# Research Papers — 2026-01 · bundle page 59 of 189

> Full-text markdown of papers #2901–2950 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/59
[← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/58.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-01/page/60.md)

### One-Shot Federated Clustering of Non-Independent Completely Distributed Data

Authors: Yiqun Zhang, Shenghong Cai, Zihua Yang, Sen Feng, Yuzhu Ji, Haijun Zhang
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17512v1

Federated Learning (FL) that extracts data knowledge while protecting the privacy of multiple clients has achieved remarkable results in distributed privacy-preserving IoT systems, including smart traffic flow monitoring, smart grid load balancing, and so on. Since most data collected from edge devices are unlabeled, unsupervised Federated Clustering (FC) is becoming increasingly popular for exploring pattern knowledge from complex distributed data. However, due to the lack of label guidance, the common Non-Independent and Identically Distributed (Non-IID) issue of clients have greatly challenged FC by posing the following problems: How to fuse pattern knowledge (i.e., cluster distribution) from Non-IID clients; How are the cluster distributions among clients related; and How does this relationship connect with the global knowledge fusion? In this paper, a more tricky but overlooked phenomenon in Non-IID is revealed, which bottlenecks the clustering performance of the existing FC approaches. That is, different clients could fragment a cluster, and accordingly, a more generalized Non-IID concept, i.e., Non-ICD (Non-Independent Completely Distributed), is derived. To tackle the above FC challenges, a new framework named GOLD (Global Oriented Local Distribution Learning) is proposed. GOLD first finely explores the potential incomplete local cluster distributions of clients, then uploads the distribution summarization to the server for global fusion, and finally performs local cluster enhancement under the guidance of the global distribution. Extensive experiments, including significance tests, ablation studies, scalability evaluations, qualitative results, etc., have been conducted to show the superiority of GOLD.

---

### "Rebuilding" Statistics in the Age of AI: A Town Hall Discussion on Culture, Infrastructure, and Training

Authors: David L. Donoho, Jian Kang, Xihong Lin, Bhramar Mukherjee, Dan Nettleton, Rebecca Nugent, Abel Rodriguez, Eric P. Xing, Tian Zheng, Hongtu Zhu
Published: 2026-01-24
Categories: stat.ML, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.17510v1

This article presents the full, original record of the 2024 Joint Statistical Meetings (JSM) town hall, "Statistics in the Age of AI," which convened leading statisticians to discuss how the field is evolving in response to advances in artificial intelligence, foundation models, large-scale empirical modeling, and data-intensive infrastructures. The town hall was structured around open panel discussion and extensive audience Q&A, with the aim of eliciting candid, experience-driven perspectives rather than formal presentations or prepared statements. This document preserves the extended exchanges among panelists and audience members, with minimal editorial intervention, and organizes the conversation around five recurring questions concerning disciplinary culture and practices, data curation and "data work," engagement with modern empirical modeling, training for large-scale AI applications, and partnerships with key AI stakeholders. By providing an archival record of this discussion, the preprint aims to support transparency, community reflection, and ongoing dialogue about the evolving role of statistics in the data- and AI-centric future.

---

### BMDS-Net:Deployment-aware multi-modal brain tumor segmentation with adaptive fusion,decoder regularization,and Bayesian calibration

Authors: Yan Zhou, Suncheng Xiang, Zhen Huang, Yue Ouyang, Yingqiu Li, Zehua Wang
Published: 2026-01-24
Categories: cs.CV, q-bio.QM
arXiv: https://arxiv.org/abs/2601.17504v2

Multi-modal MRI enables detailed brain tumor sub-region segmentation, but clinical deployment remains affected by missing sequences,boundary errors, and overconfident predictions. We present BMDS-Net, a two-stage Swin UNETR framework that combines adaptivemodality fusion, boundary-aware decoder regularization, and efficient Bayesian calibration. The deterministic stage adds zero-initializedmultimodal contextual fusion (MMCF) and residual-gated deep decoder supervision (DDS) to modulate modality contributions andregularize hierarchical boundary reconstruction. The second stage replaces only the final convolutional layer with a Bayesian layer,enabling Monte Carlo uncertainty estimates with limited additional training cost. On BraTS 2021 (1,251 cases), BMDS-Net achieved Dice scores of 0.9293, 0.9098, and 0.8675 for WT, TC, and ET, with HD95 values of 2.27, 2.22, and 3.27 mm. Configuration comparisons showed that auxiliary supervision favored complete-input segmentation, whereas the complete MMCF+gated-DDS configuration was most robust under single-modality removal. Bayesian BMDS-Net achieved an expected calibration error of 0.0037, close to a three-model ensemble (0.0035) at approximately 60% lower training cost. On BraTS 2020, the same protocol preserved the leading internal ranking of the complete configuration among five Swin UNETR-controlled variants. The implementation code is available at https://github.com/RyanZhou168/BMDS-Net.

---

### To Case or Not to Case: An Empirical Study in Learned Sparse Retrieval

Authors: Emmanouil Georgios Lionis, Jia-Huei Ju, Angelos Nalmpantis, Casper Thuis, Sean MacAvaney, Andrew Yates
Published: 2026-01-24
Categories: cs.IR, cs.CL
arXiv: https://arxiv.org/abs/2601.17500v1

Learned Sparse Retrieval (LSR) methods construct sparse lexical representations of queries and documents that can be efficiently searched using inverted indexes. Existing LSR approaches have relied almost exclusively on uncased backbone models, whose vocabularies exclude case-sensitive distinctions, thereby reducing vocabulary mismatch. However, the most recent state-of-the-art language models are only available in cased versions. Despite this shift, the impact of backbone model casing on LSR has not been studied, potentially posing a risk to the viability of the method going forward. To fill this gap, we systematically evaluate paired cased and uncased versions of the same backbone models across multiple datasets to assess their suitability for LSR. Our findings show that LSR models with cased backbone models by default perform substantially worse than their uncased counterparts; however, this gap can be eliminated by pre-processing the text to lowercase. Moreover, our token-level analysis reveals that, under lowercasing, cased models almost entirely suppress cased vocabulary items and behave effectively as uncased models, explaining their restored performance. This result broadens the applicability of recent cased models to the LSR setting and facilitates the integration of stronger backbone architectures into sparse retrieval. The complete code and implementation for this project are available at: https://github.com/lionisakis/Uncased-vs-cased-models-in-LSR

---

### PEARL: Prototype-Enhanced Alignment for Label-Efficient Representation Learning with Deployment-Driven Insights from Digital Governance Communication Systems

Authors: Ruiyu Zhang, Lin Nie, Wai-Fung Lam, Qihao Wang, Xin Zhao
Published: 2026-01-24
Categories: cs.LG, cs.AI, cs.CL, cs.IR
arXiv: https://arxiv.org/abs/2601.17495v1

In many deployed systems, new text inputs are handled by retrieving similar past cases, for example when routing and responding to citizen messages in digital governance platforms. When these systems fail, the problem is often not the language model itself, but that the nearest neighbors in the embedding space correspond to the wrong cases. Modern machine learning systems increasingly rely on fixed, high-dimensional embeddings produced by large pretrained models and sentence encoders. In real-world deployments, labels are scarce, domains shift over time, and retraining the base encoder is expensive or infeasible. As a result, downstream performance depends heavily on embedding geometry. Yet raw embeddings are often poorly aligned with the local neighborhood structure required by nearest-neighbor retrieval, similarity search, and lightweight classifiers that operate directly on embeddings. We propose PEARL (Prototype-Enhanced Aligned Representation Learning), a label-efficient approach that uses limited supervision to softly align embeddings toward class prototypes. The method reshapes local neighborhood geometry while preserving dimensionality and avoiding aggressive projection or collapse. Its aim is to bridge the gap between purely unsupervised post-processing, which offers limited and inconsistent gains, and fully supervised projections that require substantial labeled data. We evaluate PEARL under controlled label regimes ranging from extreme label scarcity to higher-label settings. In the label-scarce condition, PEARL substantially improves local neighborhood quality, yielding 25.7% gains over raw embeddings and more than 21.1% gains relative to strong unsupervised post-processing, precisely in the regime where similarity-based systems are most brittle.

