# Research Papers — 2025-10 · bundle page 144 of 217

> Full-text markdown of papers #7151–7200 of 10843 published in 2025-10, newest first. AI-readable mirrors of the full 10843-paper 2025-10 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-10/page/144
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/143.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/145.md)

### Differential Analysis of Pseudo Haptic Feedback: Novel Comparative Study of Visual and Auditory Cue Integration for Psychophysical Evaluation

Authors: Nishant Gautam, Somya Sharma, Peter Corcoran, Kaspar Althoefer
Published: 2025-10-10
Categories: cs.HC, cs.GR, cs.NE, cs.RO, physics.med-ph
arXiv: https://arxiv.org/abs/2510.09570v1

Pseudo-haptics exploit carefully crafted visual or auditory cues to trick the brain into "feeling" forces that are never physically applied, offering a low-cost alternative to traditional haptic hardware. Here, we present a comparative psychophysical study that quantifies how visual and auditory stimuli combine to evoke pseudo-haptic pressure sensations on a commodity tablet. Using a Unity-based Rollball game, participants (n = 4) guided a virtual ball across three textured terrains while their finger forces were captured in real time with a Robotous RFT40 force-torque sensor. Each terrain was paired with a distinct rolling-sound profile spanning 440 Hz - 4.7 kHz, 440 Hz - 13.1 kHz, or 440 Hz - 8.9 kHz; crevice collisions triggered additional "knocking" bursts to heighten realism. Average tactile forces increased systematically with cue intensity: 0.40 N, 0.79 N and 0.88 N for visual-only trials and 0.41 N, 0.81 N and 0.90 N for audio-only trials on Terrains 1-3, respectively. Higher audio frequencies and denser visual textures both elicited stronger muscle activation, and their combination further reduced the force needed to perceive surface changes, confirming multisensory integration. These results demonstrate that consumer-grade isometric devices can reliably induce and measure graded pseudo-haptic feedback without specialized actuators, opening a path toward affordable rehabilitation tools, training simulators and assistive interfaces.

---

### Safe, Untrusted, "Proof-Carrying" AI Agents: toward the agentic lakehouse

Authors: Jacopo Tagliabue, Ciro Greco
Published: 2025-10-10
Categories: cs.AI, cs.DB
arXiv: https://arxiv.org/abs/2510.09567v1

Data lakehouses run sensitive workloads, where AI-driven automation raises concerns about trust, correctness, and governance. We argue that API-first, programmable lakehouses provide the right abstractions for safe-by-design, agentic workflows. Using Bauplan as a case study, we show how data branching and declarative environments extend naturally to agents, enabling reproducibility and observability while reducing the attack surface. We present a proof-of-concept in which agents repair data pipelines using correctness checks inspired by proof-carrying code. Our prototype demonstrates that untrusted AI agents can operate safely on production data and outlines a path toward a fully agentic lakehouse.

---

### Automated Evolutionary Optimization for Resource-Efficient Neural Network Training

Authors: Ilia Revin, Leon Strelkov, Vadim A. Potemkin, Ivan Kireev, Andrey Savchenko
Published: 2025-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.09566v1

There are many critical challenges in optimizing neural network models, including distributed computing, compression techniques, and efficient training, regardless of their application to specific tasks. Solving such problems is crucial because the need for scalable and resource-efficient models is increasing. To address these challenges, we have developed a new automated machine learning (AutoML) framework, Parameter Efficient Training with Robust Automation (PETRA). It applies evolutionary optimization to model architecture and training strategy. PETRA includes pruning, quantization, and loss regularization. Experimental studies on real-world data with financial event sequences, as well as image and time-series -- benchmarks, demonstrate PETRA's ability to improve neural model performance and scalability -- namely, a significant decrease in model size (up to 75%) and latency (up to 33%), and an increase in throughput (by 13%) without noticeable degradation in the target metric.

---

### TC-LoRA: Temporally Modulated Conditional LoRA for Adaptive Diffusion Control

Authors: Minkyoung Cho, Ruben Ohana, Christian Jacobsen, Adityan Jothi, Min-Hung Chen, Z. Morley Mao, Ethem Can
Published: 2025-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.09561v2

Current controllable diffusion models typically rely on fixed architectures that modify intermediate activations to inject guidance conditioned on a new modality. This approach uses a static conditioning strategy for a dynamic, multi-stage denoising process, limiting the model's ability to adapt its response as the generation evolves from coarse structure to fine detail. We introduce TC-LoRA (Temporally Modulated Conditional LoRA), a new paradigm that enables dynamic, context-aware control by conditioning the model's weights directly. Our framework uses a hypernetwork to generate LoRA adapters on-the-fly, tailoring weight modifications for the frozen backbone at each diffusion step based on time and the user's condition. This mechanism enables the model to learn and execute an explicit, adaptive strategy for applying conditional guidance throughout the entire generation process. Through experiments on various data domains, we demonstrate that this dynamic, parametric control significantly enhances generative fidelity and adherence to spatial conditions compared to static, activation-based methods. TC-LoRA establishes an alternative approach in which the model's conditioning strategy is modified through a deeper functional adaptation of its weights, allowing control to align with the dynamic demands of the task and generative stage.

---

### AutoPR: Let's Automate Your Academic Promotion!

Authors: Qiguang Chen, Zheng Yan, Mingda Yang, Libo Qin, Yixin Yuan, Hanjing Li, Jinhao Liu, Yiyan Ji, Dengyun Peng, Jiannan Guan, Mengkang Hu, Yantao Du et al.
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09558v2

As the volume of peer-reviewed research surges, scholars increasingly rely on social platforms for discovery, while authors invest considerable effort in promoting their work to ensure visibility and citations. To streamline this process and reduce the reliance on human effort, we introduce Automatic Promotion (AutoPR), a novel task that transforms research papers into accurate, engaging, and timely public content. To enable rigorous evaluation, we release PRBench, a multimodal benchmark that links 512 peer-reviewed articles to high-quality promotional posts, assessing systems along three axes: Fidelity (accuracy and tone), Engagement (audience targeting and appeal), and Alignment (timing and channel optimization). We also introduce PRAgent, a multi-agent framework that automates AutoPR in three stages: content extraction with multimodal preparation, collaborative synthesis for polished outputs, and platform-specific adaptation to optimize norms, tone, and tagging for maximum reach. When compared to direct LLM pipelines on PRBench, PRAgent demonstrates substantial improvements, including a 604% increase in total watch time, a 438% rise in likes, and at least a 2.9x boost in overall engagement. Ablation studies show that platform modeling and targeted promotion contribute the most to these gains. Our results position AutoPR as a tractable, measurable research problem and provide a roadmap for scalable, impactful automated scholarly communication.

---

### WUGNECTIVES: Novel Entity Inferences of Language Models from Discourse Connectives

Authors: Daniel Brubaker, William Sheffield, Junyi Jessy Li, Kanishka Misra
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09556v3

The role of world knowledge has been particularly crucial to predict the discourse connective that marks the discourse relation between two arguments, with language models (LMs) being generally successful at this task. We flip this premise in our work, and instead study the inverse problem of understanding whether discourse connectives can inform LMs about the world. To this end, we present WUGNECTIVES, a dataset of 8,880 stimuli that evaluates LMs' inferences about novel entities in contexts where connectives link the entities to particular attributes. On investigating 17 different LMs at various scales, and training regimens, we found that tuning an LM to show reasoning behavior yields noteworthy improvements on most connectives. At the same time, there was a large variation in LMs' overall performance across connective type, with all models systematically struggling on connectives that express a concessive meaning. Our findings pave the way for more nuanced investigations into the functional role of language cues as captured by LMs. We release WUGNECTIVES at https://github.com/kanishkamisra/wugnectives

---

### A Comprehensive Evaluation of Multilingual Chain-of-Thought Reasoning: Performance, Consistency, and Faithfulness Across Languages

Authors: Raoyuan Zhao, Yihong Liu, Hinrich Schütze, Michael A. Hedderich
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09555v1

Large reasoning models (LRMs) increasingly rely on step-by-step Chain-of-Thought (CoT) reasoning to improve task performance, particularly in high-resource languages such as English. While recent work has examined final-answer accuracy in multilingual settings, the thinking traces themselves, i.e., the intermediate steps that lead to the final answer, remain underexplored. In this paper, we present the first comprehensive study of multilingual CoT reasoning, evaluating three key dimensions: performance, consistency, and faithfulness. We begin by measuring language compliance, answer accuracy, and answer consistency when LRMs are explicitly instructed or prompt-hacked to think in a target language, revealing strong language preferences and divergent performance across languages. Next, we assess crosslingual consistency of thinking traces by interchanging them between languages. We find that the quality and effectiveness of thinking traces vary substantially depending on the prompt language. Finally, we adapt perturbation-based techniques -- i.e., truncation and error injection -- to probe the faithfulness of thinking traces across languages, showing that models rely on traces to varying degrees. We release our code and data to support future research.

