# Research Papers — 2026-01 · bundle page 142 of 189

> Full-text markdown of papers #7051–7100 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/142
[← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/141.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-01/page/143.md)

### Semantic NLP Pipelines for Interoperable Patient Digital Twins from Unstructured EHRs

Authors: Rafael Brens, Yuqiao Meng, Luoxi Tang, Zhaohan Xi
Published: 2026-01-09
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.05847v1

Digital twins -- virtual replicas of physical entities -- are gaining traction in healthcare for personalized monitoring, predictive modeling, and clinical decision support. However, generating interoperable patient digital twins from unstructured electronic health records (EHRs) remains challenging due to variability in clinical documentation and lack of standardized mappings. This paper presents a semantic NLP-driven pipeline that transforms free-text EHR notes into FHIR-compliant digital twin representations. The pipeline leverages named entity recognition (NER) to extract clinical concepts, concept normalization to map entities to SNOMED-CT or ICD-10, and relation extraction to capture structured associations between conditions, medications, and observations. Evaluation on MIMIC-IV Clinical Database Demo with validation against MIMIC-IV-on-FHIR reference mappings demonstrates high F1-scores for entity and relation extraction, with improved schema completeness and interoperability compared to baseline methods.

---

### A New Family of Poisson Non-negative Matrix Factorization Methods Using the Shifted Log Link

Authors: Eric Weine, Peter Carbonetto, Rafael A. Irizarry, Matthew Stephens
Published: 2026-01-09
Categories: cs.LG, stat.ME
arXiv: https://arxiv.org/abs/2601.05845v1

Poisson non-negative matrix factorization (NMF) is a widely used method to find interpretable "parts-based" decompositions of count data. While many variants of Poisson NMF exist, existing methods assume that the "parts" in the decomposition combine additively. This assumption may be natural in some settings, but not in others. Here we introduce Poisson NMF with the shifted-log link function to relax this assumption. The shifted-log link function has a single tuning parameter, and as this parameter varies the model changes from assuming that parts combine additively (i.e., standard Poisson NMF) to assuming that parts combine more multiplicatively. We provide an algorithm to fit this model by maximum likelihood, and also an approximation that substantially reduces computation time for large, sparse datasets (computations scale with the number of non-zero entries in the data matrix). We illustrate these new methods on a variety of real datasets. Our examples show how the choice of link function in Poisson NMF can substantively impact the results, and how in some settings the use of a shifted-log link function may improve interpretability compared with the standard, additive link.

---

### DexterCap: An Affordable and Automated System for Capturing Dexterous Hand-Object Manipulation

Authors: Yutong Liang, Shiyi Xu, Yulong Zhang, Bowen Zhan, He Zhang, Libin Liu
Published: 2026-01-09
Categories: cs.GR, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2601.05844v2

Capturing fine-grained hand-object interactions is challenging due to severe self-occlusion from closely spaced fingers and the subtlety of in-hand manipulation motions. Existing optical motion capture systems rely on expensive camera setups and extensive manual post-processing, while low-cost vision-based methods often suffer from reduced accuracy and reliability under occlusion. To address these challenges, we present DexterCap, a low-cost optical capture system for dexterous in-hand manipulation. DexterCap uses dense, character-coded marker patches to achieve robust tracking under severe self-occlusion, together with an automated reconstruction pipeline that requires minimal manual effort. With DexterCap, we introduce DexterHand, a dataset of fine-grained hand-object interactions covering diverse manipulation behaviors and objects, from simple primitives to complex articulated objects such as a Rubik's Cube. We release the dataset and code to support future research on dexterous hand-object interaction. Project website: https://pku-mocca.github.io/Dextercap-Page/

---

### DexterCap: An Affordable and Automated System for Capturing Dexterous Hand-Object Manipulation

Authors: Yutong Liang, Shiyi Xu, Yulong Zhang, Bowen Zhan, He Zhang, Libin Liu
Published: 2026-01-09
Categories: cs.GR, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2601.05844v1

Capturing fine-grained hand-object interactions is challenging due to severe self-occlusion from closely spaced fingers and the subtlety of in-hand manipulation motions. Existing optical motion capture systems rely on expensive camera setups and extensive manual post-processing, while low-cost vision-based methods often suffer from reduced accuracy and reliability under occlusion. To address these challenges, we present DexterCap, a low-cost optical capture system for dexterous in-hand manipulation. DexterCap uses dense, character-coded marker patches to achieve robust tracking under severe self-occlusion, together with an automated reconstruction pipeline that requires minimal manual effort. With DexterCap, we introduce DexterHand, a dataset of fine-grained hand-object interactions covering diverse manipulation behaviors and objects, from simple primitives to complex articulated objects such as a Rubik's Cube. We release the dataset and code to support future research on dexterous hand-object interaction.

---

### GeoSurDepth: Harnessing Foundation Model for Spatial Geometry Consistency-Oriented Self-Supervised Surround-View Depth Estimation

Authors: Weimin Liu, Wenjun Wang, Joshua H. Meng
Published: 2026-01-09
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.05839v2

Accurate surround-view depth estimation provides a competitive alternative to laser-based sensors and is essential for 3D scene understanding in autonomous driving. While empirical studies have proposed various approaches that primarily focus on enforcing cross-view constraints at photometric level, few explicitly exploit the rich geometric structure inherent in both monocular and surround-view setting. In this work, we propose GeoSurDepth, a framework that leverages geometry consistency as the primary cue for surround-view depth estimation. Concretely, we utilize vision foundation models as pseudo geometry priors and feature representation enhancement tool to guide the network to maintain surface normal consistency in spatial 3D space and regularize object- and texture-consistent depth estimation in 2D. In addition, we introduce a novel view synthesis pipeline where 2D-3D lifting is achieved with dense depth reconstructed via spatial warping, encouraging additional photometric supervision across temporal and spatial contexts, and compensating for the limitations of target-view image reconstruction. Finally, a newly-proposed adaptive joint motion learning strategy enables the network to adaptively emphasize informative spatial geometry cues for improved motion reasoning. Extensive experiments on KITTI, DDAD and nuScenes demonstrate that GeoSurDepth achieves SoTA performance, validating the effectiveness of our approach. Our framework highlights the importance of exploiting geometry coherence and consistency for robust self-supervised depth estimation.

---

### GeoSurDepth: Spatial Geometry-Consistent Self-Supervised Depth Estimation for Surround-View Cameras

Authors: Weimin Liu, Wenjun Wang, Joshua H. Meng
Published: 2026-01-09
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.05839v1

Accurate surround-view depth estimation provides a competitive alternative to laser-based sensors and is essential for 3D scene understanding in autonomous driving. While prior studies have proposed various approaches that primarily focus on enforcing cross-view constraints at the photometric level, few explicitly exploit the rich geometric structure inherent in both monocular and surround-view setting. In this work, we propose GeoSurDepth, a framework that leverages geometry consistency as the primary cue for surround-view depth estimation. Concretely, we utilize foundation models as a pseudo geometry prior and feature representation enhancement tool to guide the network to maintain surface normal consistency in spatial 3D space and regularize object- and texture-consistent depth estimation in 2D. In addition, we introduce a novel view synthesis pipeline where 2D-3D lifting is achieved with dense depth reconstructed via spatial warping, encouraging additional photometric supervision across temporal, spatial, and spatial-temporal contexts, and compensating for the limitations of single-view image reconstruction. Finally, a newly-proposed adaptive joint motion learning strategy enables the network to adaptively emphasize informative spatial geometry cues for improved motion reasoning. Extensive experiments on DDAD and nuScenes demonstrate that GeoSurDepth achieves state-of-the-art performance, validating the effectiveness of our approach. Our framework highlights the importance of exploiting geometry coherence and consistency for robust self-supervised multi-view depth estimation.