---

### SpatialMath: Spatial Comprehension-Infused Symbolic Reasoning for Mathematical Problem-Solving

Authors: Ashutosh Bajpai, Akshat Bhandari, Akshay Nambi, Tanmoy Chakraborty
Published: 2026-01-24
Categories: cs.LG, cs.CL, cs.CV
arXiv: https://arxiv.org/abs/2601.17489v1

Multimodal Small-to-Medium sized Language Models (MSLMs) have demonstrated strong capabilities in integrating visual and textual information but still face significant limitations in visual comprehension and mathematical reasoning, particularly in geometric problems with diverse levels of visual infusion. Current models struggle to accurately decompose intricate visual inputs and connect perception with structured reasoning, leading to suboptimal performance. To address these challenges, we propose SpatialMath, a novel Spatial Comprehension-Infused Symbolic Reasoning Framework designed to integrate spatial representations into structured symbolic reasoning chains. SpatialMath employs a specialized perception module to extract spatially-grounded representations from visual diagrams, capturing critical geometric structures and spatial relationships. These representations are then methodically infused into symbolic reasoning chains, facilitating visual comprehension-aware structured reasoning. To this end, we introduce MATHVERSE-PLUS, a novel dataset containing structured visual interpretations and step-by-step reasoning paths for vision-intensive mathematical problems. SpatialMath significantly outperforms strong multimodal baselines, achieving up to 10 percentage points improvement over supervised fine-tuning with data augmentation in vision-intensive settings. Robustness analysis reveals that enhanced spatial representations directly improve reasoning accuracy, reinforcing the need for structured perception-to-reasoning pipelines in MSLMs.

---

### Automatic Stability and Recovery for Neural Network Training

Authors: Barak Or
Published: 2026-01-24
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.17483v2

Training modern neural networks is increasingly fragile, with rare but severe destabilizing updates often causing irreversible divergence or silent performance degradation. Existing optimization methods primarily rely on preventive mechanisms embedded within the optimizer, offering limited ability to detect and recover from instability once it occurs. We introduce a supervisory runtime stability framework that treats optimization as a controlled stochastic process. By isolating an innovation signal derived from secondary measurements, such as validation probes, the framework enables automatic detection and recovery from destabilizing updates without modifying the underlying optimizer. We provide theoretical runtime safety guarantees that formalize bounded degradation and recovery. Our implementation incurs minimal overhead and is compatible with memory-constrained training settings.

---

### Lattice: Generative Guardrails for Conversational Agents

Authors: Emily Broadhurst, Tawab Safi, Joseph Edell, Vashisht Ganesh, Karime Maamari
Published: 2026-01-24
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.17481v1

Conversational AI systems require guardrails to prevent harmful outputs, yet existing approaches use static rules that cannot adapt to new threats or deployment contexts. We introduce Lattice, a framework for self-constructing and continuously improving guardrails. Lattice operates in two stages: construction builds initial guardrails from labeled examples through iterative simulation and optimization; continuous improvement autonomously adapts deployed guardrails through risk assessment, adversarial testing, and consolidation. Evaluated on the ProsocialDialog dataset, Lattice achieves 91% F1 on held-out data, outperforming keyword baselines by 43pp, LlamaGuard by 25pp, and NeMo by 4pp. The continuous improvement stage achieves 7pp F1 improvement on cross-domain data through closed-loop optimization. Our framework shows that effective guardrails can be self-constructed through iterative optimization.

---

### Unintended Memorization of Sensitive Information in Fine-Tuned Language Models

Authors: Marton Szep, Jorge Marin Ruiz, Georgios Kaissis, Paulina Seidl, Rüdiger von Eisenhart-Rothe, Florian Hinterwimmer, Daniel Rueckert
Published: 2026-01-24
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.17480v1

Fine-tuning Large Language Models (LLMs) on sensitive datasets carries a substantial risk of unintended memorization and leakage of Personally Identifiable Information (PII), which can violate privacy regulations and compromise individual safety. In this work, we systematically investigate a critical and underexplored vulnerability: the exposure of PII that appears only in model inputs, not in training targets. Using both synthetic and real-world datasets, we design controlled extraction probes to quantify unintended PII memorization and study how factors such as language, PII frequency, task type, and model size influence memorization behavior. We further benchmark four privacy-preserving approaches including differential privacy, machine unlearning, regularization, and preference alignment, evaluating their trade-offs between privacy and task performance. Our results show that post-training methods generally provide more consistent privacy-utility trade-offs, while differential privacy achieves strong reduction in leakage in specific settings, although it can introduce training instability. These findings highlight the persistent challenge of memorization in fine-tuned LLMs and emphasize the need for robust, scalable privacy-preserving techniques.

---

### LeanTutor: Towards a Verified AI Mathematical Proof Tutor

Authors: Manooshree Patel, Rayna Bhattacharyya, Thomas Lu, Arnav Mehta, Niels Voss, Narges Norouzi, Gireeja Ranade
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17473v2

This paper considers the development of an AI-based provably-correct mathematical proof tutor. While Large Language Models (LLMs) allow seamless communication in natural language, they are error prone. Theorem provers such as Lean allow for provable-correctness, but these are hard for students to learn. We present a proof-of-concept system (LeanTutor) by combining the complementary strengths of LLMs and theorem provers. LeanTutor is composed of three modules: (i) an autoformalizer/proof-checker, (ii) a next-step generator, and (iii) a natural language feedback generator. To evaluate the system, we introduce PeanoBench, a dataset of 371 Peano Arithmetic proofs in human-written natural language and formal language, derived from the Natural Numbers Game.

---

### PhaSR: Generalized Image Shadow Removal with Physically Aligned Priors

Authors: Chia-Ming Lee, Yu-Fan Lin, Yu-Jou Hsiao, Jin-Hui Jiang, Yu-Lun Liu, Chih-Chung Hsu
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17470v4

Shadow removal under diverse lighting conditions requires disentangling illumination from intrinsic reflectance, a challenge compounded when physical priors are not properly aligned. We propose PhaSR (Physically Aligned Shadow Removal), addressing this through dual-level prior alignment to enable robust performance from single-light shadows to multi-source ambient lighting. First, Physically Aligned Normalization (PAN) performs closed-form illumination correction via Gray-world normalization, log-domain Retinex decomposition, and dynamic range recombination, suppressing chromatic bias. Second, Geometric-Semantic Rectification Attention (GSRA) extends differential attention to cross-modal alignment, harmonizing depth-derived geometry with DINO-v2 semantic embeddings to resolve modal conflicts under varying illumination. Experiments show competitive performance in shadow removal with lower complexity and generalization to ambient lighting where traditional methods fail under multi-source illumination. Our source code is available at https://github.com/ming053l/PhaSR.

---

### Identifying and Correcting Label Noise for Robust GNNs via Influence Contradiction

Authors: Wei Ju, Wei Zhang, Siyu Yi, Zhengyang Mao, Yifan Wang, Jingyang Yuan, Zhiping Xiao, Ziyue Qiao, Ming Zhang
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17469v2

Graph Neural Networks (GNNs) have shown remarkable capabilities in learning from graph-structured data with various applications such as social analysis and bioinformatics. However, the presence of label noise in real scenarios poses a significant challenge in learning robust GNNs, and their effectiveness can be severely impacted when dealing with noisy labels on graphs, often stemming from annotation errors or inconsistencies. To address this, in this paper we propose a novel approach called ICGNN that harnesses the structure information of the graph to effectively alleviate the challenges posed by noisy labels. Specifically, we first design a novel noise indicator that measures the influence contradiction score (ICS) based on the graph diffusion matrix to quantify the credibility of nodes with clean labels, such that nodes with higher ICS values are more likely to be detected as having noisy labels. Then we leverage the Gaussian mixture model to precisely detect whether the label of a node is noisy or not. Additionally, we develop a soft strategy to combine the predictions from neighboring nodes on the graph to correct the detected noisy labels. At last, pseudo-labeling for abundant unlabeled nodes is incorporated to provide auxiliary supervision signals and guide the model optimization. Experiments on benchmark datasets show the superiority of our approach over competitive baselines in noisy label scenarios.