---

### Hierarchical Indexing with Knowledge Enrichment for Multilingual Video Corpus Retrieval

Authors: Yu Wang, Tianhao Tan, Yifei Wang
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09553v1

Retrieving relevant instructional videos from multilingual medical archives is crucial for answering complex, multi-hop questions across language boundaries. However, existing systems either compress hour-long videos into coarse embeddings or incur prohibitive costs for fine-grained matching. We tackle the Multilingual Video Corpus Retrieval (mVCR) task in the NLPCC-2025 M4IVQA challenge with a multi-stage framework that integrates multilingual semantics, domain terminology, and efficient long-form processing. Video subtitles are divided into semantically coherent chunks, enriched with concise knowledge-graph (KG) facts, and organized into a hierarchical tree whose node embeddings are generated by a language-agnostic multilingual encoder. At query time, the same encoder embeds the input question; a coarse-to-fine tree search prunes irrelevant branches, and only the top-ranked chunks are re-scored by a lightweight large language model (LLM). This design avoids exhaustive cross-encoder scoring while preserving chunk-level precision. Experiments on the mVCR test set demonstrate state-of-the-art performance, and ablation studies confirm the complementary contributions of KG enrichment, hierarchical indexing, and targeted LLM re-ranking. The proposed method offers an accurate and scalable solution for multilingual retrieval in specialized medical video collections.

---

### Titans Revisited: A Lightweight Reimplementation and Critical Analysis of a Test-Time Memory Model

Authors: Gavriel Di Nepi, Federico Siciliano, Fabrizio Silvestri
Published: 2025-10-10
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.09551v1

By the end of 2024, Google researchers introduced Titans: Learning at Test Time, a neural memory model achieving strong empirical results across multiple tasks. However, the lack of publicly available code and ambiguities in the original description hinder reproducibility. In this work, we present a lightweight reimplementation of Titans and conduct a comprehensive evaluation on Masked Language Modeling, Time Series Forecasting, and Recommendation tasks. Our results reveal that Titans does not always outperform established baselines due to chunking. However, its Neural Memory component consistently improves performance compared to attention-only models. These findings confirm the model's innovative potential while highlighting its practical limitations and raising questions for future research.

---

### Beyond Surface Reasoning: Unveiling the True Long Chain-of-Thought Capacity of Diffusion Large Language Models

Authors: Qiguang Chen, Hanjing Li, Libo Qin, Dengyun Peng, Jinhao Liu, Jiangyi Wang, Chengyue Wu, Xie Chen, Yantao Du, Wanxiang Che
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09544v1

Recently, Diffusion Large Language Models (DLLMs) have offered high throughput and effective sequential reasoning, making them a competitive alternative to autoregressive LLMs (ALLMs). However, parallel decoding, which enables simultaneous token updates, conflicts with the causal order often required for rigorous reasoning. We first identify this conflict as the core Parallel-Sequential Contradiction (PSC). Behavioral analyses in both simple and complex reasoning tasks show that DLLMs exhibit genuine parallelism only for directly decidable outputs. As task difficulty increases, they revert to autoregressive-like behavior, a limitation exacerbated by autoregressive prompting, which nearly doubles the number of decoding steps with remasking without improving quality. Moreover, PSC restricts DLLMs' self-reflection, reasoning depth, and exploratory breadth. To further characterize PSC, we introduce three scaling dimensions for DLLMs: parallel, diffusion, and sequential. Empirically, while parallel scaling yields consistent improvements, diffusion and sequential scaling are constrained by PSC. Based on these findings, we propose several practical mitigations, parallel-oriented prompting, diffusion early stopping, and parallel scaling, to reduce PSC-induced ineffectiveness and inefficiencies.

---

### SPG: Sandwiched Policy Gradient for Masked Diffusion Language Models

Authors: Chenyu Wang, Paria Rashidinejad, DiJia Su, Song Jiang, Sid Wang, Siyan Zhao, Cai Zhou, Shannon Zejiang Shen, Feiyu Chen, Tommi Jaakkola, Yuandong Tian, Bo Liu
Published: 2025-10-10
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.09541v3

Diffusion large language models (dLLMs) are emerging as an efficient alternative to autoregressive models due to their ability to decode multiple tokens in parallel. However, aligning dLLMs with human preferences or task-specific rewards via reinforcement learning (RL) is challenging because their intractable log-likelihood precludes the direct application of standard policy gradient methods. While prior work uses surrogates like the evidence lower bound (ELBO), these one-sided approximations can introduce significant policy gradient bias. To address this, we propose the Sandwiched Policy Gradient (SPG) that leverages both an upper and a lower bound of the true log-likelihood. Experiments show that SPG significantly outperforms baselines based on ELBO or one-step estimation. Specifically, SPG improves the accuracy over state-of-the-art RL methods for dLLMs by 3.6% in GSM8K, 2.6% in MATH500, 18.4% in Countdown and 27.0% in Sudoku.

---

### FLOWING: Implicit Neural Flows for Structure-Preserving Morphing

Authors: Arthur Bizzi, Matias Grynberg, Vitor Matias, Daniel Perazzo, João Paulo Lima, Luiz Velho, Nuno Gonçalves, João Pereira, Guilherme Schardong, Tiago Novello
Published: 2025-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.09537v1

Morphing is a long-standing problem in vision and computer graphics, requiring a time-dependent warping for feature alignment and a blending for smooth interpolation. Recently, multilayer perceptrons (MLPs) have been explored as implicit neural representations (INRs) for modeling such deformations, due to their meshlessness and differentiability; however, extracting coherent and accurate morphings from standard MLPs typically relies on costly regularizations, which often lead to unstable training and prevent effective feature alignment. To overcome these limitations, we propose FLOWING (FLOW morphING), a framework that recasts warping as the construction of a differential vector flow, naturally ensuring continuity, invertibility, and temporal coherence by encoding structural flow properties directly into the network architectures. This flow-centric approach yields principled and stable transformations, enabling accurate and structure-preserving morphing of both 2D images and 3D shapes. Extensive experiments across a range of applications - including face and image morphing, as well as Gaussian Splatting morphing - show that FLOWING achieves state-of-the-art morphing quality with faster convergence. Code and pretrained models are available at http://schardong.github.io/flowing.

---

### Evaluating Robustness of Large Language Models Against Multilingual Typographical Errors

Authors: Raoyuan Zhao, Yihong Liu, Lena Altinger, Hinrich Schütze, Michael A. Hedderich
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09536v3

Large language models (LLMs) are increasingly deployed in multilingual, real-world applications with user inputs -- naturally introducing \emph{typographical errors} (typos). Yet most benchmarks assume clean input, leaving the robustness of LLMs to typos across languages largely underexplored. To address this gap, we introduce MulTypo, a multilingual typo generation algorithm that simulates human-like errors based on language-specific keyboard layouts and typing behavior. We evaluate 18 open-source LLMs across three model families and five downstream tasks spanning language inference, multi-choice question answering, mathematical reasoning, and machine translation tasks. Our results show that typos consistently degrade performance, particularly in generative tasks and those requiring reasoning -- while the natural language inference task is comparatively more robust. Instruction tuning improves clean-input performance but may increase brittleness under noise. We also observe language-dependent robustness: high-resource languages are generally more robust than low-resource ones, and translation from English is more robust than translation into English. Our findings underscore the need for noise-aware training and multilingual robustness evaluation. We release a Python package for MulTypo and make the source code publicly available at https://github.com/cisnlp/multypo.

