# Research Papers — 2025-10 · bundle page 35 of 217

> Full-text markdown of papers #1701–1750 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/35
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/34.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/36.md)

### Through the Lens: Benchmarking Deepfake Detectors Against Moiré-Induced Distortions

Authors: Razaib Tariq, Minji Heo, Simon S. Woo, Shahroz Tariq
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23225v2

Deepfake detection remains a pressing challenge, particularly in real-world settings where smartphone-captured media from digital screens often introduces Moiré artifacts that can distort detection outcomes. This study systematically evaluates state-of-the-art (SOTA) deepfake detectors on Moiré-affected videos, an issue that has received little attention. We collected a dataset of 12,832 videos, spanning 35.64 hours, from the Celeb-DF, DFD, DFDC, UADFV, and FF++ datasets, capturing footage under diverse real-world conditions, including varying screens, smartphones, lighting setups, and camera angles. To further examine the influence of Moiré patterns on deepfake detection, we conducted additional experiments using our DeepMoiréFake, referred to as (DMF) dataset and two synthetic Moiré generation techniques. Across 15 top-performing detectors, our results show that Moiré artifacts degrade performance by as much as 25.4%, while synthetically generated Moiré patterns lead to a 21.4% drop in accuracy. Surprisingly, demoiréing methods, intended as a mitigation approach, instead worsened the problem, reducing accuracy by up to 17.2%. These findings underscore the urgent need for detection models that can robustly handle Moiré distortions alongside other realworld challenges, such as compression, sharpening, and blurring. By introducing the DMF dataset, we aim to drive future research toward closing the gap between controlled experiments and practical deepfake detection.

---

### Accurate and Scalable Multimodal Pathology Retrieval via Attentive Vision-Language Alignment

Authors: Hongyi Wang, Zhengjie Zhu, Junlin Hou, Jiabo Ma, Fang Wang, Yue Shi, Qiuyu Cai, Jili Wang, Bo Luo, Zhizhong Chai, Zhengyu Zhang, Li Liang et al.
Published: 2025-10-27
Categories: cs.CV, cs.IR
arXiv: https://arxiv.org/abs/2510.23224v2

The rapid digitization of histopathology slides has opened new opportunities for computational tools in clinical and research workflows. Content-based slide retrieval can help pathologists identify morphologically and semantically related precedent cases, supporting expert diagnosis and example-based education. Effective retrieval of whole-slide images (WSIs), however, remains challenging because gigapixel slides contain abundant irrelevant content, focal diagnostic patterns and slide-level semantic information that must be represented at a practicable search cost. Here we present PathSearch, a retrieval framework that combines fine-grained attentive mosaics with slide-level embeddings aligned through vision-language contrastive learning. Trained on 6,926 slide-report pairs, PathSearch captures both fine-grained morphological cues and high-level semantic patterns to enable accurate and flexible retrieval. The framework supports two key functionalities: (1) mosaic-based image-to-image (I2I) retrieval, ensuring accurate and efficient slide search; and (2) multimodal retrieval, where text queries can directly retrieve relevant slides. PathSearch was evaluated on eight tasks comprising 5,021 evaluation slides, spanning malignancy assessment on frozen and hematoxylin and eosin (H\&E)-stained slides, lymph-node metastasis detection, tumor subtyping, mixed-gallery rare-cancer retrieval, and hepatocellular carcinoma (HCC) risk stratification. Internal and external experimental results demonstrate that PathSearch consistently outperforms the strongest existing methods without compromising multimodal accuracy. A multi-center reader study further demonstrated increases in task-level mean diagnostic accuracy, confidence, and inter-observer agreement with PathSearch's support. Together, these results support the effectiveness of PathSearch across diverse retrieval tasks and evaluation settings.

---

### Accelerating IC Thermal Simulation Data Generation via Block Krylov and Operator Action

Authors: Hong Wang, Wenkai Yang, Jie Wang, Huanshuo Dong, Zijie Geng, Zhen Huang, Depeng Xie, Zhezheng Hao, Hande Dong
Published: 2025-10-27
Categories: cs.AI, physics.comp-ph
arXiv: https://arxiv.org/abs/2510.23221v1