---

### ReflexSplit: Single Image Reflection Separation via Layer Fusion-Separation

Authors: Chia-Ming Lee, Yu-Fan Lin, Jin-Hui Jiang, Yu-Jou Hsiao, Chih-Chung Hsu, Yu-Lun Liu
Published: 2026-01-24
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2601.17468v3

Single Image Reflection Separation (SIRS) disentangles mixed images into transmission and reflection layers. Existing methods suffer from transmission-reflection confusion under nonlinear mixing, particularly in deep decoder layers, due to implicit fusion mechanisms and inadequate multi-scale coordination. We propose ReflexSplit, a dual-stream framework with three key innovations. (1) Cross-scale Gated Fusion (CrGF) adaptively aggregates semantic priors, texture details, and decoder context across hierarchical depths, stabilizing gradient flow and maintaining feature consistency. (2) Layer Fusion-Separation Blocks (LFSB) alternate between fusion for shared structure extraction and differential separation for layer-specific disentanglement. Inspired by Differential Transformer, we extend attention cancellation to dual-stream separation via cross-stream subtraction. (3) Curriculum training progressively strengthens differential separation through depth-dependent initialization and epoch-wise warmup. Extensive experiments on synthetic and real-world benchmarks demonstrate state-of-the-art performance with superior perceptual quality and robust generalization. Our code is available at https://github.com/wuw2135/ReflexSplit.

---

### Harnessing Reasoning Trajectories for Hallucination Detection via Answer-agreement Representation Shaping

Authors: Jianxiong Zhang, Bing Guo, Yuming Jiang, Haobo Wang, Bo An, Sean Du
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17467v3

Large reasoning models (LRMs) often generate long, seemingly coherent reasoning traces yet still produce incorrect answers, making hallucination detection challenging. Although trajectories contain useful signals, directly using trace text or vanilla hidden states for detection is brittle: traces vary in form and detectors can overfit to superficial patterns rather than answer validity. We introduce Answer-agreement Representation Shaping (ARS), which learns detection-friendly trace-conditioned representations by explicitly encoding answer stability. ARS generates counterfactual answers through small latent interventions, specifically, perturbing the trace-boundary embedding, and labels each perturbation by whether the resulting answer agrees with the original. It then learns representations that bring answer-agreeing states together and separate answer-disagreeing ones, exposing latent instability indicative of hallucination risk. The shaped embeddings are plug-and-play with existing embedding-based detectors and require no human annotations during training. Experiments demonstrate that ARS consistently improves detection and achieves substantial gains over strong baselines. Code is available at: https://github.com/radiolab-ntu/ars_icml2026.

---

### Bayesian quantum sensing using graybox machine learning

Authors: Akram Youssry, Stefan Todd, Patrick Murton, Muhammad Junaid Arshad, Nicholas Werren, Alberto Peruzzo, Cristian Bonato
Published: 2026-01-24
Categories: quant-ph, cs.LG, eess.SP
arXiv: https://arxiv.org/abs/2601.17465v2

Quantum sensors offer significant advantages over classical devices in spatial resolution and sensitivity, enabling transformative applications across materials science, healthcare, and beyond. Their practical performance, however, is often constrained by unmodelled effects, including noise, imperfect state preparation, and non-ideal control fields. In this work, we report the first experimental implementation of a graybox modelling strategy for a solid-state open quantum system. The graybox framework integrates a physics-based system model with a data-driven description of experimental imperfections, achieving higher fidelity than purely analytical (whitebox) approaches while requiring fewer training resources than fully deep-learning blackbox models. We experimentally validate the method on the task of estimating a static magnetic field using a single-spin quantum sensor, performing Bayesian inference with a graybox model trained on prior experimental data. Using roughly 10,000 training datapoints, the graybox model yields several orders of magnitude improvement in mean squared error over the corresponding physics-only model, as well as significant improvement over a blackbox model with comparable size. These results are broadly applicable to a wide range of quantum sensing platforms, not limited to single-spin systems, and are particularly valuable for real-time adaptive protocols, where model inaccuracies can otherwise lead to suboptimal control and degraded performance.

---

### Entropy-Guided Agreement-Diversity: A Semi-Supervised Active Learning Framework for Fetal Head Segmentation in Ultrasound

Authors: Fangyijie Wang, Siteng Ma, Guénolé Silvestre, Kathleen M. Curran
Published: 2026-01-24
Categories: eess.IV, cs.CV
arXiv: https://arxiv.org/abs/2601.17460v1

Fetal ultrasound (US) data is often limited due to privacy and regulatory restrictions, posing challenges for training deep learning (DL) models. While semi-supervised learning (SSL) is commonly used for fetal US image analysis, existing SSL methods typically rely on random limited selection, which can lead to suboptimal model performance by overfitting to homogeneous labeled data. To address this, we propose a two-stage Active Learning (AL) sampler, Entropy-Guided Agreement-Diversity (EGAD), for fetal head segmentation. Our method first selects the most uncertain samples using predictive entropy, and then refines the final selection using the agreement-diversity score combining cosine similarity and mutual information. Additionally, our SSL framework employs a consistency learning strategy with feature downsampling to further enhance segmentation performance. In experiments, SSL-EGAD achieves an average Dice score of 94.57\% and 96.32\% on two public datasets for fetal head segmentation, using 5\% and 10\% labeled data for training, respectively. Our method outperforms current SSL models and showcases consistent robustness across diverse pregnancy stage data. The code is available on \href{https://github.com/13204942/Semi-supervised-EGAD}{GitHub}.

---

### Embodiment-Induced Coordination Regimes in Tabular Multi-Agent Q-Learning

Authors: Muhammad Ahmed Atif, Nehal Naeem Haji, Mohammad Shahid Shaikh, Muhammad Ebad Atif
Published: 2026-01-24
Categories: cs.MA, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.17454v2

Centralized value learning underlies a broad class of multi-agent reinforcement learning methods, but its claimed advantage is typically evaluated in settings that confound coordination structure with function approximation and partial observability. We isolate coordination structure in a fully tabular 8x8 predator-prey gridworld with explicit speed and stamina constraints, comparing all four pairings of Independent and Centralized Q-Learning across three kinematic regimes over 10 seeds. Fully independent learning (IQL-IQL) yields shorter episodes and higher predator returns than the fully centralized configuration (CQL-CQL) in every regime and every seed (Wilcoxon p = 0.00195, Cliff's delta = 1.0). Asymmetric IQL-CQL pairings produce coordination breakdowns that persist across the 40,000 episode training budget rather than resolving as transient instability. A best-response test against frozen IQL-IQL predators shows that even the strongest configuration has not converged to equilibrium at 40k episodes, so between-configuration differences reflect learning dynamics under a shared budget rather than end-state performance. We propose a mechanism we call temporal synchronization lock: a shared value function couples all agent decisions, so when one agent is stamina-limited the joint Q-value forces capable partners into suboptimal waits, while independent learners continue asynchronous pursuit. Because this pathology arises from credit assignment rather than function approximation, we conjecture analogous effects may arise in deep MARL methods that centralize credit during training. Centralized coordination is not uniformly beneficial; its advantage over independent learning shrinks or disappears under embodiment constraints, and mixed centralized-independent pairings can perform worse than either uniform choice.

---

### DREAM: Dual-Standard Semantic Homogeneity with Dynamic Optimization for Graph Learning with Label Noise

Authors: Yusheng Zhao, Jiaye Xie, Qixin Zhang, Weizhi Zhang, Xiao Luo, Zhiping Xiao, Philip S. Yu, Ming Zhang
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17449v1

Graph neural networks (GNNs) have been widely used in various graph machine learning scenarios. Existing literature primarily assumes well-annotated training graphs, while the reliability of labels is not guaranteed in real-world scenarios. Recently, efforts have been made to address the problem of graph learning with label noise. However, existing methods often (i) struggle to distinguish between reliable and unreliable nodes, and (ii) overlook the relational information embedded in the graph topology. To tackle this problem, this paper proposes a novel method, Dual-Standard Semantic Homogeneity with Dynamic Optimization (DREAM), for reliable, relation-informed optimization on graphs with label noise. Specifically, we design a relation-informed dynamic optimization framework that iteratively reevaluates the reliability of each labeled node in the graph during the optimization process according to the relation of the target node and other nodes. To measure this relation comprehensively, we propose a dual-standard selection strategy that selects a set of anchor nodes based on both node proximity and graph topology. Subsequently, we compute the semantic homogeneity between the target node and the anchor nodes, which serves as guidance for optimization. We also provide a rigorous theoretical analysis to justify the design of DREAM. Extensive experiments are performed on six graph datasets across various domains under three types of graph label noise against competing baselines, and the results demonstrate the effectiveness of the proposed DREAM.