---

### Intelligent Singularity Avoidance in UR10 Robotic Arm Path Planning Using Hybrid Fuzzy Logic and Reinforcement Learning

Authors: Sheng-Kai Chen, Jyh-Horng Wu
Published: 2026-01-09
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2601.05836v1

This paper presents a comprehensive approach to singularity detection and avoidance in UR10 robotic arm path planning through the integration of fuzzy logic safety systems and reinforcement learning algorithms. The proposed system addresses critical challenges in robotic manipulation where singularities can cause loss of control and potential equipment damage. Our hybrid approach combines real-time singularity detection using manipulability measures, condition number analysis, and fuzzy logic decision-making with a stable reinforcement learning framework for adaptive path planning. Experimental results demonstrate a 90% success rate in reaching target positions while maintaining safe distances from singular configurations. The system integrates PyBullet simulation for training data collection and URSim connectivity for real-world deployment.

---

### Left, Right, or Center? Evaluating LLM Framing in News Classification and Generation

Authors: Molly Kennedy, Ali Parker, Yihong Liu, Hinrich Schütze
Published: 2026-01-09
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.05835v1

Large Language Model (LLM) based summarization and text generation are increasingly used for producing and rewriting text, raising concerns about political framing in journalism where subtle wording choices can shape interpretation. Across nine state-of-the-art LLMs, we study political framing by testing whether LLMs' classification-based bias signals align with framing behavior in their generated summaries. We first compare few-shot ideology predictions against LEFT/CENTER/RIGHT labels. We then generate "steered" summaries under FAITHFUL, CENTRIST, LEFT, and RIGHT prompts, and score all outputs using a single fixed ideology evaluator. We find pervasive ideological center-collapse in both article-level ratings and generated text, indicating a systematic tendency toward centrist framing. Among evaluated models, Grok 4 is by far the most ideologically expressive generator, while Claude Sonnet 4.5 and Llama 3.1 achieve the strongest bias-rating performance among commercial and open-weight models, respectively.

---

### Peek2: Regex-free Byte-level Byte-Pair Encoding Pretokenizer for LLM Inference on Edge Devices

Authors: Liu Zai, Iraklis Klampanos
Published: 2026-01-09
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.05833v2

Pretokenization is a crucial, sequential pass in Byte-level BPE tokenizers, yet little work has been done to optimize it for edge-side inference. Our proposed new implementation, Peek2, serves as a drop-in replacement for cl100k-like pretokenizers used in GPT-3, LLaMa-3, and Qwen-2.5. After breaking down and analyzing the logic of the original cl100k pretokenizer, we introduced a new pretokenization algorithm with linear time complexity and constant, trivial memory usage, suited for edge scenarios. Test results show that it increases microbenchmarking throughput by up to $ 2.48\times $ and delivers a $ 1.14\times $ improvement in overall throughput across the entire Byte-level BPE encoding process, depending on the dataset, while providing identical results as the baseline Regex-based tokenizer.

---

### Peek2: A Regex-free implementation of pretokenizers for Byte-level BPE

Authors: Liu Zai
Published: 2026-01-09
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.05833v1

Pretokenization is a crucial, sequential pass in Byte-level BPE tokenizers. Our proposed new implementation, Peek2, serves as a drop-in replacement for cl100k-like pretokenizers used in GPT-3, LLaMa-3, and Qwen-2.5. Designed with performance and safety in mind, Peek2 is Regex-free and delivers a $ 1.11\times $ improvement in overall throughput across the entire Byte-level BPE encoding process. This algorithm runs entirely on the CPU, has stable linear complexity $ O(n) $, and provides presegmentation results identical to those of the original Regex-based pretokenizer.

---

### Influence of Parallelism in Vector-Multiplication Units on Correlation Power Analysis

Authors: Manuel Brosch, Matthias Probst, Stefan Kögler, Georg Sigl
Published: 2026-01-09
Categories: cs.CR, cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2601.05828v1

The use of neural networks in edge devices is increasing, which introduces new security challenges related to the neural networks' confidentiality. As edge devices often offer physical access, attacks targeting the hardware, such as side-channel analysis, must be considered. To enhance the performance of neural network inference, hardware accelerators are commonly employed. This work investigates the influence of parallel processing within such accelerators on correlation-based side-channel attacks that exploit power consumption. The focus is on neurons that are part of the same fully-connected layer, which run parallel and simultaneously process the same input value. The theoretical impact of concurrent multiply-and-accumulate operations on overall power consumption is evaluated, as well as the success rate of correlation power analysis. Based on the observed behavior, equations are derived that describe how the correlation decreases with increasing levels of parallelism. The applicability of these equations is validated using a vector-multiplication unit implemented on an FPGA.

---

### Decoding Workload and Agreement From EEG During Spoken Dialogue With Conversational AI

Authors: Lucija Mihić Zidar, Philipp Wicke, Praneel Bhatia, Rosa Lutz, Marius Klug, Thorsten O. Zander
Published: 2026-01-09
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2601.05825v2

Passive brain-computer interfaces offer a potential source of implicit feedback for alignment of large language models, but most mental state decoding has been done in controlled tasks. This paper investigates whether established EEG classifiers for mental workload and implicit agreement can be transferred to spoken human-AI dialogue. We introduce two conversational paradigms - a Spelling Bee task and a sentence completion task- and an end-to-end pipeline for transcribing, annotating, and aligning word-level conversational events with continuous EEG classifier output. In a pilot study, workload decoding showed interpretable trends during spoken interaction, supporting cross-paradigm transfer. For implicit agreement, we demonstrate continuous application and precise temporal alignment to conversational events, while identifying limitations related to construct transfer and asynchronous application of event-based classifiers. Overall, the results establish feasibility and constraints for integrating passive BCI signals into conversational AI systems.

---

### Decoding Workload and Agreement From EEG During Spoken Dialogue With Conversational AI

Authors: Lucija Mihić Zidar, Philipp Wicke, Praneel Bhatia, Rosa Lutz, Marius Klug, Thorsten O. Zander
Published: 2026-01-09
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2601.05825v1

Passive brain-computer interfaces offer a potential source of implicit feedback for alignment of large language models, but most mental state decoding has been done in controlled tasks. This paper investigates whether established EEG classifiers for mental workload and implicit agreement can be transferred to spoken human-AI dialogue. We introduce two conversational paradigms - a Spelling Bee task and a sentence completion task- and an end-to-end pipeline for transcribing, annotating, and aligning word-level conversational events with continuous EEG classifier output. In a pilot study, workload decoding showed interpretable trends during spoken interaction, supporting cross-paradigm transfer. For implicit agreement, we demonstrate continuous application and precise temporal alignment to conversational events, while identifying limitations related to construct transfer and asynchronous application of event-based classifiers. Overall, the results establish feasibility and constraints for integrating passive BCI signals into conversational AI systems.

---

### Boosting Latent Diffusion Models via Disentangled Representation Alignment

Authors: John Page, Xuesong Niu, Kai Wu, Kun Gai
Published: 2026-01-09
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.05823v2

Latent Diffusion Models (LDMs) rely heavily on the compressed latent space provided by Variational Autoencoders (VAEs) for high-quality image generation. Recent studies have attempted to obtain generation-friendly VAEs by directly adopting alignment strategies from LDM training, leveraging Vision Foundation Models (VFMs) as representation alignment targets. However, such alignment paradigms overlook the fundamental differences in representational requirements between LDMs and VAEs. Simple feature mapping from local patches to high-dimensional semantics can induce semantic collapse, leading to the loss of fine-grained attributes. In this paper, we reveal a key insight: unlike LDMs that benefit from high-level global semantics, a generation-friendly VAE must possess strong semantic disentanglement capabilities to preserve fine-grained, attribute-level information in a structured manner. To address this discrepancy, we propose the Semantic-Disentangled VAE (Send-VAE). Deviating from previous shallow alignment approaches, Send-VAE introduces a non-linear mapping architecture to effectively bridge the local structures of VAEs and the dense semantics of VFMs, thereby encouraging emergent disentangled properties in the latent space without explicit regularization. Extensive experiments establish a new paradigm for evaluating VAE latent spaces via low-level attribute separability and demonstrate that Send-VAE achieves state-of-the-art generation quality (FID of 1.21) on ImageNet 256x256.