Recent advances in data-driven approaches, such as neural operators (NOs), have shown substantial efficacy in reducing the solution time for integrated circuit (IC) thermal simulations. However, a limitation of these approaches is requiring a large amount of high-fidelity training data, such as chip parameters and temperature distributions, thereby incurring significant computational costs. To address this challenge, we propose a novel algorithm for the generation of IC thermal simulation data, named block Krylov and operator action (BlocKOA), which simultaneously accelerates the data generation process and enhances the precision of generated data. BlocKOA is specifically designed for IC applications. Initially, we use the block Krylov algorithm based on the structure of the heat equation to quickly obtain a few basic solutions. Then we combine them to get numerous temperature distributions that satisfy the physical constraints. Finally, we apply heat operators on these functions to determine the heat source distributions, efficiently generating precise data points. Theoretical analysis shows that the time complexity of BlocKOA is one order lower than the existing method. Experimental results further validate its efficiency, showing that BlocKOA achieves a 420-fold speedup in generating thermal simulation data for 5000 chips with varying physical parameters and IC structures. Even with just 4% of the generation time, data-driven approaches trained on the data generated by BlocKOA exhibits comparable performance to that using the existing method.

---

### Process Reward Models for Sentence-Level Verification of LVLM Radiology Reports

Authors: Alois Thomas, Maya Varma, Jean-Benoit Delbrouck, Curtis P. Langlotz
Published: 2025-10-27
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.23217v1

Automating radiology report generation with Large Vision-Language Models (LVLMs) holds great potential, yet these models often produce clinically critical hallucinations, posing serious risks. Existing hallucination detection methods frequently lack the necessary sentence-level granularity or robust generalization across different LVLM generators. We introduce a novel approach: a sentence-level Process Reward Model (PRM) adapted for this vision-language task. Our PRM predicts the factual correctness of each generated sentence, conditioned on clinical context and preceding text. When fine-tuned on MIMIC-CXR with weakly-supervised labels, a lightweight 0.5B-parameter PRM outperforms existing verification techniques, demonstrating, for instance, relative improvements of 7.5% in Matthews Correlation Coefficient and 1.8% in AUROC over strong white-box baselines on outputs from one LVLM. Unlike methods reliant on internal model states, our PRM demonstrates strong generalization to an unseen LVLM. We further show its practical utility: PRM scores effectively filter low-quality reports, improving F1-CheXbert scores by 4.5% (when discarding the worst 10% of reports). Moreover, when guiding a novel weighted best-of-N selection process on the MIMIC-CXR test set, our PRM show relative improvements in clinical metrics of 7.4% for F1-CheXbert and 0.6% for BERTScore. These results demonstrate that a lightweight, context-aware PRM provides a model-agnostic safety layer for clinical LVLMs without access to internal activations

---

### Human-Like Goalkeeping in a Realistic Football Simulation: a Sample-Efficient Reinforcement Learning Approach

Authors: Alessandro Sestini, Joakim Bergdahl, Jean-Philippe Barrette-LaPierre, Florian Fuchs, Brady Chen, Fabio Zinno, Michael Jones, Linus Gisslén
Published: 2025-10-27
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.23216v4

While several high profile video games have served as testbeds for Deep Reinforcement Learning (DRL), this technique has rarely been employed by the game industry for crafting authentic AI behaviors. Previous research focuses on training super-human agents with large models, which is impractical for game studios with limited resources aiming for human-like agents. This paper proposes a sample-efficient DRL method tailored for training and fine-tuning agents in industrial settings such as the video game industry. Our method improves sample efficiency of value-based DRL by leveraging pre-collected data and increasing network plasticity. We evaluate our method training a goalkeeper agent in EA SPORTS FC 25, one of the best-selling football simulations today. Our agent outperforms the game's built-in AI by 10% in ball saving rate. Ablation studies show that our method trains agents 50% faster compared to standard DRL methods. Finally, qualitative evaluation from domain experts indicates that our approach creates more human-like gameplay compared to hand-crafted agents. As a testament to the impact of the approach, the method has been adopted for use in the most recent release of the series.

---

### Accelerating Eigenvalue Dataset Generation via Chebyshev Subspace Filter

Authors: Hong Wang, Jie Wang, Jian Luo, huanshuo dong, Yeqiu Chen, Runmin Jiang, Zhen huang
Published: 2025-10-27
Categories: cs.LG, cs.AI, math.NA
arXiv: https://arxiv.org/abs/2510.23215v2

Eigenvalue problems are among the most important topics in many scientific disciplines. With the recent surge and development of machine learning, neural eigenvalue methods have attracted significant attention as a forward pass of inference requires only a tiny fraction of the computation time compared to traditional solvers. However, a key limitation is the requirement for large amounts of labeled data in training, including operators and their eigenvalues. To tackle this limitation, we propose a novel method, named Sorting Chebyshev Subspace Filter (SCSF), which significantly accelerates eigenvalue data generation by leveraging similarities between operators -- a factor overlooked by existing methods. Specifically, SCSF employs truncated fast Fourier transform sorting to group operators with similar eigenvalue distributions and constructs a Chebyshev subspace filter that leverages eigenpairs from previously solved problems to assist in solving subsequent ones, reducing redundant computations. To the best of our knowledge, SCSF is the first method to accelerate eigenvalue data generation. Experimental results show that SCSF achieves up to a 3.5 times speedup compared to various numerical solvers.