---

### Clustering-driven Memory Compression for On-device Large Language Models

Authors: Ondrej Bohdal, Pramit Saha, Umberto Michieli, Mete Ozay, Taha Ceritli
Published: 2026-01-24
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.17443v1

Large language models (LLMs) often rely on user-specific memories distilled from past interactions to enable personalized generation. A common practice is to concatenate these memories with the input prompt, but this approach quickly exhausts the limited context available in on-device LLMs. Compressing memories by averaging can mitigate context growth, yet it frequently harms performance due to semantic conflicts across heterogeneous memories. In this work, we introduce a clustering-based memory compression strategy that balances context efficiency and personalization quality. Our method groups memories by similarity and merges them within clusters prior to concatenation, thereby preserving coherence while reducing redundancy. Experiments demonstrate that our approach substantially lowers the number of memory tokens while outperforming baseline strategies such as naive averaging or direct concatenation. Furthermore, for a fixed context budget, clustering-driven merging yields more compact memory representations and consistently enhances generation quality.

---

### A new approach for combined model class selection and parameters learning for auto-regressive neural models

Authors: Corrado Sgadari, Alessio La Bella, Marcello Farina
Published: 2026-01-24
Categories: eess.SY, cs.LG
arXiv: https://arxiv.org/abs/2601.17442v1

This work introduces a novel approach for the joint selection of model structure and parameter learning for nonlinear dynamical systems identification. Focusing on a specific Recurrent Neural Networks (RNNs) family, i.e., Nonlinear Auto-Regressive with eXogenous inputs Echo State Networks (NARXESNs), the method allows to simultaneously select the optimal model class and learn model parameters from data through a new set-membership (SM) based procedure. The results show the effectiveness of the approach in identifying parsimonious yet accurate models suitable for control applications. Moreover, the proposed framework enables a robust training strategy that explicitly accounts for bounded measurement noise and enhances model robustness by allowing data-consistent evaluation of simulation performance during parameter learning, a process generally NP-hard for models with autoregressive components.

---

### Data-driven Clustering and Merging of Adapters for On-device Large Language Models

Authors: Ondrej Bohdal, Taha Ceritli, Mete Ozay, Jijoong Moon, Kyeng-Hun Lee, Hyeonmok Ko, Umberto Michieli
Published: 2026-01-24
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2601.17441v1

On-device large language models commonly employ task-specific adapters (e.g., LoRAs) to deliver strong performance on downstream tasks. While storing all available adapters is impractical due to memory constraints, mobile devices typically have sufficient capacity to store a limited number of these parameters. This raises a critical challenge: how to select representative adapters that generalize well across multiple tasks - a problem that remains unexplored in existing literature. We propose a novel method D2C for adapter clustering that leverages minimal task-specific examples (e.g., 10 per task) and employs an iterative optimization process to refine cluster assignments. The adapters within each cluster are merged, creating multi-task adapters deployable on resource-constrained devices. Experimental results demonstrate that our method effectively boosts performance for considered storage budgets.

---

### UniGRec: Unified Generative Recommendation with Soft Identifiers for End-to-End Optimization

Authors: Jialei Li, Yang Zhang, Yimeng Bai, Shuai Zhu, Ziqi Xue, Xiaoyan Zhao, Dingxian Wang, Frank Yang, Andrew Rabinovich, Xiangnan He
Published: 2026-01-24
Categories: cs.IR, cs.LG
arXiv: https://arxiv.org/abs/2601.17438v1

Generative recommendation has recently emerged as a transformative paradigm that directly generates target items, surpassing traditional cascaded approaches. It typically involves two components: a tokenizer that learns item identifiers and a recommender trained on them. Existing methods often decouple tokenization from recommendation or rely on asynchronous alternating optimization, limiting full end-to-end alignment. To address this, we unify the tokenizer and recommender under the ultimate recommendation objective via differentiable soft item identifiers, enabling joint end-to-end training. However, this introduces three challenges: training-inference discrepancy due to soft-to-hard mismatch, item identifier collapse from codeword usage imbalance, and collaborative signal deficiency due to an overemphasis on fine-grained token-level semantics. To tackle these challenges, we propose UniGRec, a unified generative recommendation framework that addresses them from three perspectives. UniGRec employs Annealed Inference Alignment during tokenization to smoothly bridge soft training and hard inference, a Codeword Uniformity Regularization to prevent identifier collapse and encourage codebook diversity, and a Dual Collaborative Distillation mechanism that distills collaborative priors from a lightweight teacher model to jointly guide both the tokenizer and the recommender. Extensive experiments on real-world datasets demonstrate that UniGRec consistently outperforms state-of-the-art baseline methods. Our codes are available at https://github.com/Jialei-03/UniGRec.

---

### Towards a Declarative Agentic Layer for Intelligent Agents in MCP-Based Server Ecosystems

Authors: Maria Jesus Rodriguez-Sanchez, Manuel Noguera, Angel Ruiz-Zafra, Kawtar Benghazi
Published: 2026-01-24
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2601.17435v1

Recent advances in Large Language Models (LLMs) have enabled the development of increasingly complex agentic and multi-agent systems capable of planning, tool use and task decomposition. However, empirical evidence shows that many of these systems suffer from fundamental reliability issues, including hallucinated actions, unexecutable plans and brittle coordination. Crucially, these failures do not stem from limitations of the underlying models themselves, but from the absence of explicit architectural structure linking goals, capabilities and execution. This paper presents a declarative, model-independent architectural layer for grounded agentic workflows that addresses this gap. The proposed layer, referred to as DALIA (Declarative Agentic Layer for Intelligent Agents), formalises executable capabilities, exposes tasks through a declarative discovery protocol, maintains a federated directory of agents and their execution resources, and constructs deterministic task graphs grounded exclusively in declared operations. By enforcing a clear separation between discovery, planning and execution, the architecture constrains agent behaviour to a verifiable operational space, reducing reliance on speculative reasoning and free-form coordination. We present the architecture and design principles of the proposed layer and illustrate its operation through a representative task-oriented scenario, demonstrating how declarative grounding enables reproducible and verifiable agentic workflows across heterogeneous environments.

---

### The 17% Gap: Quantifying Epistemic Decay in AI-Assisted Survey Papers

Authors: H. Kemal İlter
Published: 2026-01-24
Categories: cs.CY, cs.AI, cs.CL, cs.DL
arXiv: https://arxiv.org/abs/2601.17431v1

The adoption of Large Language Models (LLMs) in scientific writing promises efficiency but risks introducing informational entropy. While "hallucinated papers" are a known artifact, the systematic degradation of valid citation chains remains unquantified. We conducted a forensic audit of 50 recent survey papers in Artificial Intelligence (N=5,514 citations) published between September 2024 and January 2026. We utilized a hybrid verification pipeline combining DOI resolution, Crossref metadata analysis, Semantic Scholar queries, and fuzzy text matching to distinguish between formatting errors ("Sloppiness") and verifiable non-existence ("Phantoms). We detect a persistent 17.0% Phantom Rate -- citations that cannot be resolved to any digital object despite aggressive forensic recovery. Diagnostic categorization reveals three distinct failure modes: pure hallucinations (5.1%), hallucinated identifiers with valid titles (16.4%), and parsing-induced matching failures (78.5%). Longitudinal analysis reveals a flat trend (+0.07 pp/month), suggesting that high-entropy citation practices have stabilized as an endemic feature of the field. The scientific citation graph in AI survey literature exhibits "link rot" at scale. This suggests a mechanism where AI tools act as "lazy research assistants," retrieving correct titles but hallucinating metadata, thereby severing the digital chain of custody required for reproducible science.