---

### Mitigating Overthinking through Reasoning Shaping

Authors: Feifan Song, Shaohang Wei, Bofei Gao, Yejie Wang, Wen Luo, Wei Li, Linli Yao, Weimin Xiong, Liang Chen, Tianyu Liu, Houfeng Wang
Published: 2025-10-10
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.09535v1

Large reasoning models (LRMs) boosted by Reinforcement Learning from Verifier Reward (RLVR) have shown great power in problem solving, yet they often cause overthinking: excessive, meandering reasoning that inflates computational cost. Prior designs of penalization in RLVR manage to reduce token consumption while often harming model performance, which arises from the oversimplicity of token-level supervision. In this paper, we argue that the granularity of supervision plays a crucial role in balancing efficiency and accuracy, and propose Group Relative Segment Penalization (GRSP), a step-level method to regularize reasoning. Since preliminary analyses show that reasoning segments are strongly correlated with token consumption and model performance, we design a length-aware weighting mechanism across segment clusters. Extensive experiments demonstrate that GRSP achieves superior token efficiency without heavily compromising accuracy, especially the advantages with harder problems. Moreover, GRSP stabilizes RL training and scales effectively across model sizes.

---

### Conditional Flow Matching for Bayesian Posterior Inference

Authors: Percy S. Zhai, So Won Jeong, Veronika Ročková
Published: 2025-10-10
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.09534v3

We propose a generative multivariate posterior sampler via flow matching. It offers a simple training objective, and does not require access to likelihood evaluation. The method learns a dynamic, block-triangular velocity field in the joint space of data and parameters, which results in a deterministic transport map from a source distribution to the desired posterior. The inverse map, named vector rank, is accessible by reversibly integrating the velocity over time. It is advantageous to leverage the dynamic design: proper constraints on the velocity yield a monotone map, which leads to a conditional Brenier map, enabling a fast and simultaneous generation of Bayesian credible sets whose contours correspond to level sets of Monge-Kantorovich data depth. Our approach is computationally lighter compared to GAN-based and diffusion-based counterparts, and is capable of capturing complex posterior structures. Finally, frequentist theoretical guarantee on the consistency of the recovered posterior distribution, and of the corresponding Bayesian credible sets, is provided.

---

### PRNet: Original Information Is All You Have

Authors: PeiHuang Zheng, Yunlong Zhao, Zheng Cui, Yang Li
Published: 2025-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.09531v1

Small object detection in aerial images suffers from severe information degradation during feature extraction due to limited pixel representations, where shallow spatial details fail to align effectively with semantic information, leading to frequent misses and false positives. Existing FPN-based methods attempt to mitigate these losses through post-processing enhancements, but the reconstructed details often deviate from the original image information, impeding their fusion with semantic content. To address this limitation, we propose PRNet, a real-time detection framework that prioritizes the preservation and efficient utilization of primitive shallow spatial features to enhance small object representations. PRNet achieves this via two modules:the Progressive Refinement Neck (PRN) for spatial-semantic alignment through backbone reuse and iterative refinement, and the Enhanced SliceSamp (ESSamp) for preserving shallow information during downsampling via optimized rearrangement and convolution. Extensive experiments on the VisDrone, AI-TOD, and UAVDT datasets demonstrate that PRNet outperforms state-of-the-art methods under comparable computational constraints, achieving superior accuracy-efficiency trade-offs.

---

### Accent-Invariant Automatic Speech Recognition via Saliency-Driven Spectrogram Masking

Authors: Mohammad Hossein Sameti, Sepehr Harfi Moridani, Ali Zarean, Hossein Sameti
Published: 2025-10-10
Categories: cs.CL, cs.SD, eess.AS
arXiv: https://arxiv.org/abs/2510.09528v1

Pre-trained transformer-based models have significantly advanced automatic speech recognition (ASR), yet they remain sensitive to accent and dialectal variations, resulting in elevated word error rates (WER) in linguistically diverse languages such as English and Persian. To address this challenge, we propose an accent-invariant ASR framework that integrates accent and dialect classification into the recognition pipeline. Our approach involves training a spectrogram-based classifier to capture accent-specific cues, masking the regions most influential to its predictions, and using the masked spectrograms for data augmentation. This enhances the robustness of ASR models against accent variability. We evaluate the method using both English and Persian speech. For Persian, we introduce a newly collected dataset spanning multiple regional accents, establishing the first systematic benchmark for accent variation in Persian ASR that fills a critical gap in multilingual speech research and provides a foundation for future studies on low-resource, linguistically diverse languages. Experimental results with the Whisper model demonstrate that our masking and augmentation strategy yields substantial WER reductions in both English and Persian settings, confirming the effectiveness of the approach. This research advances the development of multilingual ASR systems that are resilient to accent and dialect diversity. Code and dataset are publicly available at: https://github.com/MH-Sameti/Accent_invariant_ASR

---

### Can We Reliably Rank Model Performance across Domains without Labeled Data?

Authors: Veronica Rammouz, Aaron Gonzalez, Carlos Cruzportillo, Adrian Tan, Nicole Beebe, Anthony Rios
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09519v1

Estimating model performance without labels is an important goal for understanding how NLP models generalize. While prior work has proposed measures based on dataset similarity or predicted correctness, it remains unclear when these estimates produce reliable performance rankings across domains. In this paper, we analyze the factors that affect ranking reliability using a two-step evaluation setup with four base classifiers and several large language models as error predictors. Experiments on the GeoOLID and Amazon Reviews datasets, spanning 15 domains, show that large language model-based error predictors produce stronger and more consistent rank correlations with true accuracy than drift-based or zero-shot baselines. Our analysis reveals two key findings: ranking is more reliable when performance differences across domains are larger, and when the error model's predictions align with the base model's true failure patterns. These results clarify when performance estimation methods can be trusted and provide guidance for their use in cross-domain model evaluation.

---

### StatEval: A Comprehensive Benchmark for Large Language Models in Statistics

Authors: Yuchen Lu, Run Yang, Yichen Zhang, Shuguang Yu, Ziwei Wang, Jiayi Xiang, Wenxin E, Changyu Zhu, Fan Zhou
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09517v2

Despite rapid advances in large language models (LLMs), statistical reasoning remains underrepresented in existing LLM benchmarks, which often do not reflect the layered, proof-driven nature of real statistical practice. To address this gap, we introduce \textbf{StatEval}, the first large-scale benchmark for statistical reasoning across curricular and research-level settings. StatEval includes over 100,000 curated problems, with 20,000+ foundational questions spanning undergraduate and graduate curricula and 80,000+ research-level proof tasks extracted from leading statistical journals. To construct StatEval, we develop \textbf{TRACE} (Topology and Reasoning-Aware Context Extractor), a multi-agent pipeline with human-in-the-loop validation that converts unstructured academic texts into self-contained theorem-level reasoning tasks. We also propose an Adaptive Process-Based Scoring Pipeline for complex statistical proofs, enabling fine-grained evaluation beyond final-answer matching. Experiments show that while LLMs perform reasonably on foundational tasks, they struggle with rigorous research-level reasoning. Beyond evaluation, StatEval serves as a resource for improving reasoning, as retrieval-augmented generation and domain-specific alignment consistently enhance performance. Together, these results establish StatEval as both a benchmark and an infrastructure for advancing statistical reasoning in LLMs.

---

### Interpretable Generative and Discriminative Learning for Multimodal and Incomplete Clinical Data

Authors: Albert Belenguer-Llorens, Carlos Sevilla-Salcedo, Janaina Mourao-Miranda, Vanessa Gómez-Verdejo
Published: 2025-10-10
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.09513v1

Real-world clinical problems are often characterized by multimodal data, usually associated with incomplete views and limited sample sizes in their cohorts, posing significant limitations for machine learning algorithms. In this work, we propose a Bayesian approach designed to efficiently handle these challenges while providing interpretable solutions. Our approach integrates (1) a generative formulation to capture cross-view relationships with a semi-supervised strategy, and (2) a discriminative task-oriented formulation to identify relevant information for specific downstream objectives. This dual generative-discriminative formulation offers both general understanding and task-specific insights; thus, it provides an automatic imputation of the missing views while enabling robust inference across different data sources. The potential of this approach becomes evident when applied to the multimodal clinical data, where our algorithm is able to capture and disentangle the complex interactions among biological, psychological, and sociodemographic modalities.