---

### AUPO -- Abstracted Until Proven Otherwise: A Reward Distribution Based Abstraction Algorithm

Authors: Robin Schmöcker, Alexander Dockhorn, Bodo Rosenhahn
Published: 2025-10-27
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.23214v1

We introduce a novel, drop-in modification to Monte Carlo Tree Search's (MCTS) decision policy that we call AUPO. Comparisons based on a range of IPPC benchmark problems show that AUPO clearly outperforms MCTS. AUPO is an automatic action abstraction algorithm that solely relies on reward distribution statistics acquired during the MCTS. Thus, unlike other automatic abstraction algorithms, AUPO requires neither access to transition probabilities nor does AUPO require a directed acyclic search graph to build its abstraction, allowing AUPO to detect symmetric actions that state-of-the-art frameworks like ASAP struggle with when the resulting symmetric states are far apart in state space. Furthermore, as AUPO only affects the decision policy, it is not mutually exclusive with other abstraction techniques that only affect the tree search.

---

### Increasing LLM Coding Capabilities through Diverse Synthetic Coding Tasks

Authors: Amal Abed, Ivan Lukic, Jörg K. H. Franke, Frank Hutter
Published: 2025-10-27
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.23208v1

Large language models (LLMs) have shown impressive promise in code generation, yet their progress remains limited by the shortage of large-scale datasets that are both diverse and well-aligned with human reasoning. Most existing resources pair problems with solutions, but omit the intermediate thought process that guides coding. To close this gap, we present a scalable synthetic data generation pipeline that produces nearly 800k instruction-reasoning-code-test quadruplets. Each sample combines a task, a step-by-step reasoning trace, a working solution, and executable tests, enabling models to learn not just the what but also the how of problem solving. Our pipeline combines four key components: curated contest problems, web-mined content filtered by relevance classifiers, data expansion guided by reasoning patterns, and multi-stage execution-based validation. A genetic mutation algorithm further increases task diversity while maintaining consistency between reasoning traces and code implementations. Our key finding is that fine-tuning LLMs on this dataset yields consistent improvements on coding benchmarks. Beyond raw accuracy, reasoning-aware data can substitute for model scaling, generalize across architectures, and outperform leading open-source alternatives under identical sample budgets. Our work establishes reasoning-centered synthetic data generation as an efficient approach for advancing coding capabilities in LLMs. We publish our dataset and generation pipeline to facilitate further research.

---

### VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting

Authors: Hoonhee Cho, Jae-Young Kang, Giwon Lee, Hyemin Yang, Heejun Park, Seokwoo Jung, Kuk-Jin Yoon
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23205v1

End-to-end autonomous driving (E2E-AD) has emerged as a promising paradigm that unifies perception, prediction, and planning into a holistic, data-driven framework. However, achieving robustness to varying camera viewpoints, a common real-world challenge due to diverse vehicle configurations, remains an open problem. In this work, we propose VR-Drive, a novel E2E-AD framework that addresses viewpoint generalization by jointly learning 3D scene reconstruction as an auxiliary task to enable planning-aware view synthesis. Unlike prior scene-specific synthesis approaches, VR-Drive adopts a feed-forward inference strategy that supports online training-time augmentation from sparse views without additional annotations. To further improve viewpoint consistency, we introduce a viewpoint-mixed memory bank that facilitates temporal interaction across multiple viewpoints and a viewpoint-consistent distillation strategy that transfers knowledge from original to synthesized views. Trained in a fully end-to-end manner, VR-Drive effectively mitigates synthesis-induced noise and improves planning under viewpoint shifts. In addition, we release a new benchmark dataset to evaluate E2E-AD performance under novel camera viewpoints, enabling comprehensive analysis. Our results demonstrate that VR-Drive is a scalable and robust solution for the real-world deployment of end-to-end autonomous driving systems.