---

### Active Hypothesis Testing for Correlated Combinatorial Anomaly Detection

Authors: Zichuan Yang, Yiming Xing
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17430v1

We study the problem of identifying an anomalous subset of streams under correlated noise, motivated by monitoring and security in cyber-physical systems. This problem can be viewed as a form of combinatorial pure exploration, where each stream plays the role of an arm and measurements must be allocated sequentially under uncertainty. Existing combinatorial bandit and hypothesis testing methods typically assume independent observations and fail to exploit correlation for efficient measurement design. We propose ECC-AHT, an adaptive algorithm that selects continuous, constrained measurements to maximize Chernoff information between competing hypotheses, enabling active noise cancellation through differential sensing. ECC-AHT achieves optimal sample complexity guarantees and significantly outperforms state-of-the-art baselines in both synthetic and real-world correlated environments. The code is available on https://github.com/VincentdeCristo/ECC-AHT

---

### Coronary Artery Segmentation and Vessel-Type Classification in X-Ray Angiography

Authors: Mehdi Yousefzadeh, Siavash Shirzadeh Barough, Ashkan Fakharifar, Yashar Tayyarazad, Narges Eghbali, Mohaddeseh Mozaffari, Hoda Taeb, Negar Sadat Rafiee Tabatabaee, Parsa Esfahanian, Ghazaleh Sadeghi Gohar, Amineh Safavirad, Saeideh Mazloomzadeh et al.
Published: 2026-01-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.17429v1

X-ray coronary angiography (XCA) is the clinical reference standard for assessing coronary artery disease, yet quantitative analysis is limited by the difficulty of robust vessel segmentation in routine data. Low contrast, motion, foreshortening, overlap, and catheter confounding degrade segmentation and contribute to domain shift across centers. Reliable segmentation, together with vessel-type labeling, enables vessel-specific coronary analytics and downstream measurements that depend on anatomical localization. From 670 cine sequences (407 subjects), we select a best frame near peak opacification using a low-intensity histogram criterion and apply joint super-resolution and enhancement. We benchmark classical Meijering, Frangi, and Sato vesselness filters under per-image oracle tuning, a single global mean setting, and per-image parameter prediction via Support Vector Regression (SVR). Neural baselines include U-Net, FPN, and a Swin Transformer, trained with coronary-only and merged coronary+catheter supervision. A second stage assigns vessel identity (LAD, LCX, RCA). External evaluation uses the public DCA1 cohort. SVR per-image tuning improves Dice over global means for all classical filters (e.g., Frangi: 0.759 vs. 0.741). Among deep models, FPN attains 0.914+/-0.007 Dice (coronary-only), and merged coronary+catheter labels further improve to 0.931+/-0.006. On DCA1 as a strict external test, Dice drops to 0.798 (coronary-only) and 0.814 (merged), while light in-domain fine-tuning recovers to 0.881+/-0.014 and 0.882+/-0.015. Vessel-type labeling achieves 98.5% accuracy (Dice 0.844) for RCA, 95.4% (0.786) for LAD, and 96.2% (0.794) for LCX. Learned per-image tuning strengthens classical pipelines, while high-resolution FPN models and merged-label supervision improve stability and external transfer with modest adaptation.

---

### A Syllogistic Probe: Tracing the Evolution of Logic Reasoning in Large Language Models

Authors: Zhengqing Zang, Yuqi Ding, Yanmei Gu, Changkai Song, Zhengkai Yang, Guoping Du, Junbo Zhao, Haobo Wang
Published: 2026-01-24
Categories: cs.AI, cs.LO
arXiv: https://arxiv.org/abs/2601.17426v1

Human logic has gradually shifted from intuition-driven inference to rigorous formal systems. Motivated by recent advances in large language models (LLMs), we explore whether LLMs exhibit a similar evolution in the underlying logical framework. Using existential import as a probe, we for evaluate syllogism under traditional and modern logic. Through extensive experiments of testing SOTA LLMs on a new syllogism dataset, we have some interesting findings: (i) Model size scaling promotes the shift toward modern logic; (ii) Thinking serves as an efficient accelerator beyond parameter scaling; (iii) the Base model plays a crucial role in determining how easily and stably this shift can emerge. Beyond these core factors, we conduct additional experiments for in-depth analysis of properties of current LLMs on syllogistic reasoning.

---

### Oops, Wait: Discourse Tokens Matter in Reasoning Model

Authors: Jaehui Hwang, Byeongho Heo, Sangdoo Yun, Dongyoon Han
Published: 2026-01-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.17421v2

Recent studies suggest that even data-efficient training with ($\simeq$1K) reasoning trajectories can induce non-trivial reasoning capabilities in large language models through post-training. Such training corpora often contain iconic tokens such as "wait", "so", and "alternatively", which frequently appear in reasoning trajectories and may play a role in this process. This paper focuses on characterizing observable token-level patterns in post-training and a case study of how data-efficient supervised fine-tuning (SFT) differs from, and falls short of, large-scale post-training. To this end, we first identify tokens that correlate with correct answers along reasoning trajectories across models and training setups. We then focus on the distribution and (functional) roles of the "wait" token to primarily study the model trained in a data-efficient manner compared with the counterpart. Our study finds that discourse tokens are associated with correctness and a reasoning accuracy jump, even in data-efficient SFT. This suggests data-efficient SFT can partially reproduce discourse-token patterns to mimic meaningful reasoning behavior, but the patterns are less aligned with high-confidence answer transitions than those from large-scale post-training.

---

### CoT-Seg: Rethinking Segmentation with Chain-of-Thought Reasoning and Self-Correction

Authors: Shiu-hong Kao, Chak Ho Huang, Huaiqian Liu, Yu-Wing Tai, Chi-Keung Tang
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17420v1

Existing works of reasoning segmentation often fall short in complex cases, particularly when addressing complicated queries and out-of-domain images. Inspired by the chain-of-thought reasoning, where harder problems require longer thinking steps/time, this paper aims to explore a system that can think step-by-step, look up information if needed, generate results, self-evaluate its own results, and refine the results, in the same way humans approach harder questions. We introduce CoT-Seg, a training-free framework that rethinks reasoning segmentation by combining chain-of-thought reasoning with self-correction. Instead of fine-tuning, CoT-Seg leverages the inherent reasoning ability of pre-trained MLLMs (GPT-4o) to decompose queries into meta-instructions, extract fine-grained semantics from images, and identify target objects even under implicit or complex prompts. Moreover, CoT-Seg incorporates a self-correction stage: the model evaluates its own segmentation against the original query and reasoning trace, identifies mismatches, and iteratively refines the mask. This tight integration of reasoning and correction significantly improves reliability and robustness, especially in ambiguous or error-prone cases. Furthermore, our CoT-Seg framework allows easy incorporation of retrieval-augmented reasoning, enabling the system to access external knowledge when the input lacks sufficient information. To showcase CoT-Seg's ability to handle very challenging cases ,we introduce a new dataset ReasonSeg-Hard. Our results highlight that combining chain-of-thought reasoning, self-correction, offers a powerful paradigm for vision-language integration driven segmentation.

---

### Semantic-Aware Task Clustering for Federated Cooperative Multi-Task Semantic Communication

Authors: Ahmad Halimi Razlighi, Pallavi Dhingra, Edgar Beck, Bho Matthiesen, Armin Dekorsy
Published: 2026-01-24
Categories: eess.SP, cs.IT, stat.ML
arXiv: https://arxiv.org/abs/2601.17419v1

Task-oriented semantic communication (SemCom) prioritizes task execution over accurate symbol reconstruction and is well-suited to emerging intelligent applications. Cooperative multi-task SemCom (CMT-SemCom) further improves task execution performance. However, (1) demonstrates that cooperative multi-tasking can be either constructive or destructive. Moreover, the existing CMT-SemCom framework is not directly applicable to distributed multi-user scenarios, such as non-terrestrial satellite networks, where each satellite employs an individual semantic encoder. In this paper, we extend our earlier CMT-SemCom framework to distributed settings by proposing a federated learning (FL) based CMT-SemCom that enables cooperative multi-tasking across distributed users. Moreover, to address performance degradation caused by negative information transfer among heterogeneous tasks, we propose a semantic-aware task clustering method integrated in the FL process to ensure constructive cooperation based on an information-theoretic approach. Unlike common clustering methods that rely on high-dimensional data or feature space similarity, our proposed approach operates in the low-dimensional semantic domain to identify meaningful task relationships. Simulation results based on a LEO satellite network setup demonstrate the effectiveness of our approach and performance gain over unclustered FL and individual single-task SemCom.