---

### Boosting Latent Diffusion Models via Disentangled Representation Alignment

Authors: John Page, Xuesong Niu, Kai Wu, Kun Gai
Published: 2026-01-09
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.05823v1

Latent Diffusion Models (LDMs) generate high-quality images by operating in a compressed latent space, typically obtained through image tokenizers such as Variational Autoencoders (VAEs). In pursuit of a generation-friendly VAE, recent studies have explored leveraging Vision Foundation Models (VFMs) as representation alignment targets for VAEs, mirroring the approach commonly adopted for LDMs. Although this yields certain performance gains, using the same alignment target for both VAEs and LDMs overlooks their fundamentally different representational requirements. We advocate that while LDMs benefit from latents retaining high-level semantic concepts, VAEs should excel in semantic disentanglement, enabling encoding of attribute-level information in a structured way. To address this, we propose the Semantic disentangled VAE (Send-VAE), explicitly optimized for disentangled representation learning through aligning its latent space with the semantic hierarchy of pre-trained VFMs. Our approach employs a non-linear mapper network to transform VAE latents, aligning them with VFMs to bridge the gap between attribute-level disentanglement and high-level semantics, facilitating effective guidance for VAE learning. We evaluate semantic disentanglement via linear probing on attribute prediction tasks, showing strong correlation with improved generation performance. Finally, using Send-VAE, we train flow-based transformers SiTs; experiments show Send-VAE significantly speeds up training and achieves a state-of-the-art FID of 1.21 and 1.75 with and without classifier-free guidance on ImageNet 256x256.

---

### LLMs as Science Journalists: Supporting Early-stage Researchers in Communicating Their Science to the Public

Authors: Milad Alshomary, Grace Li, Anubhav Jangra, Yufang Hou, Kathleen McKeown, Smaranda Muresan
Published: 2026-01-09
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.05821v1

The scientific community needs tools that help early-stage researchers effectively communicate their findings and innovations to the public. Although existing general-purpose Large Language Models (LLMs) can assist in this endeavor, they are not optimally aligned for it. To address this, we propose a framework for training LLMs to emulate the role of a science journalist that can be used by early-stage researchers to learn how to properly communicate their papers to the general public. We evaluate the usefulness of our trained LLM Journalists in leading conversations with both simulated and human researchers. %compared to the general-purpose ones. Our experiments indicate that LLMs trained using our framework ask more relevant questions that address the societal impact of research, prompting researchers to clarify and elaborate on their findings. In the user study, the majority of participants who interacted with our trained LLM Journalist appreciated it more than interacting with general-purpose LLMs.

---

### A Dual Pipeline Machine Learning Framework for Automated Multi Class Sleep Disorder Screening Using Hybrid Resampling and Ensemble Learning

Authors: Md Sultanul Islam Ovi, Muhsina Tarannum Munfa, G. M. M Miftahul Alam Adib, Syed Sabbir Hasan
Published: 2026-01-09
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.05814v2

Accurate classification of sleep disorders, particularly insomnia and sleep apnea, is important for reducing long term health risks and improving patient quality of life. However, clinical sleep studies are resource intensive and are difficult to scale for population level screening. This paper presents a Dual Pipeline Machine Learning Framework for multi class sleep disorder screening using the Sleep Health and Lifestyle dataset. The framework consists of two parallel processing streams: a statistical pipeline that targets linear separability using Mutual Information and Linear Discriminant Analysis, and a wrapper based pipeline that applies Boruta feature selection with an autoencoder for non linear representation learning. To address class imbalance, we use the hybrid SMOTETomek resampling strategy. In experiments, Extra Trees and K Nearest Neighbors achieved an accuracy of 98.67%, outperforming recent baselines on the same dataset. Statistical testing using the Wilcoxon Signed Rank Test indicates that the improvement over baseline configurations is significant, and inference latency remains below 400 milliseconds. These results suggest that the proposed dual pipeline design supports accurate and efficient automated screening for non invasive sleep disorder risk stratification.

---

### A Dual Pipeline Machine Learning Framework for Automated Multi Class Sleep Disorder Screening Using Hybrid Resampling and Ensemble Learning

Authors: Md Sultanul Islam Ovi, Muhsina Tarannum Munfa, Miftahul Alam Adib, Syed Sabbir Hasan
Published: 2026-01-09
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.05814v1

Accurate classification of sleep disorders, particularly insomnia and sleep apnea, is important for reducing long term health risks and improving patient quality of life. However, clinical sleep studies are resource intensive and are difficult to scale for population level screening. This paper presents a Dual Pipeline Machine Learning Framework for multi class sleep disorder screening using the Sleep Health and Lifestyle dataset. The framework consists of two parallel processing streams: a statistical pipeline that targets linear separability using Mutual Information and Linear Discriminant Analysis, and a wrapper based pipeline that applies Boruta feature selection with an autoencoder for non linear representation learning. To address class imbalance, we use the hybrid SMOTETomek resampling strategy. In experiments, Extra Trees and K Nearest Neighbors achieved an accuracy of 98.67%, outperforming recent baselines on the same dataset. Statistical testing using the Wilcoxon Signed Rank Test indicates that the improvement over baseline configurations is significant, and inference latency remains below 400 milliseconds. These results suggest that the proposed dual pipeline design supports accurate and efficient automated screening for non invasive sleep disorder risk stratification.

---

### Detecting Autism Spectrum Disorder with Deep Eye Movement Features

Authors: Zhanpei Huang, Taochen chen, Fangqing Gu, Yiqun Zhang
Published: 2026-01-09
Categories: cs.LG
arXiv: https://arxiv.org/abs/2601.05812v1

Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social communication and behavioral patterns. Eye movement data offers a non-invasive diagnostic tool for ASD detection, as it is inherently discrete and exhibits short-term temporal dependencies, reflecting localized gaze focus between fixation points. These characteristics enable the data to provide deeper insights into subtle behavioral markers, distinguishing ASD-related patterns from typical development. Eye movement signals mainly contain short-term and localized dependencies. However, despite the widespread application of stacked attention layers in Transformer-based models for capturing long-range dependencies, our experimental results indicate that this approach yields only limited benefits when applied to eye movement data. This may be because discrete fixation points and short-term dependencies in gaze focus reduce the utility of global attention mechanisms, making them less efficient than architectures focusing on local temporal patterns. To efficiently capture subtle and complex eye movement patterns, distinguishing ASD from typically developing (TD) individuals, a discrete short-term sequential (DSTS) modeling framework is designed with Class-aware Representation and Imbalance-aware Mechanisms. Through extensive experiments on several eye movement datasets, DSTS outperforms both traditional machine learning techniques and more sophisticated deep learning models.

---

### Learning Reconstructive Embeddings in Reproducing Kernel Hilbert Spaces via the Representer Theorem

Authors: Enrique Feito-Casares, Francisco M. Melgarejo-Meseguer, José-Luis Rojo-Álvarez
Published: 2026-01-09
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.05811v1

Motivated by the growing interest in representation learning approaches that uncover the latent structure of high-dimensional data, this work proposes new algorithms for reconstruction-based manifold learning within Reproducing-Kernel Hilbert Spaces (RKHS). Each observation is first reconstructed as a linear combination of the other samples in the RKHS, by optimizing a vector form of the Representer Theorem for their autorepresentation property. A separable operator-valued kernel extends the formulation to vector-valued data while retaining the simplicity of a single scalar similarity function. A subsequent kernel-alignment task projects the data into a lower-dimensional latent space whose Gram matrix aims to match the high-dimensional reconstruction kernel, thus transferring the auto-reconstruction geometry of the RKHS to the embedding. Therefore, the proposed algorithms represent an extended approach to the autorepresentation property, exhibited by many natural data, by using and adapting well-known results of Kernel Learning Theory. Numerical experiments on both simulated (concentric circles and swiss-roll) and real (cancer molecular activity and IoT network intrusions) datasets provide empirical evidence of the practical effectiveness of the proposed approach.