---

### Diagonal Artifacts in Samsung Images: PRNU Challenges and Solutions

Authors: David Vázquez-Padín, Fernando Pérez-González, Alejandro Martín-Del-Río
Published: 2025-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.09509v1

We investigate diagonal artifacts present in images captured by several Samsung smartphones and their impact on PRNU-based camera source verification. We first show that certain Galaxy S series models share a common pattern causing fingerprint collisions, with a similar issue also found in some Galaxy A models. Next, we demonstrate that reliable PRNU verification remains feasible for devices supporting PRO mode with raw capture, since raw images bypass the processing pipeline that introduces artifacts. This option, however, is not available for the mid-range A series models or in forensic cases without access to raw images. Finally, we outline potential forensic applications of the diagonal artifacts, such as reducing misdetections in HDR images and localizing regions affected by synthetic bokeh in portrait-mode images.

---

### PhysToolBench: Benchmarking Physical Tool Understanding for MLLMs

Authors: Zixin Zhang, Kanghao Chen, Xingwang Lin, Lutao Jiang, Xu Zheng, Yuanhuiyi Lyu, Litao Guo, Yinchuan Li, Ying-Cong Chen
Published: 2025-10-10
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2510.09507v1

The ability to use, understand, and create tools is a hallmark of human intelligence, enabling sophisticated interaction with the physical world. For any general-purpose intelligent agent to achieve true versatility, it must also master these fundamental skills. While modern Multimodal Large Language Models (MLLMs) leverage their extensive common knowledge for high-level planning in embodied AI and in downstream Vision-Language-Action (VLA) models, the extent of their true understanding of physical tools remains unquantified. To bridge this gap, we present PhysToolBench, the first benchmark dedicated to evaluating the comprehension of physical tools by MLLMs. Our benchmark is structured as a Visual Question Answering (VQA) dataset comprising over 1,000 image-text pairs. It assesses capabilities across three distinct difficulty levels: (1) Tool Recognition: Requiring the recognition of a tool's primary function. (2) Tool Understanding: Testing the ability to grasp the underlying principles of a tool's operation. (3) Tool Creation: Challenging the model to fashion a new tool from surrounding objects when conventional options are unavailable. Our comprehensive evaluation of 32 MLLMs-spanning proprietary, open-source, specialized embodied, and backbones in VLAs-reveals a significant deficiency in tool understanding. Furthermore, we provide an in-depth analysis and propose preliminary solutions. Code and dataset are publicly available.

---

### Geo-Aware Models for Stream Temperature Prediction across Different Spatial Regions and Scales

Authors: Shiyuan Luo, Runlong Yu, Shengyu Chen, Yingda Fan, Yiqun Xie, Yanhua Li, Xiaowei Jia
Published: 2025-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.09500v1

Understanding environmental ecosystems is vital for the sustainable management of our planet. However,existing physics-based and data-driven models often fail to generalize to varying spatial regions and scales due to the inherent data heterogeneity presented in real environmental ecosystems. This generalization issue is further exacerbated by the limited observation samples available for model training. To address these issues, we propose Geo-STARS, a geo-aware spatio-temporal modeling framework for predicting stream water temperature across different watersheds and spatial scales. The major innovation of Geo-STARS is the introduction of geo-aware embedding, which leverages geographic information to explicitly capture shared principles and patterns across spatial regions and scales. We further integrate the geo-aware embedding into a gated spatio-temporal graph neural network. This design enables the model to learn complex spatial and temporal patterns guided by geographic and hydrological context, even with sparse or no observational data. We evaluate Geo-STARS's efficacy in predicting stream water temperature, which is a master factor for water quality. Using real-world datasets spanning 37 years across multiple watersheds along the eastern coast of the United States, Geo-STARS demonstrates its superior generalization performance across both regions and scales, outperforming state-of-the-art baselines. These results highlight the promise of Geo-STARS for scalable, data-efficient environmental monitoring and decision-making.

---

### A methodology for clinically driven interactive segmentation evaluation

Authors: Parhom Esmaeili, Virginia Fernandez, Pedro Borges, Eli Gibson, Sebastien Ourselin, M. Jorge Cardoso
Published: 2025-10-10
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.09499v1

Interactive segmentation is a promising strategy for building robust, generalisable algorithms for volumetric medical image segmentation. However, inconsistent and clinically unrealistic evaluation hinders fair comparison and misrepresents real-world performance. We propose a clinically grounded methodology for defining evaluation tasks and metrics, and built a software framework for constructing standardised evaluation pipelines. We evaluate state-of-the-art algorithms across heterogeneous and complex tasks and observe that (i) minimising information loss when processing user interactions is critical for model robustness, (ii) adaptive-zooming mechanisms boost robustness and speed convergence, (iii) performance drops if validation prompting behaviour/budgets differ from training, (iv) 2D methods perform well with slab-like images and coarse targets, but 3D context helps with large or irregularly shaped targets, (v) performance of non-medical-domain models (e.g. SAM2) degrades with poor contrast and complex shapes.

---

### Unsupervised full-field Bayesian inference of orthotropic hyperelasticity from a single biaxial test: a myocardial case study

Authors: Rogier P. Krijnen, Akshay Joshi, Siddhant Kumar, Mathias Peirlinck
Published: 2025-10-10
Categories: q-bio.TO, cs.CE, cs.LG
arXiv: https://arxiv.org/abs/2510.09498v3

Cardiac muscle tissue exhibits highly non-linear hyperelastic and orthotropic material behavior during passive deformation. Traditional constitutive identification protocols therefore combine multiple loading modes and typically require multiple specimens and substantial handling. In soft living tissues, such protocols are challenged by inter- and intra-sample variability and by manipulation-induced alterations of mechanical response, which can bias inverse calibration. In this work we exploit spatially heterogeneous full-field kinematics as an information-rich alternative to multimodal testing. We recast EUCLID, an unsupervised method for the automated discovery of constitutive models, towards Bayesian parameter inference for highly nonlinear, orthotropic constitutive models. Using synthetic myocardial tissue slabs, we demonstrate that a single heterogeneous biaxial experiment, combined with sparse reaction-force measurements, enables robust recovery of Holzapfel-Ogden parameters with quantified uncertainty, across multiple noise levels. The inferred responses agree closely with ground-truth simulations and yield credible intervals that reflect the impact of measurement noise on orthotropic material model inference. Our work supports single-shot, uncertainty-aware characterization of nonlinear orthotropic material models from a single biaxial test, reducing sample demand and experimental manipulation.

---

### Toward Autonomous Soft Robotic Endovascular Navigation via Imitation Learning

Authors: Noah Barnes, Ji Woong Kim, Lingyun Di, Hannah Qu, Anuruddha Bhattacharjee, Miroslaw Janowski, Dheeraj Gandhi, Bailey Felix, Shaopeng Jiang, Olivia Young, Mark Fuge, Ryan D. Sochol et al.
Published: 2025-10-10
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2510.09497v2

In endovascular surgery, endovascular interventionists push a thin tube called a catheter, guided by a thin wire to a treatment site inside the patient's blood vessels to treat various conditions such as blood clots, aneurysms, and malformations. Robotic guidewires can enhance maneuverability but are difficult to model and control. Autonomous soft robotic guidewire navigation has the potential to overcome these challenges, increasing the precision and safety of endovascular navigation. As a first step, we establish a large-scale, 2D-projected environment for autonomous navigation. In other surgical domains, end-to-end imitation learning has shown promising results. Thus, we develop a transformer-based imitation learning framework with goal conditioning, relative action outputs, and automatic contrast dye injections to enable generalizable soft robot navigation in an aneurysm targeting task. We train the policy on 36 different modular bifurcated geometries, generating 647 total demonstrations under simulated fluoroscopy, and evaluate it on three previously unseen vascular geometries. The policy reaches the aneurysm with a success rate of 83% on the unseen geometries, outperforming several baselines. In addition, ablation and baseline studies evaluate the effectiveness of each design and data collection choice. Lastly, we extend the policy to achieve 75% success on an unseen patient-derived geometry. Project website: https://softrobotnavigation.github.io/

---

### Precoder Design in Multi-User FDD Systems with VQ-VAE and GNN

Authors: Srikar Allaparapu, Michael Baur, Benedikt Böck, Michael Joham, Wolfgang Utschick
Published: 2025-10-10
Categories: cs.IT, cs.AI, eess.SP
arXiv: https://arxiv.org/abs/2510.09495v1

Robust precoding is efficiently feasible in frequency division duplex (FDD) systems by incorporating the learnt statistics of the propagation environment through a generative model. We build on previous work that successfully designed site-specific precoders based on a combination of Gaussian mixture models (GMMs) and graph neural networks (GNNs). In this paper, by utilizing a vector quantized-variational autoencoder (VQ-VAE), we circumvent one of the key drawbacks of GMMs, i.e., the number of GMM components scales exponentially to the feedback bits. In addition, the deep learning architecture of the VQ-VAE allows us to jointly train the GNN together with VQ-VAE along with pilot optimization forming an end-to-end (E2E) model, resulting in considerable performance gains in sum rate for multi-user wireless systems. Simulations demonstrate the superiority of the proposed frameworks over the conventional methods involving the sub-discrete Fourier transform (DFT) pilot matrix and iterative precoder algorithms enabling the deployment of systems characterized by fewer pilots or feedback bits.