---

### Cloud-Enabled IoT System for Real-Time Environmental Monitoring and Remote Device Control Using Firebase

Authors: Abdul Hasib, A. S. M. Ahsanul Sarkar Akib
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17414v1

The proliferation of Internet of Things (IoT) devices has created unprecedented opportunities for remote monitoring and control applications across various domains. Traditional monitoring systems often suffer from limitations in real-time data accessibility, remote controllability, and cloud integration. This paper presents a cloud-enabled IoT system that leverages Google's Firebase Realtime Database for synchronized environmental monitoring and device control. The system utilizes an ESP32 microcontroller to interface with a DHT22 temperature/humidity sensor and an HC-SR04 ultrasonic distance sensor, while enabling remote control of two LED indicators through a cloud-based interface. Real-time sensor data is transmitted to Firebase, providing a synchronized platform accessible from multiple devices simultaneously. Experimental results demonstrate reliable data transmission with 99.2\% success rate, real-time control latency under 1.5 seconds, and persistent data storage for historical analysis. The system architecture offers a scalable framework for various IoT applications, from smart home automation to industrial monitoring, with a total implementation cost of \$32.50. The integration of Firebase provides robust cloud capabilities without requiring complex server infrastructure, making advanced IoT applications accessible to developers and researchers with limited resources.

---

### Source-Free Domain Adaptation by Optimizing Batch-Wise Cosine Similarity

Authors: Harsharaj Pathak, Vineeth N Balasubramanian
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17408v1

Source-Free Domain Adaptation (SFDA) is an emerging area of research that aims to adapt a model trained on a labeled source domain to an unlabeled target domain without accessing the source data. Most of the successful methods in this area rely on the concept of neighborhood consistency but are prone to errors due to misleading neighborhood information. In this paper, we explore this approach from the point of view of learning more informative clusters and mitigating the effect of noisy neighbors using a concept called neighborhood signature, and demonstrate that adaptation can be achieved using just a single loss term tailored to optimize the similarity and dissimilarity of predictions of samples in the target domain. In particular, our proposed method outperforms existing methods in the challenging VisDA dataset while also yielding competitive results on other benchmark datasets.

---

### Efficient Dilated Squeeze and Excitation Neural Operator for Differential Equations

Authors: Prajwal Chauhan, Salah Eddine Choutri, Saif Eddin Jabari
Published: 2026-01-24
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.17407v1

Fast and accurate surrogates for physics-driven partial differential equations (PDEs) are essential in fields such as aerodynamics, porous media design, and flow control. However, many transformer-based models and existing neural operators remain parameter-heavy, resulting in costly training and sluggish deployment. We propose D-SENO (Dilated Squeeze-Excitation Neural Operator), a lightweight operator learning framework for efficiently solving a wide range of PDEs, including airfoil potential flow, Darcy flow in porous media, pipe Poiseuille flow, and incompressible Navier Stokes vortical fields. D-SENO combines dilated convolution (DC) blocks with squeeze-and-excitation (SE) modules to jointly capture wide receptive fields and dynamics alongside channel-wise attention, enabling both accurate and efficient PDE inference. Carefully chosen dilation rates allow the receptive field to focus on critical regions, effectively modeling long-range physical dependencies. Meanwhile, the SE modules adaptively recalibrate feature channels to emphasize dynamically relevant scales. Our model achieves training speed of up to approximately $20\times$ faster than standard transformer-based models and neural operators, while also surpassing (or matching) them in accuracy across multiple PDE benchmarks. Ablation studies show that removing the SE modules leads to a slight drop in performance.

---

### HAAF: Hierarchical Adaptation and Alignment of Foundation Models for Few-Shot Pathology Anomaly Detection

Authors: Chunze Yang, Wenjie Zhao, Yue Tang, Junbo Lu, Jiusong Ge, Qidong Liu, Zeyu Gao, Chen Li
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17405v1

Precision pathology relies on detecting fine-grained morphological abnormalities within specific Regions of Interest (ROIs), as these local, texture-rich cues - rather than global slide contexts - drive expert diagnostic reasoning. While Vision-Language (V-L) models promise data efficiency by leveraging semantic priors, adapting them faces a critical Granularity Mismatch, where generic representations fail to resolve such subtle defects. Current adaptation methods often treat modalities as independent streams, failing to ground semantic prompts in ROI-specific visual contexts. To bridge this gap, we propose the Hierarchical Adaptation and Alignment Framework (HAAF). At its core is a novel Cross-Level Scaled Alignment (CLSA) mechanism that enforces a sequential calibration order: visual features first inject context into text prompts to generate content-adaptive descriptors, which then spatially guide the visual encoder to spotlight anomalies. Additionally, a dual-branch inference strategy integrates semantic scores with geometric prototypes to ensure stability in few-shot settings. Experiments on four benchmarks show HAAF significantly outperforms state-of-the-art methods and effectively scales with domain-specific backbones (e.g., CONCH) in low-resource scenarios.

---

### ReLE: A Scalable System and Structured Benchmark for Diagnosing Capability Anisotropy in Chinese LLMs

Authors: Rui Fang, Jian Li, Wei Chen, Bin Hu, Ying-Cong Chen, Xin Tang, Liang Diao
Published: 2026-01-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.17399v1

Large Language Models (LLMs) have achieved rapid progress in Chinese language understanding, yet accurately evaluating their capabilities remains challenged by benchmark saturation and prohibitive computational costs. While static leaderboards provide snapshot rankings, they often mask the structural trade-offs between capabilities. In this work, we present ReLE (Robust Efficient Live Evaluation), a scalable system designed to diagnose Capability Anisotropy, the non-uniformity of model performance across domains. Using ReLE, we evaluate 304 models (189 commercial, 115 open-source) across a Domain $\times$ Capability orthogonal matrix comprising 207,843 samples. We introduce two methodological contributions to address current evaluation pitfalls: (1) A Symbolic-Grounded Hybrid Scoring Mechanism that eliminates embedding-based false positives in reasoning tasks; (2) A Dynamic Variance-Aware Scheduler based on Neyman allocation with noise correction, which reduces compute costs by 70\% compared to full-pass evaluations while maintaining a ranking correlation of $ρ=0.96$. Our analysis reveals that aggregate rankings are highly sensitive to weighting schemes: models exhibit a Rank Stability Amplitude (RSA) of 11.4 in ReLE versus $\sim$5.0 in traditional benchmarks, confirming that modern models are highly specialized rather than generally superior. We position ReLE not as a replacement for comprehensive static benchmarks, but as a high-frequency diagnostic monitor for the evolving model landscape.

---

### CLM-Bench: Benchmarking and Analyzing Cross-lingual Misalignment of LLMs in Knowledge Editing

Authors: Yucheng Hu, Wei Zhou, Juesi Xiao
Published: 2026-01-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.17397v1

Knowledge Editing (KE) has emerged as a promising paradigm for updating facts in Large Language Models (LLMs) without retraining. However, progress in Multilingual Knowledge Editing (MKE) is currently hindered by biased evaluation frameworks. We observe that existing MKE benchmarks are typically constructed by mechanically translating English-centric datasets into target languages (e.g., English-to-Chinese). This approach introduces translation artifacts and neglects culturally specific entities native to the target language, failing to reflect the true knowledge distribution of LLMs. To address this, we propose CLM-Bench, a culture-aware benchmark constructed using a native Chinese-first methodology. We curate 1,010 high-quality CounterFact pairs rooted in Chinese cultural contexts and align them with English counterparts. Using CLM-Bench, we conduct extensive experiments on representative LLMs (e.g., Llama-3, Qwen2) and reveal a significant Cross-lingual Misalignment: edits in one language function independently and fail to propagate to the other. We further provide a geometric explanation via layer-wise representation analysis, demonstrating that edit vectors for Chinese and English are nearly orthogonal -- residing in disjoint subspaces -- while mixed-lingual editing exhibits linear additivity of these vectors. Our findings challenge the effectiveness of current methods in cross-lingual transfer and underscore the importance of culturally native benchmarks.