---

### SceneFoundry: Generating Interactive Infinite 3D Worlds

Authors: ChunTeng Chen, YiChen Hsu, YiWen Liu, WeiFang Sun, TsaiChing Ni, ChunYi Lee, Min Sun, YuanFu Yang
Published: 2026-01-09
Categories: cs.CV, cs.AI, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2601.05810v2

The ability to automatically generate large-scale, interactive, and physically realistic 3D environments is crucial for advancing robotic learning and embodied intelligence. However, existing generative approaches often fail to capture the functional complexity of real-world interiors, particularly those containing articulated objects with movable parts essential for manipulation and navigation. This paper presents SceneFoundry, a language-guided diffusion framework that generates apartment-scale 3D worlds with functionally articulated furniture and semantically diverse layouts for robotic training. From natural language prompts, an LLM module controls floor layout generation, while diffusion-based posterior sampling efficiently populates the scene with articulated assets from large-scale 3D repositories. To ensure physical usability, SceneFoundry employs differentiable guidance functions to regulate object quantity, prevent articulation collisions, and maintain sufficient walkable space for robotic navigation. Extensive experiments demonstrate that our framework generates structurally valid, semantically coherent, and functionally interactive environments across diverse scene types and conditions, enabling scalable embodied AI research. project page: https://anc891203.github.io/SceneFoundry-Demo/

---

### SceneFoundry: Generating Interactive Infinite 3D Worlds

Authors: ChunTeng Chen, YiChen Hsu, YiWen Liu, WeiFang Sun, TsaiChing Ni, ChunYi Lee, Min Sun, YuanFu Yang
Published: 2026-01-09
Categories: cs.CV, cs.AI, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2601.05810v1

The ability to automatically generate large-scale, interactive, and physically realistic 3D environments is crucial for advancing robotic learning and embodied intelligence. However, existing generative approaches often fail to capture the functional complexity of real-world interiors, particularly those containing articulated objects with movable parts essential for manipulation and navigation. This paper presents SceneFoundry, a language-guided diffusion framework that generates apartment-scale 3D worlds with functionally articulated furniture and semantically diverse layouts for robotic training. From natural language prompts, an LLM module controls floor layout generation, while diffusion-based posterior sampling efficiently populates the scene with articulated assets from large-scale 3D repositories. To ensure physical usability, SceneFoundry employs differentiable guidance functions to regulate object quantity, prevent articulation collisions, and maintain sufficient walkable space for robotic navigation. Extensive experiments demonstrate that our framework generates structurally valid, semantically coherent, and functionally interactive environments across diverse scene types and conditions, enabling scalable embodied AI research.

---

### EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis

Authors: Xiaoshuai Song, Haofei Chang, Guanting Dong, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou
Published: 2026-01-09
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.05808v2

Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but this process relies on rich and varied tool-interaction sandboxes. However, access to real systems is often restricted; LLM-simulated environments are prone to hallucinations and inconsistencies; and manually built sandboxes are hard to scale. In this paper, we propose EnvScaler, an automated framework for scalable tool-interaction environments via programmatic synthesis. EnvScaler comprises two components. First, SkelBuilder constructs diverse environment skeletons through topic mining, logic modeling, and quality evaluation. Then, ScenGenerator generates multiple task scenarios and rule-based trajectory validation functions for each environment. With EnvScaler, we synthesize 191 environments and about 7K scenarios, and apply them to Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) for Qwen3 series models. Results on three benchmarks show that EnvScaler significantly improves LLMs' ability to solve tasks in complex environments involving multi-turn, multi-tool interactions. We release our code and data at https://github.com/RUC-NLPIR/EnvScaler.

---

### EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis

Authors: Xiaoshuai Song, Haofei Chang, Guanting Dong, Yutao Zhu, Zhicheng Dou, Ji-Rong Wen
Published: 2026-01-09
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.05808v1

Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but this process relies on rich and varied tool-interaction sandboxes. However, access to real systems is often restricted; LLM-simulated environments are prone to hallucinations and inconsistencies; and manually built sandboxes are hard to scale. In this paper, we propose EnvScaler, an automated framework for scalable tool-interaction environments via programmatic synthesis. EnvScaler comprises two components. First, SkelBuilder constructs diverse environment skeletons through topic mining, logic modeling, and quality evaluation. Then, ScenGenerator generates multiple task scenarios and rule-based trajectory validation functions for each environment. With EnvScaler, we synthesize 191 environments and about 7K scenarios, and apply them to Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) for Qwen3 series models. Results on three benchmarks show that EnvScaler significantly improves LLMs' ability to solve tasks in complex environments involving multi-turn, multi-tool interactions. We release our code and data at https://github.com/RUC-NLPIR/EnvScaler.

---

### Fusion Matters: Length-Aware Analysis of Positional-Encoding Fusion in Transformers

Authors: Mohamed Amine Hallam, Kuo-Kun Tseng
Published: 2026-01-09
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2601.05807v1

Transformers require positional encodings to represent sequence order, yet most prior work focuses on designing new positional encodings rather than examining how positional information is fused with token embeddings. In this paper, we study whether the fusion mechanism itself affects performance, particularly in long-sequence settings. We conduct a controlled empirical study comparing three canonical fusion strategies--element-wise addition, concatenation with projection, and scalar gated fusion--under identical Transformer architectures, data splits, and random seeds. Experiments on three text classification datasets spanning short (AG News), medium (IMDB), and long (ArXiv) sequences show that fusion choice has negligible impact on short texts but produces consistent gains on long documents. To verify that these gains are structural rather than stochastic, we perform paired-seed analysis and cross-dataset comparison across sequence-length regimes. Additional experiments on the ArXiv dataset indicate that the benefit of learnable fusion generalizes across multiple positional encoding families. Finally, we explore a lightweight convolutional gating mechanism that introduces local inductive bias at the fusion level, evaluated on long documents only. Our results indicate that positional-encoding fusion is a non-trivial design choice for long-sequence Transformers and should be treated as an explicit modeling decision rather than a fixed default.

---

### Simplify-This: A Comparative Analysis of Prompt-Based and Fine-Tuned LLMs

Authors: Eilam Cohen, Itamar Bul, Danielle Inbar, Omri Loewenbach
Published: 2026-01-09
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2601.05794v1

Large language models (LLMs) enable strong text generation, and in general there is a practical tradeoff between fine-tuning and prompt engineering. We introduce Simplify-This, a comparative study evaluating both paradigms for text simplification with encoder-decoder LLMs across multiple benchmarks, using a range of evaluation metrics. Fine-tuned models consistently deliver stronger structural simplification, whereas prompting often attains higher semantic similarity scores yet tends to copy inputs. A human evaluation favors fine-tuned outputs overall. We release code, a cleaned derivative dataset used in our study, checkpoints of fine-tuned models, and prompt templates to facilitate reproducibility and future work.

---

### Tensor-DTI: Enhancing Biomolecular Interaction Prediction with Contrastive Embedding Learning

Authors: Manel Gil-Sorribes, Júlia Vilalta-Mor, Isaac Filella-Mercè, Robert Soliva, Álvaro Ciudad, Víctor Guallar, Alexis Molina
Published: 2026-01-09
Categories: cs.LG, cs.AI, q-bio.BM
arXiv: https://arxiv.org/abs/2601.05792v1

Accurate drug-target interaction (DTI) prediction is essential for computational drug discovery, yet existing models often rely on single-modality predefined molecular descriptors or sequence-based embeddings with limited representativeness. We propose Tensor-DTI, a contrastive learning framework that integrates multimodal embeddings from molecular graphs, protein language models, and binding-site predictions to improve interaction modeling. Tensor-DTI employs a siamese dual-encoder architecture, enabling it to capture both chemical and structural interaction features while distinguishing interacting from non-interacting pairs. Evaluations on multiple DTI benchmarks demonstrate that Tensor-DTI outperforms existing sequence-based and graph-based models. We also conduct large-scale inference experiments on CDK2 across billion-scale chemical libraries, where Tensor-DTI produces chemically plausible hit distributions even when CDK2 is withheld from training. In enrichment studies against Glide docking and Boltz-2 co-folder, Tensor-DTI remains competitive on CDK2 and improves the screening budget required to recover moderate fractions of high-affinity ligands on out-of-family targets under strict family-holdout splits. Additionally, we explore its applicability to protein-RNA and peptide-protein interactions. Our findings highlight the benefits of integrating multimodal information with contrastive objectives to enhance interaction-prediction accuracy and to provide more interpretable and reliability-aware models for virtual screening.