---

### DecoDINO: 3D Human-Scene Contact Prediction with Semantic Classification

Authors: Lukas Bierling, Davide Pasero, Fleur Dolmans, Helia Ghasemi, Angelo Broere
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23203v1

Accurate vertex-level contact prediction between humans and surrounding objects is a prerequisite for high fidelity human object interaction models used in robotics, AR/VR, and behavioral simulation. DECO was the first in the wild estimator for this task but is limited to binary contact maps and struggles with soft surfaces, occlusions, children, and false-positive foot contacts. We address these issues and introduce DecoDINO, a three-branch network based on DECO's framework. It uses two DINOv2 ViT-g/14 encoders, class-balanced loss weighting to reduce bias, and patch-level cross-attention for improved local reasoning. Vertex features are finally passed through a lightweight MLP with a softmax to assign semantic contact labels. We also tested a vision-language model (VLM) to integrate text features, but the simpler architecture performed better and was used instead. On the DAMON benchmark, DecoDINO (i) raises the binary-contact F1 score by 7$\%$, (ii) halves the geodesic error, and (iii) augments predictions with object-level semantic labels. Ablation studies show that LoRA fine-tuning and the dual encoders are key to these improvements. DecoDINO outperformed the challenge baseline in both tasks of the DAMON Challenge. Our code is available at https://github.com/DavidePasero/deco/tree/main.

---

### Rate-optimal Design for Anytime Best Arm Identification

Authors: Junpei Komiyama, Kyoungseok Jang, Junya Honda
Published: 2025-10-27
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.23199v3

We consider the best arm identification problem, where the goal is to identify the arm with the highest mean reward from a set of $K$ arms under a limited sampling budget. This problem models many practical scenarios such as A/B testing. We consider a class of algorithms for this problem, which is provably minimax optimal up to a constant factor. This idea is a generalization of existing works in fixed-budget best arm identification, which are limited to a particular choice of risk measures. Based on the framework, we propose Almost Tracking, a closed-form algorithm that has a provable guarantee on the popular risk measure $H_1$. Unlike existing algorithms, Almost Tracking does not require the total budget in advance nor does it need to discard a significant part of samples, which gives a practical advantage. Through experiments on synthetic and real-world datasets, we show that our algorithm outperforms existing anytime algorithms as well as fixed-budget algorithms.

---

### PTPP-Aware Adaptation Scaling Laws: Predicting Domain-Adaptation Performance at Unseen Pre-Training Budgets

Authors: Etienne Goffinet, Shane Bergsma, Avraham Sheinin, Natalia Vassilieva, Shaheer Muhammad, Preslav Nakov, Gurpreet Gosal
Published: 2025-10-27
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.23198v1

Continual pre-training (CPT) for domain adaptation must balance target-domain gains with stability on the base domain. Existing CPT scaling laws typically assume a fixed pre-training budget, which limits their ability to forecast adaptation outcomes for models trained at different tokens-per-parameter (PTPP). We present \emph{PTPP-aware} adaptation scaling laws that make the pre-training budget an explicit variable, enabling accurate \emph{prediction} of adaptation loss at unseen \ptpp. On a multilingual setup (English/Arabic $\rightarrow$ French), PTPP-aware formulations trained on early stages (\ptpp{}=\{15,31\}) predict target loss at \ptpp{}=279 and outperform a PTPP-agnostic \dcpt{} transfer baseline on metrics (Huber-on-log, MAE$_\mathrm{rel}$, calibration slope); full diagnostics (RMSE, MAPE) are in the appendix. Beyond forecasting, we show a practical use case: planning replay ratios and adaptation token budgets that satisfy target and forgetting constraints under compute limits.

---

### Model-free filtering in high dimensions via projection and score-based diffusions

Authors: Sören Christensen, Jan Kallsen, Claudia Strauch, Lukas Trottner
Published: 2025-10-27
Categories: math.ST, stat.ML
arXiv: https://arxiv.org/abs/2510.23197v1

We consider the problem of recovering a latent signal $X$ from its noisy observation $Y$. The unknown law $\mathbb{P}^X$ of $X$, and in particular its support $\mathscr{M}$, are accessible only through a large sample of i.i.d.\ observations. We further assume $\mathscr{M}$ to be a low-dimensional submanifold of a high-dimensional Euclidean space $\mathbb{R}^d$. As a filter or denoiser $\widehat X$, we suggest an estimator of the metric projection $π_{\mathscr{M}}(Y)$ of $Y$ onto the manifold $\mathscr{M}$. To compute this estimator, we study an auxiliary semiparametric model in which $Y$ is obtained by adding isotropic Laplace noise to $X$. Using score matching within a corresponding diffusion model, we obtain an estimator of the Bayesian posterior $\mathbb{P}^{X \mid Y}$ in this setup. Our main theoretical results show that, in the limit of high dimension $d$, this posterior $\mathbb{P}^{X\mid Y}$ is concentrated near the desired metric projection $π_{\mathscr{M}}(Y)$.