---

### Performance Analysis of Machine Learning Algorithms in Chronic Kidney Disease Prediction

Authors: Iftekhar Ahmed, Tanzil Ebad Chowdhury, Biggo Bushon Routh, Nafisa Tasmiya, Shadman Sakib, Adil Ahmed Chowdhury
Published: 2025-10-10
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.09493v1

Kidneys are the filter of the human body. About 10% of the global population is thought to be affected by Chronic Kidney Disease (CKD), which causes kidney function to decline. To protect in danger patients from additional kidney damage, effective risk evaluation of CKD and appropriate CKD monitoring are crucial. Due to quick and precise detection capabilities, Machine Learning models can help practitioners accomplish this goal efficiently; therefore, an enormous number of diagnosis systems and processes in the healthcare sector nowadays are relying on machine learning due to its disease prediction capability. In this study, we designed and suggested disease predictive computer-aided designs for the diagnosis of CKD. The dataset for CKD is attained from the repository of machine learning of UCL, with a few missing values; those are filled in using "mean-mode" and "Random sampling method" strategies. After successfully achieving the missing data, eight ML techniques (Random Forest, SVM, Naive Bayes, Logistic Regression, KNN, XGBoost, Decision Tree, and AdaBoost) were used to establish models, and the performance evaluation comparisons among the result accuracies are measured by the techniques to find the machine learning models with the highest accuracy. Among them, Random Forest as well as Logistic Regression showed an outstanding 99% accuracy, followed by the Ada Boost, XGBoost, Naive Bayes, Decision Tree, and SVM, whereas the KNN classifier model stands last with an accuracy of 73%.

---

### Near-Optimal Second-Order Guarantees for Model-Based Adversarial Imitation Learning

Authors: Shangzhe Li, Dongruo Zhou, Weitong Zhang
Published: 2025-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.09487v3

We study online adversarial imitation learning (AIL), where an agent learns from offline expert demonstrations and interacts with the environment online without access to rewards. Despite strong empirical results, the benefits of online interaction and the impact of stochasticity remain poorly understood. We address these gaps by introducing a model-based AIL algorithm (MB-AIL) and establish its horizon-free, second-order sample-complexity guarantees under general function approximations for both expert data and reward-free interactions. These second-order bounds provide an instance-dependent result that can scale with the variance of returns under the relevant policies and therefore tighten as the system approaches determinism. Together with second-order, information-theoretic lower bounds on a newly constructed hard-instance family, we show that MB-AIL attains minimax-optimal sample complexity for online interaction (up to logarithmic factors) with limited expert demonstrations and matches the lower bound for expert demonstrations in terms of the dependence on horizon $H$, precision $ε$ and the policy variance $σ^2$. Experiments further validate our theoretical findings and demonstrate that a practical implementation of MB-AIL matches or surpasses the sample efficiency of existing methods.

---

### Locally Optimal Private Sampling: Beyond the Global Minimax

Authors: Hrad Ghoukasian, Bonwoo Lee, Shahab Asoodeh
Published: 2025-10-10
Categories: cs.LG, cs.CR, cs.CY, cs.IT
arXiv: https://arxiv.org/abs/2510.09485v1

We study the problem of sampling from a distribution under local differential privacy (LDP). Given a private distribution $P \in \mathcal{P}$, the goal is to generate a single sample from a distribution that remains close to $P$ in $f$-divergence while satisfying the constraints of LDP. This task captures the fundamental challenge of producing realistic-looking data under strong privacy guarantees. While prior work by Park et al. (NeurIPS'24) focuses on global minimax-optimality across a class of distributions, we take a local perspective. Specifically, we examine the minimax risk in a neighborhood around a fixed distribution $P_0$, and characterize its exact value, which depends on both $P_0$ and the privacy level. Our main result shows that the local minimax risk is determined by the global minimax risk when the distribution class $\mathcal{P}$ is restricted to a neighborhood around $P_0$. To establish this, we (1) extend previous work from pure LDP to the more general functional LDP framework, and (2) prove that the globally optimal functional LDP sampler yields the optimal local sampler when constrained to distributions near $P_0$. Building on this, we also derive a simple closed-form expression for the locally minimax-optimal samplers which does not depend on the choice of $f$-divergence. We further argue that this local framework naturally models private sampling with public data, where the public data distribution is represented by $P_0$. In this setting, we empirically compare our locally optimal sampler to existing global methods, and demonstrate that it consistently outperforms global minimax samplers.

---

### CRPS-LAM: Probabilistic Regional Weather Forecasting with Continuous Ranked Probability Score

Authors: Erik Larsson, Joel Oskarsson, Tomas Landelius, Fredrik Lindsten
Published: 2025-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.09484v3

Limited-Area Models (LAMs) enable weather forecasting over regional domains at higher resolutions than what is computationally feasible for global models. At such high resolutions, machine learning approaches for weather prediction increasingly rely on ensemble methods to produce probabilistic forecasts. However, existing machine learning LAMs are not scalable due to relying on computationally costly diffusion models or inefficient graph neural networks. We tackle this by introducing a new hybrid CNN/GNN architecture, tailored to the LAM weather forecasting problem. Using this architecture, we construct the DET-LAM deterministic model, producing LAM forecasts both more efficiently and accurately than its graph-based competitor. We then tackle the ensemble forecasting problem, by using this architecture as a backbone for the generative model CRPS-LAM. CRPS-LAM is trained using a Continuous Ranked Probability Score (CRPS) objective, enabling efficient training and sampling in a single forward pass. This yields a speedup of $\approx \times 39$ compared to diffusion-based baselines. We evaluate our approach on regional domains in northern Europe, demonstrating that CRPS-LAM produces skillful and well-calibrated forecasts across a range of atmospheric variables.

---

### Efficient Autoregressive Inference for Transformer Probabilistic Models

Authors: Conor Hassan, Nasrulloh Loka, Cen-You Li, Daolang Huang, Paul E. Chang, Yang Yang, Francesco Silvestrin, Samuel Kaski, Luigi Acerbi
Published: 2025-10-10
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.09477v2

Set-based transformer models for amortized probabilistic inference and meta-learning, such as neural processes, prior-fitted networks, and tabular foundation models, excel at single-pass marginal prediction. However, many applications require joint distributions over multiple predictions. Purely autoregressive architectures generate these efficiently but sacrifice flexible set-conditioning. Obtaining joint distributions from set-based models requires re-encoding the entire context at each autoregressive step, which scales poorly. We introduce a causal autoregressive buffer that combines the strengths of both paradigms. The model encodes the context once and caches it; a lightweight causal buffer captures dependencies among generated targets, with each new prediction attending to both the cached context and all previously predicted targets added to the buffer. This enables efficient batched autoregressive sampling and joint predictive density evaluation. Training integrates set-based and autoregressive modes through masked attention at minimal overhead. Across synthetic functions, EEG time series, a Bayesian model comparison task, and tabular regression, our method closely matches the performance of full context re-encoding while delivering up to $20\times$ faster joint sampling and density evaluation, and up to $7\times$ lower memory usage.

---

### Few-shot multi-token DreamBooth with LoRa for style-consistent character generation

Authors: Ruben Pascual, Mikel Sesma-Sara, Aranzazu Jurio, Daniel Paternain, Mikel Galar
Published: 2025-10-10
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.09475v1

The audiovisual industry is undergoing a profound transformation as it is integrating AI developments not only to automate routine tasks but also to inspire new forms of art. This paper addresses the problem of producing a virtually unlimited number of novel characters that preserve the artistic style and shared visual traits of a small set of human-designed reference characters, thus broadening creative possibilities in animation, gaming, and related domains. Our solution builds upon DreamBooth, a well-established fine-tuning technique for text-to-image diffusion models, and adapts it to tackle two core challenges: capturing intricate character details beyond textual prompts and the few-shot nature of the training data. To achieve this, we propose a multi-token strategy, using clustering to assign separate tokens to individual characters and their collective style, combined with LoRA-based parameter-efficient fine-tuning. By removing the class-specific regularization set and introducing random tokens and embeddings during generation, our approach allows for unlimited character creation while preserving the learned style. We evaluate our method on five small specialized datasets, comparing it to relevant baselines using both quantitative metrics and a human evaluation study. Our results demonstrate that our approach produces high-quality, diverse characters while preserving the distinctive aesthetic features of the reference characters, with human evaluation further reinforcing its effectiveness and highlighting the potential of our method.