---

### SAFE: Secure and Accurate Federated Learning for Privacy-Preserving Brain-Computer Interfaces

Authors: Tianwang Jia, Xiaoqing Chen, Dongrui Wu
Published: 2026-01-09
Categories: cs.HC, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2601.05789v1

Electroencephalogram (EEG)-based brain-computer interfaces (BCIs) are widely adopted due to their efficiency and portability; however, their decoding algorithms still face multiple challenges, including inadequate generalization, adversarial vulnerability, and privacy leakage. This paper proposes Secure and Accurate FEderated learning (SAFE), a federated learning-based approach that protects user privacy by keeping data local during model training. SAFE employs local batch-specific normalization to mitigate cross-subject feature distribution shifts and hence improves model generalization. It further enhances adversarial robustness by introducing perturbations in both the input space and the parameter space through federated adversarial training and adversarial weight perturbation. Experiments on five EEG datasets from motor imagery (MI) and event-related potential (ERP) BCI paradigms demonstrated that SAFE consistently outperformed 14 state-of-the-art approaches in both decoding accuracy and adversarial robustness, while ensuring privacy protection. Notably, it even outperformed centralized training approaches that do not consider privacy protection at all. To our knowledge, SAFE is the first algorithm to simultaneously achieve high decoding accuracy, strong adversarial robustness, and reliable privacy protection without using any calibration data from the target subject, making it highly desirable for real-world BCIs.

---

### From Off-Policy to On-Policy: Enhancing GUI Agents via Bi-level Expert-to-Policy Assimilation

Authors: Zezhou Wang, Ziyun Zhang, Xiaoyi Zhang, Zhuzhong Qian, Yan Lu
Published: 2026-01-09
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.05787v2

Vision-language models are increasingly deployed as computer-use agents (CUAs) that operate desktops and browsers. Top-performing CUAs are framework-based systems that decompose planning and execution, while end-to-end screenshot-to-action policies are easier to deploy but lag behind on benchmarks such as OSWorld-Verified. GUI datasets like OSWorld pose two bottlenecks: they expose only a few hundred interactive, verifiable tasks and environments, and expert trajectories must be gathered by interacting with these environments, making such data hard to scale. We therefore ask how reinforcement learning from verifiable rewards (RLVR) can best exploit a small pool of exist expert trajectories to train end-to-end policies. Naively mixing these off-policy traces into on-policy RLVR is brittle: even after format conversion, expert trajectories exhibit structural mismatch and distribution shift from the learner. We propose BEPA (Bi-Level Expert-to-Policy Assimilation), which turns static expert traces into policy-aligned guidance via self-rolled reachable trajectories under the base policy (LEVEL-1) and a per-task, dynamically updated cache used in RLVR (LEVEL-2). On OSWorld-Verified, BEPA improves UITARS1.5-7B success from 22.87% to 32.13% and raises a held-out split from 5.74% to 10.30%, with consistent gains on MMBench-GUI and Online-Mind2Web. Our code and data are available at: https://github.com/LEON-gittech/Verl_GUI.git

---

### From Off-Policy to On-Policy: Enhancing GUI Agents via Bi-level Expert-to-Policy Assimilation

Authors: Zezhou Wang, Ziyun Zhang, Xiaoyi Zhang, Zhuzhong Qian, Yan Lu
Published: 2026-01-09
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.05787v1

Vision-language models are increasingly deployed as computer-use agents (CUAs) that operate desktops and browsers. Top-performing CUAs are framework-based systems that decompose planning and execution, while end-to-end screenshot-to-action policies are easier to deploy but lag behind on benchmarks such as OSWorld-Verified. GUI datasets like OSWorld pose two bottlenecks: they expose only a few hundred interactive, verifiable tasks and environments, and expert trajectories must be gathered by interacting with these environments, making such data hard to scale. We therefore ask how reinforcement learning from verifiable rewards (RLVR) can best exploit a small pool of exist expert trajectories to train end-to-end policies. Naively mixing these off-policy traces into on-policy RLVR is brittle: even after format conversion, expert trajectories exhibit structural mismatch and distribution shift from the learner. We propose BEPA (Bi-Level Expert-to-Policy Assimilation), which turns static expert traces into policy-aligned guidance via self-rolled reachable trajectories under the base policy (LEVEL-1) and a per-task, dynamically updated cache used in RLVR (LEVEL-2). On OSWorld-Verified, BEPA improves UITARS1.5-7B success from 22.87% to 32.13% and raises a held-out split from 5.74% to 10.30%, with consistent gains on MMBench-GUI and Online-Mind2Web. Our code and data are available at: https://github.com/LEON-gittech/Verl_GUI.git

---

### Adaptive Disentangled Representation Learning for Incomplete Multi-View Multi-Label Classification

Authors: Quanjiang Li, Zhiming Liu, Tianxiang Xu, Tingjin Luo, Chenping Hou
Published: 2026-01-09
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2601.05785v1

Multi-view multi-label learning frequently suffers from simultaneous feature absence and incomplete annotations, due to challenges in data acquisition and cost-intensive supervision. To tackle the complex yet highly practical problem while overcoming the existing limitations of feature recovery, representation disentanglement, and label semantics modeling, we propose an Adaptive Disentangled Representation Learning method (ADRL). ADRL achieves robust view completion by propagating feature-level affinity across modalities with neighborhood awareness, and reinforces reconstruction effectiveness by leveraging a stochastic masking strategy. Through disseminating category-level association across label distributions, ADRL refines distribution parameters for capturing interdependent label prototypes. Besides, we formulate a mutual-information-based objective to promote consistency among shared representations and suppress information overlap between view-specific representation and other modalities. Theoretically, we derive the tractable bounds to train the dual-channel network. Moreover, ADRL performs prototype-specific feature selection by enabling independent interactions between label embeddings and view representations, accompanied by the generation of pseudo-labels for each category. The structural characteristics of the pseudo-label space are then exploited to guide a discriminative trade-off during view fusion. Finally, extensive experiments on public datasets and real-world applications demonstrate the superior performance of ADRL.

---

### One Script Instead of Hundreds? On Pretraining Romanized Encoder Language Models

Authors: Benedikt Ebing, Lennart Keller, Goran Glavaš
Published: 2026-01-09
Categories: cs.CL
arXiv: https://arxiv.org/abs/2601.05776v1

Exposing latent lexical overlap, script romanization has emerged as an effective strategy for improving cross-lingual transfer (XLT) in multilingual language models (mLMs). Most prior work, however, focused on setups that favor romanization the most: (1) transfer from high-resource Latin-script to low-resource non-Latin-script languages and/or (2) between genealogically closely related languages with different scripts. It thus remains unclear whether romanization is a good representation choice for pretraining general-purpose mLMs, or, more precisely, if information loss associated with romanization harms performance for high-resource languages. We address this gap by pretraining encoder LMs from scratch on both romanized and original texts for six typologically diverse high-resource languages, investigating two potential sources of degradation: (i) loss of script-specific information and (ii) negative cross-lingual interference from increased vocabulary overlap. Using two romanizers with different fidelity profiles, we observe negligible performance loss for languages with segmental scripts, whereas languages with morphosyllabic scripts (Chinese and Japanese) suffer degradation that higher-fidelity romanization mitigates but cannot fully recover. Importantly, comparing monolingual LMs with their mLM counterpart, we find no evidence that increased subword overlap induces negative interference. We further show that romanization improves encoding efficiency (i.e., fertility) for segmental scripts at a negligible performance cost.