---

### The Benchmarking Epistemology: Validity Theory for Evaluating Machine Learning Models

Authors: Timo Freiesleben, Sebastian Zezulka
Published: 2025-10-27
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2510.23191v2

Predictive benchmarking, evaluating machine learning models based on predictive performance and competitive ranking, is central to machine learning research and scientific inquiry. However, benchmark scores at best measure performance relative to a specific dataset and learning problem. Drawing substantial scientific inferences requires additional assumptions. Adapting ideas from psychological validity theory, we propose validity conditions that make these assumptions explicit. In two case studies---ImageNet and the Fragile Families Challenge---we show how benchmark results can support inferences about research progress and limits of predictability, situating predictive benchmarking as a distinct epistemic practice in machine learning.

---

### Evaluation of Vision-LLMs in Surveillance Video

Authors: Pascal Benschop, Cristian Meo, Justin Dauwels, Jelte P. Mense
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23190v1

The widespread use of cameras in our society has created an overwhelming amount of video data, far exceeding the capacity for human monitoring. This presents a critical challenge for public safety and security, as the timely detection of anomalous or criminal events is crucial for effective response and prevention. The ability for an embodied agent to recognize unexpected events is fundamentally tied to its capacity for spatial reasoning. This paper investigates the spatial reasoning of vision-language models (VLMs) by framing anomalous action recognition as a zero-shot, language-grounded task, addressing the embodied perception challenge of interpreting dynamic 3D scenes from sparse 2D video. Specifically, we investigate whether small, pre-trained vision--LLMs can act as spatially-grounded, zero-shot anomaly detectors by converting video into text descriptions and scoring labels via textual entailment. We evaluate four open models on UCF-Crime and RWF-2000 under prompting and privacy-preserving conditions. Few-shot exemplars can improve accuracy for some models, but may increase false positives, and privacy filters -- especially full-body GAN transforms -- introduce inconsistencies that degrade accuracy. These results chart where current vision--LLMs succeed (simple, spatially salient events) and where they falter (noisy spatial cues, identity obfuscation). Looking forward, we outline concrete paths to strengthen spatial grounding without task-specific training: structure-aware prompts, lightweight spatial memory across clips, scene-graph or 3D-pose priors during description, and privacy methods that preserve action-relevant geometry. This positions zero-shot, language-grounded pipelines as adaptable building blocks for embodied, real-world video understanding. Our implementation for evaluating VLMs is publicly available at: https://github.com/pascalbenschopTU/VLLM_AnomalyRecognition

---

### DREaM: Drug-Drug Relation Extraction via Transfer Learning Method

Authors: Ali Fata, Hossein Rahmani, Parinaz Soltanzadeh, Amirhossein Derakhshan, Behrouz Minaei Bidgoli
Published: 2025-10-27
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.23189v1

Relation extraction between drugs plays a crucial role in identifying drug drug interactions and predicting side effects. The advancement of machine learning methods in relation extraction, along with the development of large medical text databases, has enabled the low cost extraction of such relations compared to other approaches that typically require expert knowledge. However, to the best of our knowledge, there are limited datasets specifically designed for drug drug relation extraction currently available. Therefore, employing transfer learning becomes necessary to apply machine learning methods in this domain. In this study, we propose DREAM, a method that first employs a trained relation extraction model to discover relations between entities and then applies this model to a corpus of medical texts to construct an ontology of drug relationships. The extracted relations are subsequently validated using a large language model. Quantitative results indicate that the LLM agreed with 71 of the relations extracted from a subset of PubMed abstracts. Furthermore, our qualitative analysis indicates that this approach can uncover ambiguities in the medical domain, highlighting the challenges inherent in relation extraction in this field.