---

### GO-OSC and VASH: Geometry-Aware Representation Learning for Early Degradation Detection in Oscillatory Systems

Authors: Vashista Nobaub
Published: 2026-01-24
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.17396v1

Early-stage degradation in oscillatory systems often manifests as geometric distortions of the dynamics, such as phase jitter, frequency drift, or loss of coherence, long before changes in signal energy are detectable. In this regime, classical energy-based diagnostics and unconstrained learned representations are structurally insensitive, leading to delayed or unstable detection. We introduce GO-OSC, a geometry-aware representation learning framework for oscillatory time series that enforces a canonical and identifiable latent parameterization, enabling stable comparison and aggregation across short, unlabeled windows. Building on this representation, we define a family of invariant linear geometric probes that target degradation-relevant directions in latent space. We provide theoretical results showing that under early phase-only degradation, energy-based statistics have zero first-order detection power, whereas geometric probes achieve strictly positive sensitivity. Our analysis characterizes when and why linear probing fails under non-identifiable representations and shows how canonicalization restores statistical detectability. Experiments on synthetic benchmarks and real vibration datasets validate the theory, demonstrating earlier detection, improved data efficiency, and robustness to operating condition changes.

---

### SMV-EAR: Bring Spatiotemporal Multi-View Representation Learning into Efficient Event-Based Action Recognition

Authors: Rui Fan, Weidong Hao
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17391v1

Event cameras action recognition (EAR) offers compelling privacy-protecting and efficiency advantages, where temporal motion dynamics is of great importance. Existing spatiotemporal multi-view representation learning (SMVRL) methods for event-based object recognition (EOR) offer promising solutions by projecting H-W-T events along spatial axis H and W, yet are limited by its translation-variant spatial binning representation and naive early concatenation fusion architecture. This paper reexamines the key SMVRL design stages for EAR and propose: (i) a principled spatiotemporal multi-view representation through translation-invariant dense conversion of sparse events, (ii) a dual-branch, dynamic fusion architecture that models sample-wise complementarity between motion features from different views, and (iii) a bio-inspired temporal warping augmentation that mimics speed variability of real-world human actions. On three challenging EAR datasets of HARDVS, DailyDVS-200 and THU-EACT-50-CHL, we show +7.0%, +10.7%, and +10.2% Top-1 accuracy gains over existing SMVRL EOR method with surprising 30.1% reduced parameters and 35.7% lower computations, establishing our framework as a novel and powerful EAR paradigm.

---

### ONRW: Optimizing inversion noise for high-quality and robust watermark

Authors: Xuan Ding, Xiu Yan, Chuanlong Xie, Yao Zhu
Published: 2026-01-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.17388v1

Watermarking methods have always been effective means of protecting intellectual property, yet they face significant challenges. Although existing deep learning-based watermarking systems can hide watermarks in images with minimal impact on image quality, they often lack robustness when encountering image corruptions during transmission, which undermines their practical application value. To this end, we propose a high-quality and robust watermark framework based on the diffusion model. Our method first converts the clean image into inversion noise through a null-text optimization process, and after optimizing the inversion noise in the latent space, it produces a high-quality watermarked image through an iterative denoising process of the diffusion model. The iterative denoising process serves as a powerful purification mechanism, ensuring both the visual quality of the watermarked image and enhancing the robustness of the watermark against various corruptions. To prevent the optimizing of inversion noise from distorting the original semantics of the image, we specifically introduced self-attention constraints and pseudo-mask strategies. Extensive experimental results demonstrate the superior performance of our method against various image corruptions. In particular, our method outperforms the stable signature method by an average of 10\% across 12 different image transformations on COCO datasets. Our codes are available at https://github.com/920927/ONRW.

---

### Generation-Step-Aware Framework for Cross-Modal Representation and Control in Multilingual Speech-Text Models

Authors: Toshiki Nakai, Varsha Suresh, Vera Demberg
Published: 2026-01-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.17387v3

Multilingual speech-text models rely on cross-modal language alignment to transfer knowledge between speech and text, but it remains unclear whether this reflects shared computation for the same language or modality-specific processing. We introduce a generation-step-aware framework for evaluating cross-modal computation that (i) identifies language-selective neurons for each modality at different decoding steps, (ii) decomposes them into language-representation and language-control roles, and (iii) enables cross-modal comparison via overlap measures and causal intervention. Applying our framework to SeamlessM4T, and additionally evaluating its generality on the decoder-only model Qwen2-Audio, we find that cross-modal language-representation alignment is highly model dependent. SeamlessM4T exhibits pronounced generation-step-dependent specialization, where only 5-7% of language-representation neurons are shared across modalities and overlap shifts from same-language to typologically related languages during autoregressive generation, whereas Qwen2-Audio maintains substantially larger cross-modal sharing (45-47%) together with stable, strongly language-specific alignment across decoding steps. In contrast, language-control neurons identified at later decoding steps exhibit progressively stronger cross-modal transfer from speech to text in SeamlessM4T. Together, these results demonstrate that generation-step-aware analysis can reveal both shared and model-specific patterns of cross-modal computation.

---

### Physical Prompt Injection Attacks on Large Vision-Language Models

Authors: Chen Ling, Kai Hu, Hangcheng Liu, Xingshuo Han, Tianwei Zhang, Changhai Ou
Published: 2026-01-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.17383v1

Large Vision-Language Models (LVLMs) are increasingly deployed in real-world intelligent systems for perception and reasoning in open physical environments. While LVLMs are known to be vulnerable to prompt injection attacks, existing methods either require access to input channels or depend on knowledge of user queries, assumptions that rarely hold in practical deployments. We propose the first Physical Prompt Injection Attack (PPIA), a black-box, query-agnostic attack that embeds malicious typographic instructions into physical objects perceivable by the LVLM. PPIA requires no access to the model, its inputs, or internal pipeline, and operates solely through visual observation. It combines offline selection of highly recognizable and semantically effective visual prompts with strategic environment-aware placement guided by spatiotemporal attention, ensuring that the injected prompts are both perceivable and influential on model behavior. We evaluate PPIA across 10 state-of-the-art LVLMs in both simulated and real-world settings on tasks including visual question answering, planning, and navigation, PPIA achieves attack success rates up to 98%, with strong robustness under varying physical conditions such as distance, viewpoint, and illumination. Our code is publicly available at https://github.com/2023cghacker/Physical-Prompt-Injection-Attack.

---

### Prompt and Circumstances: Evaluating the Efficacy of Human Prompt Inference in AI-Generated Art

Authors: Khoi Trinh, Scott Seidenberger, Joseph Spracklen, Raveen Wijewickrama, Bimal Viswanath, Murtuza Jadliwala, Anindya Maiti
Published: 2026-01-24
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2601.17379v1

The emerging field of AI-generated art has witnessed the rise of prompt marketplaces, where creators can purchase, sell, or share prompts to generate unique artworks. These marketplaces often assert ownership over prompts, claiming them as intellectual property. This paper investigates whether concealed prompts sold on prompt marketplaces can be considered bona fide intellectual property, given that humans and AI tools may be able to infer the prompts based on publicly advertised sample images accompanying each prompt on sale. Specifically, our study aims to assess (i) how accurately humans can infer the original prompt solely by examining an AI-generated image, with the goal of generating images similar to the original image, and (ii) the possibility of improving upon individual human and AI prompt inferences by crafting combined human and AI prompts with the help of a large language model. Although previous research has explored AI-driven prompt inference and protection strategies, our work is the first to incorporate a human subject study and examine collaborative human-AI prompt inference in depth. Our findings indicate that while prompts inferred by humans and prompts inferred through a combined human and AI effort can generate images with a moderate level of similarity, they are not as successful as using the original prompt. Moreover, combining human- and AI-inferred prompts using our suggested merging techniques did not improve performance over purely human-inferred prompts.