---

### Weights to Code: Extracting Interpretable Algorithms from the Discrete Transformer

Authors: Yifan Zhang, Wei Bi, Kechi Zhang, Dongming Jin, Jie Fu, Zhi Jin
Published: 2026-01-09
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2601.05770v3

Algorithm extraction aims to synthesize executable programs directly from models trained on algorithmic tasks, enabling de novo recovery of executable mechanisms from weights without relying on human-written target programs. However, applying this paradigm to Transformer is complicated by representation entanglement (e.g., superposition), where features encoded in overlapping directions substantially hinder the recovery of symbolic expressions. We propose the Discrete Transformer, an architecture explicitly designed to bridge the gap between continuous representations and discrete symbolic logic. By injecting discreteness through temperature-annealed sampling, our framework effectively leverages hypothesis testing and symbolic regression to extract human-readable programs. Empirically, the Discrete Transformer achieves performance comparable to the RNN-based MIPS baseline on shared discrete tasks, while broadening extraction to tasks with continuous-valued intermediate computations. Finally, we show that architectural inductive biases provide fine-grained control over synthesized programs, establishing the Discrete Transformer as a controllable testbed for algorithm extraction and Transformer interpretability.

---

### Weights to Code: Extracting Interpretable Algorithms from the Discrete Transformer

Authors: Yifan Zhang, Wei Bi, Kechi Zhang, Dongming Jin, Jie Fu, Zhi Jin
Published: 2026-01-09
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2601.05770v1

Algorithm extraction aims to synthesize executable programs directly from models trained on specific algorithmic tasks, enabling de novo algorithm discovery without relying on human-written code. However, extending this paradigm to Transformer is hindered by superposition, where entangled features encoded in overlapping directions obstruct the extraction of symbolic expressions. In this work, we propose the Discrete Transformer, an architecture explicitly engineered to bridge the gap between continuous representations and discrete symbolic logic. By enforcing a strict functional disentanglement, which constrains Numerical Attention to information routing and Numerical MLP to element-wise arithmetic, and employing temperature-annealed sampling, our method effectively facilitates the extraction of human-readable programs. Empirically, the Discrete Transformer not only achieves performance comparable to RNN-based baselines but crucially extends interpretability to continuous variable domains. Moreover, our analysis of the annealing process shows that the efficient discrete search undergoes a clear phase transition from exploration to exploitation. We further demonstrate that our method enables fine-grained control over synthesized programs by imposing inductive biases. Collectively, these findings establish the Discrete Transformer as a robust framework for demonstration-free algorithm discovery, offering a rigorous pathway toward Transformer interpretability.

---

### Variational Autoencoders for P-wave Detection on Strong Motion Earthquake Spectrograms

Authors: Turkan Simge Ispak, Salih Tileylioglu, Erdem Akagunduz
Published: 2026-01-09
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2601.05759v1

Accurate P-wave detection is critical for earthquake early warning, yet strong-motion records pose challenges due to high noise levels, limited labeled data, and complex waveform characteristics. This study reframes P-wave arrival detection as a self-supervised anomaly detection task to evaluate how architectural variations regulate the trade-off between reconstruction fidelity and anomaly discrimination. Through a comprehensive grid search of 492 Variational Autoencoder configurations, we show that while skip connections minimize reconstruction error (Mean Absolute Error approximately 0.0012), they induce "overgeneralization", allowing the model to reconstruct noise and masking the detection signal. In contrast, attention mechanisms prioritize global context over local detail and yield the highest detection performance with an area-under-the-curve of 0.875. The attention-based Variational Autoencoder achieves an area-under-the-curve of 0.91 in the 0 to 40-kilometer near-source range, demonstrating high suitability for immediate early warning applications. These findings establish that architectural constraints favoring global context over pixel-perfect reconstruction are essential for robust, self-supervised P-wave detection.

---

### VIGIL: Defending LLM Agents Against Tool Stream Injection via Verify-Before-Commit

Authors: Junda Lin, Zhaomeng Zhou, Zhi Zheng, Shuochen Liu, Tong Xu, Yong Chen, Enhong Chen
Published: 2026-01-09
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2601.05755v2

LLM agents operating in open environments face escalating risks from indirect prompt injection, particularly within the tool stream where manipulated metadata and runtime feedback hijack execution flow. Existing defenses encounter a critical dilemma as advanced models prioritize injected rules due to strict alignment while static protection mechanisms sever the feedback loop required for adaptive reasoning. To reconcile this conflict, we propose \textbf{VIGIL}, a framework that shifts the paradigm from restrictive isolation to a verify-before-commit protocol. By facilitating speculative hypothesis generation and enforcing safety through intent-grounded verification, \textbf{VIGIL} preserves reasoning flexibility while ensuring robust control. We further introduce \textbf{SIREN}, a benchmark comprising 959 tool stream injection cases designed to simulate pervasive threats characterized by dynamic dependencies. Extensive experiments demonstrate that \textbf{VIGIL} outperforms state-of-the-art dynamic defenses by reducing the attack success rate by over 22\% while more than doubling the utility under attack compared to static baselines, thereby achieving an optimal balance between security and utility.

---

### VIGIL: Defending LLM Agents Against Tool Stream Injection via Verify-Before-Commit

Authors: Junda Lin, Zhaomeng Zhou, Zhi Zheng, Shuochen Liu, Tong Xu, Yong Chen, Enhong Chen
Published: 2026-01-09
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2601.05755v1

LLM agents operating in open environments face escalating risks from indirect prompt injection, particularly within the tool stream where manipulated metadata and runtime feedback hijack execution flow. Existing defenses encounter a critical dilemma as advanced models prioritize injected rules due to strict alignment while static protection mechanisms sever the feedback loop required for adaptive reasoning. To reconcile this conflict, we propose \textbf{VIGIL}, a framework that shifts the paradigm from restrictive isolation to a verify-before-commit protocol. By facilitating speculative hypothesis generation and enforcing safety through intent-grounded verification, \textbf{VIGIL} preserves reasoning flexibility while ensuring robust control. We further introduce \textbf{SIREN}, a benchmark comprising 959 tool stream injection cases designed to simulate pervasive threats characterized by dynamic dependencies. Extensive experiments demonstrate that \textbf{VIGIL} outperforms state-of-the-art dynamic defenses by reducing the attack success rate by over 22\% while more than doubling the utility under attack compared to static baselines, thereby achieving an optimal balance between security and utility. Code is available at https://anonymous.4open.science/r/VIGIL-378B/.

---

### AutoMonitor-Bench: Evaluating the Reliability of LLM-Based Misbehavior Monitor

Authors: Shu Yang, Jingyu Hu, Tong Li, Hanqi Yan, Wenxuan Wang, Di Wang
Published: 2026-01-09
Categories: cs.CL, cs.SE
arXiv: https://arxiv.org/abs/2601.05752v3

We introduce AutoMonitor-Bench, the first benchmark designed to systematically evaluate the reliability of LLM-based misbehavior monitors across diverse tasks and failure modes. AutoMonitor-Bench consists of 3,010 carefully annotated test samples spanning question answering, code generation, and reasoning, with paired misbehavior and benign instances. We evaluate monitors using two complementary metrics: Miss Rate (MR) and False Alarm Rate (FAR), capturing failures to detect misbehavior and oversensitivity to benign behavior, respectively. Evaluating 12 proprietary and 10 open-source LLMs, we observe substantial variability in monitoring performance and a consistent trade-off between MR and FAR, revealing an inherent safety-utility tension. To further explore the limits of monitor reliability, we construct a large-scale training corpus of 153,581 samples and fine-tune Qwen3-4B-Instruction to investigate whether training on known, relatively easy-to-construct misbehavior datasets improves monitoring performance on unseen and more implicit misbehaviors. Our results highlight the challenges of reliable, scalable misbehavior monitoring and motivate future work on task-aware designing and training strategies for LLM-based monitors.