---

### Finding 3D Scene Analogies with Multimodal Foundation Models

Authors: Junho Kim, Young Min Kim
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23184v1

Connecting current observations with prior experiences helps robots adapt and plan in new, unseen 3D environments. Recently, 3D scene analogies have been proposed to connect two 3D scenes, which are smooth maps that align scene regions with common spatial relationships. These maps enable detailed transfer of trajectories or waypoints, potentially supporting demonstration transfer for imitation learning or task plan transfer across scenes. However, existing methods for the task require additional training and fixed object vocabularies. In this work, we propose to use multimodal foundation models for finding 3D scene analogies in a zero-shot, open-vocabulary setting. Central to our approach is a hybrid neural representation of scenes that consists of a sparse graph based on vision-language model features and a feature field derived from 3D shape foundation models. 3D scene analogies are then found in a coarse-to-fine manner, by first aligning the graph and refining the correspondence with feature fields. Our method can establish accurate correspondences between complex scenes, and we showcase applications in trajectory and waypoint transfer.

---

### SI-Bench: Benchmarking Social Intelligence of Large Language Models in Human-to-Human Conversations

Authors: Shuai Huang, Wenxuan Zhao, Jun Gao
Published: 2025-10-27
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.23182v1

As large language models (LLMs) develop anthropomorphic abilities, they are increasingly being deployed as autonomous agents to interact with humans. However, evaluating their performance in realistic and complex social interactions remains a significant challenge. Most previous research built datasets through simulated agent-to-agent interactions, which fails to capture the authentic linguistic styles and relational dynamics found in real human conversations. To address this gap, we introduce SI-Bench, a novel benchmark designed to evaluate aspects of social intelligence in LLMs. Grounded in broad social science theories, SI-Bench contains 2,221 authentic multi-turn dialogues collected from a social networking application. We further selected a subset of 312 dialogues for manual annotation across 8 major models. The experiments show that SOTA models have surpassed the human expert in process reasoning under complex social situations, yet they still fall behind humans in reply quality. Moreover, introducing Chain-of-Thought (CoT) reasoning may degrade the performance of LLMs in social dialogue tasks. All datasets are openly available at https://github.com/SI-Bench/SI-Bench.git.

---

### Introducing physics-informed generative models for targeting structural novelty in the exploration of chemical space

Authors: Andrij Vasylenko, Federico Ottomano, Christopher M. Collins, Rahul Savani, Matthew S. Dyer, Matthew J. Rosseinsky
Published: 2025-10-27
Categories: cond-mat.mtrl-sci, cs.LG
arXiv: https://arxiv.org/abs/2510.23181v2

Discovering materials with new structural chemistry is key to achieving transformative functionality. Generative artificial intelligence offers a scalable route to propose candidate crystal structures. We introduce a reliable low-cost proxy for structural novelty as a conditioning property to steer generation towards novel yet physically plausible structures. We then develop a physics-informed diffusion model that embeds this descriptor of local environment diversity together with compactness as a stability metric to balance physical plausibility with structural novelty. Conditioning on these metrics improves generative performance across diffusion models, shifting generation away from structural motifs that dominate the training data. A chemically grounded validation protocol isolates those candidates that combine plausibility with structural novelty for physics-based calculation of energetic stability. Both the stability and the novelty of candidates emerging from this workflow can however change when the full potential energy surface at a candidate composition is evaluated with crystal structure prediction (CSP). This suggests a practical generative-CSP synergy for discovery-oriented exploration, where AI targets physically viable yet structurally distinct regions of chemical space for detailed physics-based assessment of novelty and stability.

---

### TARC: Time-Adaptive Robotic Control

Authors: Arnav Sukhija, Lenart Treven, Jin Cheng, Florian Dörfler, Stelian Coros, Andreas Krause
Published: 2025-10-27
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2510.23176v2

Most robotic systems rely on fixed-frequency discrete-time controllers, creating a trade-off between the efficiency of low-frequency control and the responsiveness of high-frequency feedback. As a result, systems typically default to high control rates for robustness, at the cost of wasted inference and unnecessary actuation. Addressing this, we introduce Time-Adaptive Robotic Control (TARC), a reinforcement learning framework in which the policy jointly predicts a control action and its duration of application. TARC learns temporally extended actions by optimizing task performance under soft or hard constraints on the number of control switches, enabling adaptive modulation of control rates. We evaluate TARC on two robotic hardware platforms: a high-speed RC car and the Unitree Go1 quadruped, and on a vision-language action model in simulation, where each query incurs a costly transformer forward pass. Across all settings, TARC matches the performance of high-frequency discrete-time controllers while operating at less than half their control frequency. Unlike fixed-rate controllers, TARC adapts its control frequency online, allocating high-frequency feedback only when required.