---

### Multimodal Policy Internalization for Conversational Agents

Authors: Zhenhailong Wang, Jiateng Liu, Amin Fazel, Ritesh Sarkhel, Xing Fan, Xiang Li, Chenlei Guo, Heng Ji, Ruhi Sarikaya
Published: 2025-10-10
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.09474v2

Modern conversational agents like ChatGPT and Alexa+ rely on predefined policies specifying metadata, response styles, and tool-usage rules. As these LLM-based systems expand to support diverse business and user queries, such policies, often implemented as in-context prompts, are becoming increasingly complex and lengthy, making faithful adherence difficult and imposing large fixed computational costs. With the rise of multimodal agents, policies that govern visual and multimodal behaviors are critical but remain understudied. Prior prompt-compression work mainly shortens task templates and demonstrations, while existing policy-alignment studies focus only on text-based safety rules. We introduce Multimodal Policy Internalization (MPI), a new task that internalizes reasoning-intensive multimodal policies into model parameters, enabling stronger policy-following without including the policy during inference. MPI poses unique data and algorithmic challenges. We build two datasets spanning synthetic and real-world decision-making and tool-using tasks and propose TriMPI, a three-stage training framework. TriMPI first injects policy knowledge via continual pretraining, then performs supervised finetuning, and finally applies PolicyRollout, a GRPO-style reinforcement learning extension that augments rollouts with policy-aware responses for grounded exploration. TriMPI achieves notable gains in end-to-end accuracy, generalization, and robustness to forgetting. As the first work on multimodal policy internalization, we provide datasets, training recipes, and comprehensive evaluations to foster future research. Project page: https://mikewangwzhl.github.io/TriMPI.

---

### D-TPT: Dimensional Entropy Maximization for Calibrating Test-Time Prompt Tuning in Vision-Language Models

Authors: Jisu Han, Wonjun Hwang
Published: 2025-10-10
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.09473v2

Test-time adaptation paradigm provides flexibility towards domain shifts by performing immediate adaptation on unlabeled target data from the source model. Vision-Language Models (VLMs) leverage their generalization capabilities for diverse downstream tasks, and test-time prompt tuning has emerged as a prominent solution for adapting VLMs. In this work, we explore contrastive VLMs and identify the modality gap caused by a single dominant feature dimension across modalities. We observe that the dominant dimensions in both text and image modalities exhibit high predictive sensitivity, and that constraining their influence can improve calibration error. Building on this insight, we propose dimensional entropy maximization that regularizes the distribution of textual features toward uniformity to mitigate the dependency of dominant dimensions. Our method alleviates the degradation of calibration performance in test-time prompt tuning, offering a simple yet effective solution to enhance the reliability of VLMs in real-world deployment scenarios.

---

### Hybrid Models for Natural Language Reasoning: The Case of Syllogistic Logic

Authors: Manuel Vargas Guzmán, Jakub Szymanik, Maciej Malicki
Published: 2025-10-10
Categories: cs.CL, cs.LG, cs.LO
arXiv: https://arxiv.org/abs/2510.09472v2

Despite the remarkable progress in neural models, their ability to generalize, a cornerstone for applications such as logical reasoning, remains a critical challenge. We delineate two fundamental aspects of this ability: compositionality, the capacity to abstract atomic logical rules underlying complex inferences, and recursiveness, the aptitude to build intricate representations through iterative application of inference rules. In the literature, these two aspects are often conflated under the umbrella term of generalization. To sharpen this distinction, we investigate the logical generalization capabilities of LLMs using the syllogistic fragment as a benchmark for natural language reasoning. We extend classical syllogistic forms to construct more complex structures, yielding a foundational yet expressive subset of formal logic that supports controlled evaluation of essential reasoning abilities. Our findings on this non-trivial benchmark show that, while LLMs demonstrate reasonable proficiency in recursiveness, they struggle with compositionality. This disparity is not uniform, as a more detailed analysis reveals substantial variability in generalization performance across individual syllogistic types, ranging from near-perfect accuracy to significantly lower performance. To overcome these limitations and establish a reliable logical prover, we propose a hybrid architecture integrating symbolic reasoning with neural computation. This synergistic interaction enables robust and efficient inference, neural components accelerate processing, while symbolic reasoning guarantees completeness. Our experiments further show that high efficiency is preserved even when using relatively small neural components. Overall, our analysis provides both a rationale for hybrid neuro-symbolic approaches and evidence of their potential to address key generalization barriers in neural reasoning systems.

---

### Getting Your Indices in a Row: Full-Text Search for LLM Training Data for Real World

Authors: Ines Altemir Marinas, Anastasiia Kucherenko, Alexander Sternfeld, Andrei Kucharavy
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09471v1

The performance of Large Language Models (LLMs) is determined by their training data. Despite the proliferation of open-weight LLMs, access to LLM training data has remained limited. Even for fully open LLMs, the scale of the data makes it all but inscrutable to the general scientific community, despite potentially containing critical data scraped from the internet. In this paper, we present the full-text indexing pipeline for the Apertus LLM training data. Leveraging Elasticsearch parallel indices and the Alps infrastructure, a state-of-the-art, highly energy-efficient arm64 supercluster, we were able to index 8.6T tokens out of 15.2T used to train the Apertus LLM family, creating both a critical LLM safety tool and effectively an offline, curated, open web search engine. Our contribution is threefold. First, we demonstrate that Elasticsearch can be successfully ported onto next-generation arm64-based infrastructure. Second, we demonstrate that full-text indexing at the scale of modern LLM training datasets and the entire open web is feasible and accessible. Finally, we demonstrate that such indices can be used to ensure previously inaccessible jailbreak-agnostic LLM safety. We hope that our findings will be useful to other teams attempting large-scale data indexing and facilitate the general transition towards greener computation.

---

### Towards Information-Optimized Multi-Agent Path Finding: A Hybrid Framework with Reduced Inter-Agent Information Sharing

Authors: Bharath Muppasani, Ritirupa Dey, Biplav Srivastava, Vignesh Narayanan
Published: 2025-10-10
Categories: cs.MA, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2510.09469v2

Multi-agent pathfinding (MAPF) remains a critical problem in robotics and autonomous systems, where agents must navigate shared spaces efficiently while avoiding conflicts. Traditional centralized algorithms with global information provide high-quality solutions but scale poorly in large-scale scenarios due to the combinatorial explosion of conflicts. Conversely, distributed approaches that have local information, particularly learning-based methods, offer better scalability by operating with relaxed information availability, yet often at the cost of solution quality. In realistic deployments, information is a constrained resource: broadcasting full agent states and goals can raise privacy concerns, strain limited bandwidth, and require extra sensing and communication hardware, increasing cost and energy use. We focus on the core question of how MAPF can be solved with minimal inter-agent information sharing while preserving solution feasibility. To this end, we present an information-centric formulation of the MAPF problem and introduce a hybrid framework, IO-MAPF, that integrates decentralized path planning with a lightweight centralized coordinator. In this framework, agents use reinforcement learning (RL) to plan independently, while the central coordinator provides minimal, targeted signals, such as static conflict-cell indicators or short conflict trajectories, that are dynamically shared to support efficient conflict resolution. We introduce an Information Units (IU) metric to quantify information use and show that our alert-driven design achieves 2x to 23x reduction in information sharing, compared to the state-of-the-art algorithms, while maintaining high success rates, demonstrating that reliable MAPF is achievable under strongly information-restricted, privacy-preserving conditions. We demonstrate the effectiveness of our algorithm using simulation and hardware experiments.

