We propose PC-MCL (Patient-Consistent Multi-Cycle Learning) to address these limitations by utilizing three key components: multi-cycle concatenation, a 3-label formulation, and a patient-matching auxiliary task. Our work resolves a multi-label distributional bias in respiratory sound classification, a critical issue inherent to applying multi-cycle concatenation with the conventional 2-label formulation (crackle, wheeze). This bias manifests as a systematic loss of normal signal information when normal and abnormal cycles are combined. Our proposed 3-label formulation (normal, crackle, wheeze) corrects this by preserving information from all constituent cycles in mixed samples. Furthermore, the patient-matching auxiliary task acts as a multi-task regularizer, encouraging the model to learn more robust features and improving generalization. On the ICBHI 2017 benchmark, PC-MCL achieves an ICBHI Score of 65.37%, outperforming existing baselines. Ablation studies confirm that all three components are essential, working synergistically to improve the detection of abnormal respiratory events. --- ### E2PL: Effective and Efficient Prompt Learning for Incomplete Multi-view Multi-Label Class Incremental Learning Authors: Jiajun Chen, Yue Wu, Kai Huang, Wen Xi, Yangyang Wu, Xiaoye Miao, Mengying Zhu, Meng Xi, Guanjie Cheng Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17076v1 Multi-view multi-label classification (MvMLC) is indispensable for modern web applications aggregating information from diverse sources. However, real-world web-scale settings are rife with missing views and continuously emerging classes, which pose significant obstacles to robust learning. Prevailing methods are ill-equipped for this reality, as they either lack adaptability to new classes or incur exponential parameter growth when handling all possible missing-view patterns, severely limiting their scalability in web environments. To systematically address this gap, we formally introduce a novel task, termed \emph{incomplete multi-view multi-label class incremental learning} (IMvMLCIL), which requires models to simultaneously address heterogeneous missing views and dynamic class expansion. To tackle this task, we propose \textsf{E2PL}, an Effective and Efficient Prompt Learning framework for IMvMLCIL. \textsf{E2PL} unifies two novel prompt designs: \emph{task-tailored prompts} for class-incremental adaptation and \emph{missing-aware prompts} for the flexible integration of arbitrary view-missing scenarios. To fundamentally address the exponential parameter explosion inherent in missing-aware prompts, we devise an \emph{efficient prototype tensorization} module, which leverages atomic tensor decomposition to elegantly reduce the prompt parameter complexity from exponential to linear w.r.t. the number of views. We further incorporate a \emph{dynamic contrastive learning} strategy explicitly model the complex dependencies among diverse missing-view patterns, thus enhancing the model's robustness. Extensive experiments on three benchmarks demonstrate that \textsf{E2PL} consistently outperforms state-of-the-art methods in both effectiveness and efficiency. The codes and datasets are available at https://anonymous.4open.science/r/code-for-E2PL. --- ### Physics-Encoded Inverse Modeling for Arctic Snow Depth Estimation Authors: Akila Sampath, Vandana P. Janeja, Jianwu Wang Published: 2026-01-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.17074v5 Accurate estimation of unobserved quantities in time-varying inverse problems remains challenging when observations are sparse and only indirectly related to the target variable. In Arctic climate applications, snow depth over sea ice is not directly available in commonly used reanalysis products and must instead be inferred from related physical and environmental variables. To address this challenge, we introduce Physics-Encoded Inverse Modeling (PhysE-Inv), a framework that combines sequential deep learning with a physics-encoded parameter estimation module for inverse estimation under sparse observational conditions. PhysE-Inv uses an LSTM encoder-decoder to capture temporal dependencies and incorporates contrastive learning to improve the consistency of learned representations. The framework learns structured physics-encoded parameters that are integrated with observational inputs to estimate snow depth proxies. Under the proposed proxy evaluation framework, PhysE-Inv outperforms all evaluated baselines, achieving an average MSE reduction of 24.4\% compared with baseline models and a 17.3\% improvement over the strongest baseline under the parameter estimation setting. These results demonstrate the potential of physics-encoded modeling approaches for estimating unobserved quantities in data-scarce domains where direct observations are limited. --- ### Attention-Based Variational Framework for Joint and Individual Components Learning with Applications in Brain Network Analysis Authors: Yifei Zhang, Meimei Liu, Zhengwu Zhang Published: 2026-01-23 Categories: cs.LG, cs.CV, stat.ML arXiv: https://arxiv.org/abs/2601.17073v1 Brain organization is increasingly characterized through multiple imaging modalities, most notably structural connectivity (SC) and functional connectivity (FC). Integrating these inherently distinct yet complementary data sources is essential for uncovering the cross-modal patterns that drive behavioral phenotypes. However, effective integration is hindered by the high dimensionality and non-linearity of connectome data, complex non-linear SC-FC coupling, and the challenge of disentangling shared information from modality-specific variations. To address these issues, we propose the Cross-Modal Joint-Individual Variational Network (CM-JIVNet), a unified probabilistic framework designed to learn factorized latent representations from paired SC-FC datasets. Our model utilizes a multi-head attention fusion module to capture non-linear cross-modal dependencies while isolating independent, modality-specific signals. Validated on Human Connectome Project Young Adult (HCP-YA) data, CM-JIVNet demonstrates superior performance in cross-modal reconstruction and behavioral trait prediction. By effectively disentangling joint and individual feature spaces, CM-JIVNet provides a robust, interpretable, and scalable solution for large-scale multimodal brain analysis. ---