---

### Benchmarking VQE Configurations: Architectures, Initializations, and Optimizers for Silicon Ground State Energy

Authors: Zakaria Boutakka, Nouhaila Innan, Muhammed Shafique, Mohamed Bennai, Z. Sakhi
Published: 2025-10-27
Categories: quant-ph, cs.LG
arXiv: https://arxiv.org/abs/2510.23171v1

Quantum computing presents a promising path toward precise quantum chemical simulations, particularly for systems that challenge classical methods. This work investigates the performance of the Variational Quantum Eigensolver (VQE) in estimating the ground-state energy of the silicon atom, a relatively heavy element that poses significant computational complexity. Within a hybrid quantum-classical optimization framework, we implement VQE using a range of ansatz, including Double Excitation Gates, ParticleConservingU2, UCCSD, and k-UpCCGSD, combined with various optimizers such as gradient descent, SPSA, and ADAM. The main contribution of this work lies in a systematic methodological exploration of how these configuration choices interact to influence VQE performance, establishing a structured benchmark for selecting optimal settings in quantum chemical simulations. Key findings show that parameter initialization plays a decisive role in the algorithm's stability, and that the combination of a chemically inspired ansatz with adaptive optimization yields superior convergence and precision compared to conventional approaches.

---

### MATCH: Task-Driven Code Evaluation through Contrastive Learning

Authors: Marah Ghoummaid, Vladimir Tchuiev, Ofek Glick, Michal Moshkovitz, Dotan Di Castro
Published: 2025-10-27
Categories: cs.CL, cs.SE
arXiv: https://arxiv.org/abs/2510.23169v2

AI-based code generation is increasingly prevalent, with GitHub Copilot estimated to generate 46% of the code on GitHub. Accurately evaluating how well generated code aligns with developer intent remains a critical challenge. Traditional evaluation methods, such as unit tests, are often unscalable and costly. Syntactic similarity metrics (e.g., BLEU, ROUGE) fail to capture code functionality, and metrics like CodeBERTScore require reference code, which is not always available. To address the gap in reference-free evaluation, with few alternatives such as ICE-Score, this paper introduces MATCH, a novel reference-free metric. MATCH uses Contrastive Learning to generate meaningful embeddings for code and natural language task descriptions, enabling similarity scoring that reflects how well generated code implements the task. We show that MATCH achieves stronger correlations with functional correctness and human preference than existing metrics across multiple programming languages.

---

### Guiding Skill Discovery with Foundation Models

Authors: Zhao Yang, Thomas M. Moerland, Mike Preuss, Aske Plaat, Vincent François-Lavet, Edward S. Hu
Published: 2025-10-27
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.23167v1

Learning diverse skills without hand-crafted reward functions could accelerate reinforcement learning in downstream tasks. However, existing skill discovery methods focus solely on maximizing the diversity of skills without considering human preferences, which leads to undesirable behaviors and possibly dangerous skills. For instance, a cheetah robot trained using previous methods learns to roll in all directions to maximize skill diversity, whereas we would prefer it to run without flipping or entering hazardous areas. In this work, we propose a Foundation model Guided (FoG) skill discovery method, which incorporates human intentions into skill discovery through foundation models. Specifically, FoG extracts a score function from foundation models to evaluate states based on human intentions, assigning higher values to desirable states and lower to undesirable ones. These scores are then used to re-weight the rewards of skill discovery algorithms. By optimizing the re-weighted skill discovery rewards, FoG successfully learns to eliminate undesirable behaviors, such as flipping or rolling, and to avoid hazardous areas in both state-based and pixel-based tasks. Interestingly, we show that FoG can discover skills involving behaviors that are difficult to define. Interactive visualisations are available from https://sites.google.com/view/submission-fog.

---

### Beyond Direct Generation: A Decomposed Approach to Well-Crafted Screenwriting with LLMs

Authors: Hang Lei, Shengyi Zong, Zhaoyan Li, Ziren Zhou, Hao Liu, Liang Yu
Published: 2025-10-27
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.23163v3

The screenplay serves as the foundation for television production, defining narrative structure, character development, and dialogue. While Large Language Models (LLMs) show great potential in creative writing, direct end-to-end generation approaches often fail to produce well-crafted screenplays. We argue this failure stems from forcing a single model to simultaneously master two disparate capabilities: creative narrative construction and rigid format adherence. The resulting outputs may mimic superficial style but lack the deep structural integrity and storytelling substance required for professional use. To enable LLMs to generate high-quality screenplays, we introduce Dual-Stage Refinement (DSR), a decomposed framework that decouples creative narrative generation from format conversion. The first stage transforms a brief outline into rich, novel-style prose. The second stage refines this narrative into a professionally formatted screenplay. This separation enables the model to specialize in one distinct capability at each stage. A key challenge in implementing DSR is the scarcity of paired outline-to-novel training data. We address this through hybrid data synthesis: reverse synthesis deconstructs existing screenplays into structured inputs, while forward synthesis leverages these inputs to generate high-quality narrative texts as training targets. Blind evaluations by professional screenwriters show that DSR achieves a 75% win rate against strong baselines like Gemini-2.5-Pro and reaches 82.7% of human-level performance. Our work demonstrates that decomposed generation architecture with tailored data synthesis effectively specializes LLMs in complex creative domains.