---

### AutoMonitor-Bench: Evaluating the Reliability of LLM-Based Misbehavior Monitor

Authors: Shu Yang, Jingyu Hu, Tong Li, Hanqi Yan, Wenxuan Wang, Di Wang
Published: 2026-01-09
Categories: cs.CL, cs.SE
arXiv: https://arxiv.org/abs/2601.05752v1

We introduce AutoMonitor-Bench, the first benchmark designed to systematically evaluate the reliability of LLM-based misbehavior monitors across diverse tasks and failure modes. AutoMonitor-Bench consists of 3,010 carefully annotated test samples spanning question answering, code generation, and reasoning, with paired misbehavior and benign instances. We evaluate monitors using two complementary metrics: Miss Rate (MR) and False Alarm Rate (FAR), capturing failures to detect misbehavior and oversensitivity to benign behavior, respectively. Evaluating 12 proprietary and 10 open-source LLMs, we observe substantial variability in monitoring performance and a consistent trade-off between MR and FAR, revealing an inherent safety-utility tension. To further explore the limits of monitor reliability, we construct a large-scale training corpus of 153,581 samples and fine-tune Qwen3-4B-Instruction to investigate whether training on known, relatively easy-to-construct misbehavior datasets improves monitoring performance on unseen and more implicit misbehaviors. Our results highlight the challenges of reliable, scalable misbehavior monitoring and motivate future work on task-aware designing and training strategies for LLM-based monitors.

---

### Analysing Differences in Persuasive Language in LLM-Generated Text: Uncovering Stereotypical Gender Patterns

Authors: Amalie Brogaard Pauli, Maria Barrett, Max Müller-Eberstein, Isabelle Augenstein, Ira Assent
Published: 2026-01-09
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.05751v2

Large language models (LLMs) are increasingly used for everyday communication tasks, including drafting interpersonal messages intended to influence and persuade. Prior work has shown that LLMs can successfully persuade humans and amplify persuasive language. It is therefore essential to understand how user instructions affect the generation of persuasive language, and to understand whether the generated persuasive language differs, for example, when targeting different groups. In this work, we propose a framework for evaluating how persuasive language generation is affected by recipient gender, sender intent, or output language. We evaluate 13 LLMs and 16 languages using pairwise prompt instructions. We evaluate model responses on 19 categories of persuasive language using an LLM-as-judge setup grounded in social psychology and communication science. Our results reveal significant gender differences in the persuasive language generated across all models. These patterns reflect biases consistent with gender-stereotypical linguistic tendencies documented in social psychology and sociolinguistics.

---

### Analysing Differences in Persuasive Language in LLM-Generated Text: Uncovering Stereotypical Gender Patterns

Authors: Amalie Brogaard Pauli, Maria Barrett, Max Müller-Eberstein, Isabelle Augenstein, Ira Assent
Published: 2026-01-09
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2601.05751v1

Large language models (LLMs) are increasingly used for everyday communication tasks, including drafting interpersonal messages intended to influence and persuade. Prior work has shown that LLMs can successfully persuade humans and amplify persuasive language. It is therefore essential to understand how user instructions affect the generation of persuasive language, and to understand whether the generated persuasive language differs, for example, when targeting different groups. In this work, we propose a framework for evaluating how persuasive language generation is affected by recipient gender, sender intent, or output language. We evaluate 13 LLMs and 16 languages using pairwise prompt instructions. We evaluate model responses on 19 categories of persuasive language using an LLM-as-judge setup grounded in social psychology and communication science. Our results reveal significant gender differences in the persuasive language generated across all models. These patterns reflect biases consistent with gender-stereotypical linguistic tendencies documented in social psychology and sociolinguistics.

---

### FlyPose: Towards Robust Human Pose Estimation From Aerial Views

Authors: Hassaan Farooq, Marvin Brenner, Peter Stütz
Published: 2026-01-09
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2601.05747v2

Unmanned Aerial Vehicles (UAVs) are increasingly deployed in close proximity to humans for applications such as parcel delivery, traffic monitoring, disaster response and infrastructure inspections. Ensuring safe and reliable operation in these human-populated environments demands accurate perception of human poses and actions from an aerial viewpoint. This perspective challenges existing methods with low resolution, steep viewing angles and (self-)occlusion, especially if the application demands realtime feasibile models. We train and deploy FlyPose, a lightweight top-down human pose estimation pipeline for aerial imagery. Through multi-dataset training, we achieve an average improvement of 6.8 mAP in person detection across the test-sets of Manipal-UAV, VisDrone, HIT-UAV as well as our custom dataset. For 2D human pose estimation we report an improvement of 16.3 mAP on the challenging UAV-Human dataset. FlyPose runs with an inference latency of ~20 milliseconds including preprocessing on a Jetson Orin AGX Developer Kit and is deployed onboard a quadrotor UAV during flight experiments. We also publish FlyPose-104, a small but challenging aerial human pose estimation dataset, that includes manual annotations from difficult aerial perspectives: https://github.com/farooqhassaan/FlyPose.

---

### FlyPose: Towards Robust Human Pose Estimation From Aerial Views

Authors: Hassaan Farooq, Marvin Brenner, Peter St\ütz
Published: 2026-01-09
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2601.05747v1

Unmanned Aerial Vehicles (UAVs) are increasingly deployed in close proximity to humans for applications such as parcel delivery, traffic monitoring, disaster response and infrastructure inspections. Ensuring safe and reliable operation in these human-populated environments demands accurate perception of human poses and actions from an aerial viewpoint. This perspective challenges existing methods with low resolution, steep viewing angles and (self-)occlusion, especially if the application demands realtime feasibile models. We train and deploy FlyPose, a lightweight top-down human pose estimation pipeline for aerial imagery. Through multi-dataset training, we achieve an average improvement of 6.8 mAP in person detection across the test-sets of Manipal-UAV, VisDrone, HIT-UAV as well as our custom dataset. For 2D human pose estimation we report an improvement of 16.3 mAP on the challenging UAV-Human dataset. FlyPose runs with an inference latency of ~20 milliseconds including preprocessing on a Jetson Orin AGX Developer Kit and is deployed onboard a quadrotor UAV during flight experiments. We also publish FlyPose-104, a small but challenging aerial human pose estimation dataset, that includes manual annotations from difficult aerial perspectives: https://github.com/farooqhassaan/FlyPose.

---

### DynaDebate: Breaking Homogeneity in Multi-Agent Debate with Dynamic Path Generation

Authors: Zhenghao Li, Zhi Zheng, Wei Chen, Jielun Zhao, Yong Chen, Tong Xu, Enhong Chen
Published: 2026-01-09
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.05746v2

Recent years have witnessed the rapid development of Large Language Model-based Multi-Agent Systems (MAS), which excel at collaborative decision-making and complex problem-solving. Researchers have further investigated Multi-Agent Debate (MAD) frameworks, which enhance the reasoning and collaboration capabilities of MAS through information exchange and debate among multiple agents. However, existing approaches often rely on unguided initialization, causing agents to adopt identical reasoning paths that lead to the same errors. As a result, effective debate among agents is hindered, and the final outcome frequently degenerates into simple majority voting. To solve the above problem, we introduce Dynamic Multi-Agent Debate (DynaDebate), which enhances the effectiveness of multi-agent debate through three key mechanisms: (1) Dynamic Path Generation and Allocation, which employs a dedicated Path Generation Agent to generate diverse and logical solution paths with adaptive redundancy; (2) Process-Centric Debate, which shifts the focus from surface-level outcome voting to rigorous step-by-step logic critique to ensure process correctness; (3) A Trigger-Based Verification Agent, which is activated upon disagreement and uses external tools to objectively resolve deadlocks. Experiments show that DynaDebate achieves superior or highly competitive performance across the majority of benchmarks\footnote{The code is at https://github.com/nwpuLee2021/brianstorm.}.