---

### Geodesic Calculus on Implicitly Defined Latent Manifolds

Authors: Florine Hartwig, Josua Sassen, Juliane Braunsmann, Martin Rumpf, Benedikt Wirth
Published: 2025-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.09468v3

Latent manifolds of autoencoders provide low-dimensional representations of data, which can be studied from a geometric perspective. We propose to describe these latent manifolds as implicit submanifolds of some ambient latent space. Based on this, we develop tools for a discrete Riemannian calculus approximating classical geometric operators. These tools are robust against inaccuracies of the implicit representation often occurring in practical examples. To obtain a suitable implicit representation, we propose to learn an approximate projection onto the latent manifold by minimizing a denoising objective. This approach is independent of the underlying autoencoder and supports the use of different Riemannian geometries on the latent manifolds. The framework in particular enables the computation of geodesic paths connecting given end points and shooting geodesics via the Riemannian exponential maps on latent manifolds. We evaluate our approach on various autoencoders trained on synthetic and real data.

---

### Interpretable Machine Learning for Predicting Startup Funding, Patenting, and Exits

Authors: Saeid Mashhadi, Amirhossein Saghezchi, Vesal Ghassemzadeh Kashani
Published: 2025-10-10
Categories: cs.LG, q-fin.GN
arXiv: https://arxiv.org/abs/2510.09465v2

This study develops an interpretable machine learning framework to forecast startup outcomes, including funding, patenting, and exit. A firm-quarter panel for 2010-2023 is constructed from Crunchbase and matched to U.S. Patent and Trademark Office (USPTO) data. Three horizons are evaluated: next funding within 12 months, patent-stock growth within 24 months, and exit through an initial public offering (IPO) or acquisition within 36 months. Preprocessing is fit on a development window (2010-2019) and applied without change to later cohorts to avoid leakage. Class imbalance is addressed using inverse-prevalence weights and the Synthetic Minority Oversampling Technique for Nominal and Continuous features (SMOTE-NC). Logistic regression and tree ensembles, including Random Forest, XGBoost, LightGBM, and CatBoost, are compared using the area under the precision-recall curve (PR-AUC) and the area under the receiver operating characteristic curve (AUROC). Patent, funding, and exit predictions achieve AUROC values of 0.921, 0.817, and 0.872, providing transparent and reproducible rankings for innovation finance.

---

### Adaptive Attacks on Trusted Monitors Subvert AI Control Protocols

Authors: Mikhail Terekhov, Alexander Panfilov, Daniil Dzenhaliou, Caglar Gulcehre, Maksym Andriushchenko, Ameya Prabhu, Jonas Geiping
Published: 2025-10-10
Categories: cs.LG, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2510.09462v2

AI control protocols serve as a defense mechanism to stop untrusted LLM agents from causing harm in autonomous settings. Prior work treats this as a security problem, stress testing with exploits that use the deployment context to subtly complete harmful side tasks, such as backdoor insertion. In practice, most AI control protocols are fundamentally based on LLM monitors, which can become a central point of failure. We study adaptive attacks by an untrusted model that knows the protocol and the monitor model, which is plausible if the untrusted model was trained with a later knowledge cutoff or can search for this information autonomously. We instantiate a simple adaptive attack vector by which the attacker embeds publicly known or zero-shot prompt injections in the model outputs. Using this tactic, frontier models consistently evade diverse monitors and complete malicious tasks on two main AI control benchmarks. The attack works universally against current protocols that rely on a monitor. Furthermore, the recent Defer-to-Resample protocol even backfires, as its resampling amplifies the prompt injection and effectively reframes it as a best-of-$n$ attack. In general, adaptive attacks on monitor models represent a major blind spot in current control protocols and should become a standard component of evaluations for future AI control mechanisms.

---

### Failure Prediction at Runtime for Generative Robot Policies

Authors: Ralf Römer, Adrian Kobras, Luca Worbis, Angela P. Schoellig
Published: 2025-10-10
Categories: cs.RO, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.09459v2

Imitation learning (IL) with generative models, such as diffusion and flow matching, has enabled robots to perform complex, long-horizon tasks. However, distribution shifts from unseen environments or compounding action errors can still cause unpredictable and unsafe behavior, leading to task failure. Early failure prediction during runtime is therefore essential for deploying robots in human-centered and safety-critical environments. We propose FIPER, a general framework for Failure Prediction at Runtime for generative IL policies that does not require failure data. FIPER identifies two key indicators of impending failure: (i) out-of-distribution (OOD) observations detected via random network distillation in the policy's embedding space, and (ii) high uncertainty in generated actions measured by a novel action-chunk entropy score. Both failure prediction scores are calibrated using a small set of successful rollouts via conformal prediction. A failure alarm is triggered when both indicators, aggregated over short time windows, exceed their thresholds. We evaluate FIPER across five simulation and real-world environments involving diverse failure modes. Our results demonstrate that FIPER better distinguishes actual failures from benign OOD situations and predicts failures more accurately and earlier than existing methods. We thus consider this work an important step towards more interpretable and safer generative robot policies. Code, data and videos are available at https://tum-lsy.github.io/fiper_website.

---

### SilvaScenes: Tree Detection and Species Classification from Under-Canopy Images in Natural Forests

Authors: David-Alexandre Duclos, William Guimont-Martin, Gabriel Jeanson, Arthur Larochelle-Tremblay, Martine Lapointe, Théo Defosse, Frédéric Moore, Philippe Nolet, François Pomerleau, Philippe Giguère
Published: 2025-10-10
Categories: cs.CV, cs.AI, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2510.09458v2

Interest in forestry automation is growing alongside rapid advances in deep learning. In particular, tree detection and taxonomic classification are seen as core tasks required for automating field surveys and forestry equipment. These operations must often be performed in under-canopy settings, which pose challenging conditions for perception systems, including heavy occlusion, variable lighting, and dense vegetation. Despite this necessity, current work has yet to properly establish the feasibility of simultaneously executing tree detection and taxonomic classification in natural forests, as available datasets primarily focus on urban settings or on a limited number of species. To address this gap, we present SilvaScenes, a benchmark dataset for instance segmentation of tree species from under-canopy images in natural forests. Collected across five bioclimatic domains in Quebec, Canada, our dataset features 1421 trees from 28 species, with segmentation masks for pixel-precise tree trunk detection and fine-grained species annotations from forestry experts. We demonstrate the relevance and difficult nature of SilvaScenes by evaluating modern deep learning approaches, showing that while trunk segmentation is feasible, with a top mean average precision (mAP) of 69.9% and mean average recall (mAR) of 76.4%, species-aware segmentation remains a significant challenge with an mAP and an mAR of only 39.2% and 68.6%, respectively. Alongside additional experiments, we highlight key challenges, namely that species imbalance and tree occlusion figure among the most pressing issues for precise segmentation and identification. Meanwhile, higher image resolutions contribute to significant performance gains and will likely prove fundamental to these tasks moving forward. Our dataset, source code, and models will be made available at https://github.com/norlab-ulaval/SilvaScenes.

---

### On Uniformly Scaling Flows: A Density-Aligned Approach to Deep One-Class Classification

Authors: Faried Abu Zaid, Tim Katzke, Emmanuel Müller, Daniel Neider
Published: 2025-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.09452v1

Unsupervised anomaly detection is often framed around two widely studied paradigms. Deep one-class classification, exemplified by Deep SVDD, learns compact latent representations of normality, while density estimators realized by normalizing flows directly model the likelihood of nominal data. In this work, we show that uniformly scaling flows (USFs), normalizing flows with a constant Jacobian determinant, precisely connect these approaches. Specifically, we prove how training a USF via maximum-likelihood reduces to a Deep SVDD objective with a unique regularization that inherently prevents representational collapse. This theoretical bridge implies that USFs inherit both the density faithfulness of flows and the distance-based reasoning of one-class methods. We further demonstrate that USFs induce a tighter alignment between negative log-likelihood and latent norm than either Deep SVDD or non-USFs, and how recent hybrid approaches combining one-class objectives with VAEs can be naturally extended to USFs. Consequently, we advocate using USFs as a drop-in replacement for non-USFs in modern anomaly detection architectures. Empirically, this substitution yields consistent performance gains and substantially improved training stability across multiple benchmarks and model backbones for both image-level and pixel-level detection. These results unify two major anomaly detection paradigms, advancing both theoretical understanding and practical performance.