---

### Res-MIA: A Training-Free Resolution-Based Membership Inference Attack on Federated Learning Models

Authors: Mohammad Zare, Pirooz Shamsinejadbabaki
Published: 2026-01-24
Categories: cs.CR, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.17378v1

Membership inference attacks (MIAs) pose a serious threat to the privacy of machine learning models by allowing adversaries to determine whether a specific data sample was included in the training set. Although federated learning (FL) is widely regarded as a privacy-aware training paradigm due to its decentralized nature, recent evidence shows that the final global model can still leak sensitive membership information through black-box access. In this paper, we introduce Res-MIA, a novel training-free and black-box membership inference attack that exploits the sensitivity of deep models to high-frequency input details. Res-MIA progressively degrades the input resolution using controlled downsampling and restoration operations, and analyzes the resulting confidence decay in the model's predictions. Our key insight is that training samples exhibit a significantly steeper confidence decline under resolution erosion compared to non-member samples, revealing a robust membership signal. Res-MIA requires no shadow models, no auxiliary data, and only a limited number of forward queries to the target model. We evaluate the proposed attack on a federated ResNet-18 trained on CIFAR-10, where it consistently outperforms existing training-free baselines and achieves an AUC of up to 0.88 with minimal computational overhead. These findings highlight frequency-sensitive overfitting as an important and previously underexplored source of privacy leakage in federated learning, and emphasize the need for privacy-aware model designs that reduce reliance on fine-grained, non-robust input features.

---

### WarrantScore: Modeling Warrants between Claims and Evidence for Substantiation Evaluation in Peer Reviews

Authors: Kiyotada Mori, Shohei Tanaka, Tosho Hirasawa, Tadashi Kozuno, Koichiro Yoshino, Yoshitaka Ushiku
Published: 2026-01-24
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.17377v1

The scientific peer-review process is facing a shortage of human resources due to the rapid growth in the number of submitted papers. The use of language models to reduce the human cost of peer review has been actively explored as a potential solution to this challenge. A method has been proposed to evaluate the level of substantiation in scientific reviews in a manner that is interpretable by humans. This method extracts the core components of an argument, claims and evidence, and assesses the level of substantiation based on the proportion of claims supported by evidence. The level of substantiation refers to the extent to which claims are based on objective facts. However, when assessing the level of substantiation, simply detecting the presence or absence of supporting evidence for a claim is insufficient; it is also necessary to accurately assess the logical inference between a claim and its evidence. We propose a new evaluation metric for scientific review comments that assesses the logical inference between claims and evidence. Experimental results show that the proposed method achieves a higher correlation with human scores than conventional methods, indicating its potential to better support the efficiency of the peer-review process.

---

### Diversified Scaling Inference in Time Series Foundation Models

Authors: Ruijin Hua, Zichuan Liu, Kun Zhang, Yiyuan Yang
Published: 2026-01-24
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.17376v1

The advancement of Time Series Foundation Models (TSFMs) has been driven primarily by large-scale pre-training, but inference-time compute potential remains largely untapped. This work systematically investigates two questions: how do TSFMs behave under standard sampling-based inference scaling, and can controlled sampling diversity enhance performance? We first examine the properties of TSFMs under standard sampling often fail to adhere to scaling laws due to insufficient exploration of the solution space. Building on this, we then delve into diversified inference scaling via tailored time series perturbations to expand the generative distribution's support. We theoretically analyze the diversity-fidelity trade-off and derive a critical sample threshold for diversified sampling to outperform standard sampling. Extensive experiments across various TSFMs and datasets show proper diversified inference scaling yields substantial performance gains without parameter updates, establishing inference design as a critical, compute-efficient dimension of TSFM optimization. As an application, we propose RobustMSE, a rigorous metric to quantify the headroom performance of TSFM under a fixed budget. Overall, our findings clarify these factor interactions, enabling reliable performance via diverse large-scale inference time series in parallel environments without re-training TSFMs.

---

### Error Analysis of Bayesian Inverse Problems with Generative Priors

Authors: Bamdad Hosseini, Ziqi Huang
Published: 2026-01-24
Categories: stat.ML, cs.LG, math.NA
arXiv: https://arxiv.org/abs/2601.17374v2

Data-driven methods for the solution of inverse problems have become widely popular in recent years thanks to the rise of machine learning techniques. A popular approach concerns the training of a generative model on additional data to learn a bespoke prior for the problem at hand. In this article we present an analysis for such problems by presenting quantitative error bounds for minimum Wasserstein-2 generative models for the prior. We show that under some assumptions, the error in the posterior due to the generative prior will inherit the same rate as the prior with respect to the Wasserstein-1 distance. We further present numerical experiments that verify that aspects of our error analysis manifests in some benchmarks followed by an elliptic PDE inverse problem where a generative prior is used to model a non-stationary field.

---

### Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers

Authors: Zecheng Tang, Quantong Qiu, Yi Yang, Zhiyi Hong, Haiya Xiang, Kebin Liu, Qingqing Dang, Juntao Li, Min Zhang
Published: 2026-01-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.17367v2

The quadratic complexity of standard attention mechanisms poses a significant scalability bottleneck for large language models (LLMs) in long-context scenarios. While hybrid attention strategies that combine sparse and full attention within a single model offer a viable solution, they typically employ static computation ratios (i.e., fixed proportions of sparse versus full attention) and fail to adapt to the varying sparsity sensitivities of downstream tasks during inference. To address this issue, we propose Elastic Attention, which allows the model to dynamically adjust its overall sparsity based on the input. This is achieved by integrating a lightweight Attention Router into the existing pretrained model, which dynamically assigns each attention head to different computation modes. Within only 12 hours of training on 8xA800 GPUs, our method enables models to achieve both strong performance and efficient inference. Experiments across three long-context benchmarks on widely-used LLMs demonstrate the superiority of our method.

---

### UCAD: Uncertainty-guided Contour-aware Displacement for semi-supervised medical image segmentation

Authors: Chengbo Ding, Fenghe Tang, Shaohua Kevin Zhou
Published: 2026-01-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.17366v1

Existing displacement strategies in semi-supervised segmentation only operate on rectangular regions, ignoring anatomical structures and resulting in boundary distortions and semantic inconsistency. To address these issues, we propose UCAD, an Uncertainty-Guided Contour-Aware Displacement framework for semi-supervised medical image segmentation that preserves contour-aware semantics while enhancing consistency learning. Our UCAD leverages superpixels to generate anatomically coherent regions aligned with anatomy boundaries, and an uncertainty-guided selection mechanism to selectively displace challenging regions for better consistency learning. We further propose a dynamic uncertainty-weighted consistency loss, which adaptively stabilizes training and effectively regularizes the model on unlabeled regions. Extensive experiments demonstrate that UCAD consistently outperforms state-of-the-art semi-supervised segmentation methods, achieving superior segmentation accuracy under limited annotation. The code is available at:https://github.com/dcb937/UCAD.

---

### Parameter Efficient Fine Tuning Llama 3.1 for Answering Arabic Legal Questions: A Case Study on Jordanian Laws

Authors: Mohammed Fasha, Bassam Hammo, Bilal Sowan, Husam Barham, Esam Nsour
Published: 2026-01-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.17364v1

This study uses Jordanian law as a case study to explore the fine-tuning of the Llama-3.1 large language model for Arabic question-answering. Two versions of the model - Llama-3.1-8B-bnb-4bit and Llama-3.1-8B-Instruct-bnb-4bit - were fine-tuned using parameter-efficient fine-tuning (PEFT) with LoRA adapters and 4-bit quantized models, leveraging the Unsloth framework for accelerated and resource-efficient training. A custom dataset of 6000 legal question-answer pairs was curated from Jordanian laws and formatted into structured prompts. Performance was evaluated using the BLEU and the ROUGE metrics to compare the fine-tuned models to their respective base versions. Results demonstrated improved legal reasoning and accuracy while achieving resource efficiency through quantization and optimized fine-tuning strategies. This work underscores the potential of adapting large language models for Arabic legal domains and highlights effective techniques for fine-tuning domain-specific tasks.

---

### Do readers prefer AI-generated Italian short stories?

Authors: Michael Farrell
Published: 2026-01-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.17363v3

This study investigates whether readers prefer AI-generated short stories in Italian over one written by a renowned Italian author. In a blind setup, 20 participants read and evaluated three stories, two created with ChatGPT-4o and one by Alberto Moravia, without being informed of their origin. To explore potential influencing factors, reading habits and demographic data, comprising age, gender, education and first language, were also collected. The results showed that the AI-written texts received slightly higher average ratings and were more frequently preferred, although differences were modest. No statistically significant associations were found between text preference and demographic or reading-habit variables. These findings challenge assumptions about reader preference for human-authored fiction and raise questions about the necessity of synthetic-text editing in literary contexts.

---