---

### DynaDebate: Breaking Homogeneity in Multi-Agent Debate with Dynamic Path Generation

Authors: Zhenghao Li, Zhi Zheng, Wei Chen, Jielun Zhao, Yong Chen, Tong Xu, Enhong Chen
Published: 2026-01-09
Categories: cs.AI
arXiv: https://arxiv.org/abs/2601.05746v1

Recent years have witnessed the rapid development of Large Language Model-based Multi-Agent Systems (MAS), which excel at collaborative decision-making and complex problem-solving. Recently, researchers have further investigated Multi-Agent Debate (MAD) frameworks, which enhance the reasoning and collaboration capabilities of MAS through information exchange and debate among multiple agents. However, existing approaches often rely on unguided initialization, causing agents to adopt identical reasoning paths that lead to the same errors. As a result, effective debate among agents is hindered, and the final outcome frequently degenerates into simple majority voting. To solve the above problem, in this paper, we introduce Dynamic Multi-Agent Debate (DynaDebate), which enhances the effectiveness of multi-agent debate through three key mechanisms: (1) Dynamic Path Generation and Allocation, which employs a dedicated Path Generation Agent to generate diverse and logical solution paths with adaptive redundancy; (2) Process-Centric Debate, which shifts the focus from surface-level outcome voting to rigorous step-by-step logic critique to ensure process correctness; (3) A Trigger-Based Verification Agent, which is activated upon disagreement and uses external tools to objectively resolve deadlocks. Extensive experiments demonstrate that DynaDebate achieves superior performance across various benchmarks, surpassing existing state-of-the-art MAD methods.

---

### The Echo Chamber Multi-Turn LLM Jailbreak

Authors: Ahmad Alobaid, Martí Jordà Roca, Carlos Castillo, Joan Vendrell
Published: 2026-01-09
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2601.05742v1

The availability of Large Language Models (LLMs) has led to a new generation of powerful chatbots that can be developed at relatively low cost. As companies deploy these tools, security challenges need to be addressed to prevent financial loss and reputational damage. A key security challenge is jailbreaking, the malicious manipulation of prompts and inputs to bypass a chatbot's safety guardrails. Multi-turn attacks are a relatively new form of jailbreaking involving a carefully crafted chain of interactions with a chatbot. We introduce Echo Chamber, a new multi-turn attack using a gradual escalation method. We describe this attack in detail, compare it to other multi-turn attacks, and demonstrate its performance against multiple state-of-the-art models through extensive evaluation.

---

### ViTNT-FIQA: Training-Free Face Image Quality Assessment with Vision Transformers

Authors: Guray Ozgur, Eduarda Caldeira, Tahar Chettaoui, Jan Niklas Kolf, Marco Huber, Naser Damer, Fadi Boutros
Published: 2026-01-09
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2601.05741v1

Face Image Quality Assessment (FIQA) is essential for reliable face recognition systems. Current approaches primarily exploit only final-layer representations, while training-free methods require multiple forward passes or backpropagation. We propose ViTNT-FIQA, a training-free approach that measures the stability of patch embedding evolution across intermediate Vision Transformer (ViT) blocks. We demonstrate that high-quality face images exhibit stable feature refinement trajectories across blocks, while degraded images show erratic transformations. Our method computes Euclidean distances between L2-normalized patch embeddings from consecutive transformer blocks and aggregates them into image-level quality scores. We empirically validate this correlation on a quality-labeled synthetic dataset with controlled degradation levels. Unlike existing training-free approaches, ViTNT-FIQA requires only a single forward pass without backpropagation or architectural modifications. Through extensive evaluation on eight benchmarks (LFW, AgeDB-30, CFP-FP, CALFW, Adience, CPLFW, XQLFW, IJB-C), we show that ViTNT-FIQA achieves competitive performance with state-of-the-art methods while maintaining computational efficiency and immediate applicability to any pre-trained ViT-based face recognition model.

---

### PII-VisBench: Evaluating Personally Identifiable Information Safety in Vision Language Models Along a Continuum of Visibility

Authors: G M Shahariar, Zabir Al Nazi, Md Olid Hasan Bhuiyan, Zhouxing Shi
Published: 2026-01-09
Categories: cs.AI, cs.CL, cs.CR, cs.CV
arXiv: https://arxiv.org/abs/2601.05739v1

Vision Language Models (VLMs) are increasingly integrated into privacy-critical domains, yet existing evaluations of personally identifiable information (PII) leakage largely treat privacy as a static extraction task and ignore how a subject's online presence--the volume of their data available online--influences privacy alignment. We introduce PII-VisBench, a novel benchmark containing 4000 unique probes designed to evaluate VLM safety through the continuum of online presence. The benchmark stratifies 200 subjects into four visibility categories: high, medium, low, and zero--based on the extent and nature of their information available online. We evaluate 18 open-source VLMs (0.3B-32B) based on two key metrics: percentage of PII probing queries refused (Refusal Rate) and the fraction of non-refusal responses flagged for containing PII (Conditional PII Disclosure Rate). Across models, we observe a consistent pattern: refusals increase and PII disclosures decrease (9.10% high to 5.34% low) as subject visibility drops. We identify that models are more likely to disclose PII for high-visibility subjects, alongside substantial model-family heterogeneity and PII-type disparities. Finally, paraphrasing and jailbreak-style prompts expose attack and model-dependent failures, motivating visibility-aware safety evaluation and training interventions.

---

### FeatureSLAM: Feature-enriched 3D gaussian splatting SLAM in real time

Authors: Christopher Thirgood, Oscar Mendez, Erin Ling, Jon Storey, Simon Hadfield
Published: 2026-01-09
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.05738v2

We present a real-time tracking SLAM system that unifies efficient camera tracking with photorealistic feature-enriched mapping using 3D Gaussian Splatting (3DGS). Our main contribution is integrating dense feature rasterization into the novel-view synthesis, aligned with a visual foundation model. This yields strong semantics, going beyond basic RGB-D input, aiding both tracking and mapping accuracy. Unlike previous semantic SLAM approaches (which embed pre-defined class labels) FeatureSLAM enables entirely new downstream tasks via free-viewpoint, open-set segmentation. Across standard benchmarks, our method achieves real-time tracking, on par with state-of-the-art systems while improving tracking stability and map fidelity without prohibitive compute. Quantitatively, we obtain 9\% lower pose error and 8\% higher mapping accuracy compared to recent fixed-set SLAM baselines. Our results confirm that real-time feature-embedded SLAM, is not only valuable for enabling new downstream applications. It also improves the performance of the underlying tracking and mapping subsystems, providing semantic and language masking results that are on-par with offline 3DGS models, alongside state-of-the-art tracking, depth and RGB rendering.

---

### FeatureSLAM: Feature-enriched 3D gaussian splatting SLAM in real time

Authors: Christopher Thirgood, Oscar Mendez, Erin Ling, Jon Storey, Simon Hadfield
Published: 2026-01-09
Categories: cs.CV
arXiv: https://arxiv.org/abs/2601.05738v1

We present a real-time tracking SLAM system that unifies efficient camera tracking with photorealistic feature-enriched mapping using 3D Gaussian Splatting (3DGS). Our main contribution is integrating dense feature rasterization into the novel-view synthesis, aligned with a visual foundation model. This yields strong semantics, going beyond basic RGB-D input, aiding both tracking and mapping accuracy. Unlike previous semantic SLAM approaches (which embed pre-defined class labels) FeatureSLAM enables entirely new downstream tasks via free-viewpoint, open-set segmentation. Across standard benchmarks, our method achieves real-time tracking, on par with state-of-the-art systems while improving tracking stability and map fidelity without prohibitive compute. Quantitatively, we obtain 9\% lower pose error and 8\% higher mapping accuracy compared to recent fixed-set SLAM baselines. Our results confirm that real-time feature-embedded SLAM, is not only valuable for enabling new downstream applications. It also improves the performance of the underlying tracking and mapping subsystems, providing semantic and language masking results that are on-par with offline 3DGS models, alongside state-of-the-art tracking, depth and RGB rendering.

---