---

### Adaptive Decoding via Hierarchical Neural Information Gradients in Mouse Visual Tasks

Authors: Jingyi Feng, Xiang Feng
Published: 2025-10-10
Categories: q-bio.NC, cs.NE
arXiv: https://arxiv.org/abs/2510.09451v1

Understanding the encoding and decoding mechanisms of dynamic neural responses to different visual stimuli is an important topic in exploring how the brain represents visual information. Currently, hierarchically deep neural networks (DNNs) have played a significant role as tools for mining the core features of complex data. However, most methods often overlook the dynamic generation process of neural data, such as hierarchical brain's visual data, within the brain's structure. In the decoding of brain's visual data, two main paradigms are 'fine-grained decoding tests' and 'rough-grained decoding tests', which we define as focusing on a single brain region and studying the overall structure across multiple brain regions, respectively. In this paper, we mainly use the Visual Coding Neuropixel dataset from the Allen Brain Institute, and the hierarchical information extracted from some single brain regions (i.e., fine-grained decoding tests) is provided to the proposed method for studying the adaptive topological decoding between brain regions, called the Adaptive Topological Vision Transformer, or AT-ViT. In numerous experiments, the results reveal the importance of the proposed method in hierarchical networks in the visual tasks, and also validate the hypothesis that "the hierarchical information content in brain regions of the visual system can be quantified by decoding outcomes to reflect an information hierarchy." Among them, we found that neural data collected in the hippocampus can have a random decoding performance, and this negative impact on performance still holds significant scientific value.

---

### Dynamic Weight-based Temporal Aggregation for Low-light Video Enhancement Under Extreme Noise

Authors: Ruirui Lin, Guoxi Huang, Nantheera Anantrasirichai
Published: 2025-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.09450v2

Low-light video enhancement (LLVE) is challenging due to noise, low contrast, and color degradation. While learning-based methods enable fast inference, they often fail under heavy real-world noise because they do not sufficiently exploit long-term temporal cues. We propose DWTA-Net, a novel deep-learning recurrent LLVE framework with a recurrent design. DWTA-Net adopts an integrated two-stage architecture: Stage I restores local structure and color via multi-frame alignment for temporally consistent Mamba-based enhancement, while Stage II performs recurrent refinement using a novel dynamic weight-based temporal aggregation guided by optical flow, functioning as a recurrent denoiser that adapts to motion. We further introduce a texture-adaptive loss that preserves fine details in textured regions while suppressing noise in homogeneous areas. Experiments on real-world low-light footage show that DWTA-Net achieves stronger noise suppression and fewer artifacts, delivering superior visual quality compared with state-of-the-art methods.

---

### Mono4DEditor: Text-Driven 4D Scene Editing from Monocular Video via Point-Level Localization of Language-Embedded Gaussians

Authors: Jin-Chuan Shi, Chengye Su, Jiajun Wang, Ariel Shamir, Miao Wang
Published: 2025-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.09438v1

Editing 4D scenes reconstructed from monocular videos based on text prompts is a valuable yet challenging task with broad applications in content creation and virtual environments. The key difficulty lies in achieving semantically precise edits in localized regions of complex, dynamic scenes, while preserving the integrity of unedited content. To address this, we introduce Mono4DEditor, a novel framework for flexible and accurate text-driven 4D scene editing. Our method augments 3D Gaussians with quantized CLIP features to form a language-embedded dynamic representation, enabling efficient semantic querying of arbitrary spatial regions. We further propose a two-stage point-level localization strategy that first selects candidate Gaussians via CLIP similarity and then refines their spatial extent to improve accuracy. Finally, targeted edits are performed on localized regions using a diffusion-based video editing model, with flow and scribble guidance ensuring spatial fidelity and temporal coherence. Extensive experiments demonstrate that Mono4DEditor enables high-quality, text-driven edits across diverse scenes and object types, while preserving the appearance and geometry of unedited areas and surpassing prior approaches in both flexibility and visual fidelity.

---

### Cross-attention Secretly Performs Orthogonal Alignment in Recommendation Models

Authors: Hyunin Lee, Yong Zhang, Hoang Vu Nguyen, Xiaoyi Liu, Namyong Park, Christopher Jung, Rong Jin, Yang Wang, Zhigang Wang, Somayeh Sojoudi, Xue Feng
Published: 2025-10-10
Categories: cs.LG, cs.IR
arXiv: https://arxiv.org/abs/2510.09435v1

Cross-domain sequential recommendation (CDSR) aims to align heterogeneous user behavior sequences collected from different domains. While cross-attention is widely used to enhance alignment and improve recommendation performance, its underlying mechanism is not fully understood. Most researchers interpret cross-attention as residual alignment, where the output is generated by removing redundant and preserving non-redundant information from the query input by referencing another domain data which is input key and value. Beyond the prevailing view, we introduce Orthogonal Alignment, a phenomenon in which cross-attention discovers novel information that is not present in the query input, and further argue that those two contrasting alignment mechanisms can co-exist in recommendation models We find that when the query input and output of cross-attention are orthogonal, model performance improves over 300 experiments. Notably, Orthogonal Alignment emerges naturally, without any explicit orthogonality constraints. Our key insight is that Orthogonal Alignment emerges naturally because it improves scaling law. We show that baselines additionally incorporating cross-attention module outperform parameter-matched baselines, achieving a superior accuracy-per-model parameter. We hope these findings offer new directions for parameter-efficient scaling in multi-modal research.

---

### Domain-Adapted Pre-trained Language Models for Implicit Information Extraction in Crash Narratives

Authors: Xixi Wang, Jordanka Kovaceva, Miguel Costa, Shuai Wang, Francisco Camara Pereira, Robert Thomson
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09434v1

Free-text crash narratives recorded in real-world crash databases have been shown to play a significant role in improving traffic safety. However, large-scale analyses remain difficult to implement as there are no documented tools that can batch process the unstructured, non standardized text content written by various authors with diverse experience and attention to detail. In recent years, Transformer-based pre-trained language models (PLMs), such as Bidirectional Encoder Representations from Transformers (BERT) and large language models (LLMs), have demonstrated strong capabilities across various natural language processing tasks. These models can extract explicit facts from crash narratives, but their performance declines on inference-heavy tasks in, for example, Crash Type identification, which can involve nearly 100 categories. Moreover, relying on closed LLMs through external APIs raises privacy concerns for sensitive crash data. Additionally, these black-box tools often underperform due to limited domain knowledge. Motivated by these challenges, we study whether compact open-source PLMs can support reasoning-intensive extraction from crash narratives. We target two challenging objectives: 1) identifying the Manner of Collision for a crash, and 2) Crash Type for each vehicle involved in the crash event from real-world crash narratives. To bridge domain gaps, we apply fine-tuning techniques to inject task-specific knowledge to LLMs with Low-Rank Adaption (LoRA) and BERT. Experiments on the authoritative real-world dataset Crash Investigation Sampling System (CISS) demonstrate that our fine-tuned compact models outperform strong closed LLMs, such as GPT-4o, while requiring only minimal training resources. Further analysis reveals that the fine-tuned PLMs can capture richer narrative details and even correct some mislabeled annotations in the dataset.

---

### KORMo: Korean Open Reasoning Model for Everyone

Authors: Minjun Kim, Hyeonseok Lim, Hangyeol Yoo, Inho Won, Seungwoo Song, Minkyung Cho, Junhun Yuk, Changsu Choi, Dongjae Shin, Huige Lee, Hoyun Song, Alice Oh et al.
Published: 2025-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.09426v1

This work presents the first large-scale investigation into constructing a fully open bilingual large language model (LLM) for a non-English language, specifically Korean, trained predominantly on synthetic data. We introduce KORMo-10B, a 10.8B-parameter model trained from scratch on a Korean-English corpus in which 68.74% of the Korean portion is synthetic. Through systematic experimentation, we demonstrate that synthetic data, when carefully curated with balanced linguistic coverage and diverse instruction styles, does not cause instability or degradation during large-scale pretraining. Furthermore, the model achieves performance comparable to that of contemporary open-weight multilingual baselines across a wide range of reasoning, knowledge, and instruction-following benchmarks. Our experiments reveal two key findings: (1) synthetic data can reliably sustain long-horizon pretraining without model collapse, and (2) bilingual instruction tuning enables near-native reasoning and discourse coherence in Korean. By fully releasing all components including data, code, training recipes, and logs, this work establishes a transparent framework for developing synthetic data-driven fully open models (FOMs) in low-resource settings and sets a reproducible precedent for future multilingual LLM research.

---