---

### Reclaiming Low-Quality Instruction Data with Rule-Guided LLM Reconstruction

Authors: Zile Yang, Ling Li, Na Di, Jinlong Pang, Yao Zhou, Hao Cheng, Bo Han, Jiaheng Wei
Published: 2025-10-27
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.23160v2

Data selection for supervised fine-tuning (SFT) often discards low-scoring instruction examples that may retain useful supervision. We propose a rule-guided LLM reconstruction framework to reclaim this information. The framework identifies low-scoring samples through LLM-based rating and score correction, selects semantically representative sources, and reconstructs instruction-response pairs through strategy-guided generation, criteria-based validation, and failure-guided repair. From 123,786 low-scoring samples, it constructs 54,888 training examples. Across three 7B/8B base LLMs, the reconstructed data achieves higher average downstream performance than the low-scoring source set and all evaluated data-selection baselines. Despite using substantially fewer examples than the full source dataset (approximately 300K), it achieves higher reported average scores than full-data training across all three base models. Stepwise ablations identify rule-guided reconstruction as the largest source of observed improvements over direct fusion, while a manual audit reveals remaining limitations in source integration and answer faithfulness. These findings support reconstruction as a complementary approach to selection-based SFT data curation.

---

### Enabling Vibration-Based Gesture Recognition on Everyday Furniture via Energy-Efficient FPGA Implementation of 1D Convolutional Networks

Authors: Koki Shibata, Tianheng Ling, Chao Qian, Tomokazu Matsui, Hirohiko Suwa, Keiichi Yasumoto, Gregor Schiele
Published: 2025-10-27
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.23156v2

The growing demand for smart home interfaces has increased interest in non-intrusive sensing methods like vibration-based gesture recognition. While prior studies demonstrated feasibility, they often rely on complex preprocessing and large Neural Networks (NNs) requiring costly high-performance hardware, resulting in high energy usage and limited real-world deployability. This study proposes an energy-efficient solution deploying compact NNs on low-power Field-Programmable Gate Arrays (FPGAs) to enable real-time gesture recognition with competitive accuracy. We adopt a series of optimizations: (1) We replace complex spectral preprocessing with raw waveform input, eliminating complex on-board preprocessing while reducing input size by 21x without sacrificing accuracy. (2) We design two lightweight architectures (1D-CNN and 1D-SepCNN) tailored for embedded FPGAs, reducing parameters from 369 million to as few as 216 while maintaining comparable accuracy. (3) With integer-only quantization and automated RTL generation, we achieve seamless FPGA deployment. A ping-pong buffering mechanism in 1D-SepCNN further improves deployability under tight memory constraints. (4) We extend a hardware-aware search framework to support constraint-driven model configuration selection, considering accuracy, deployability, latency, and energy consumption. Evaluated on two swipe-direction datasets with multiple users and ordinary tables, our approach achieves low-latency, energy-efficient inference on the AMD Spartan-7 XC7S25 FPGA. Under the PS data splitting setting, the selected 6-bit 1D-CNN reaches 0.970 average accuracy across users with 9.22 ms latency. The chosen 8-bit 1D-SepCNN further reduces latency to 6.83 ms (over 53x CPU speedup) with slightly lower accuracy (0.949). Both consume under 1.2 mJ per inference, demonstrating suitability for long-term edge operation.

---

### AG-Fusion: adaptive gated multimodal fusion for 3d object detection in complex scenes

Authors: Sixian Liu, Chen Xu, Qiang Wang, Donghai Shi, Yiwen Li
Published: 2025-10-27
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.23151v1

Multimodal camera-LiDAR fusion technology has found extensive application in 3D object detection, demonstrating encouraging performance. However, existing methods exhibit significant performance degradation in challenging scenarios characterized by sensor degradation or environmental disturbances. We propose a novel Adaptive Gated Fusion (AG-Fusion) approach that selectively integrates cross-modal knowledge by identifying reliable patterns for robust detection in complex scenes. Specifically, we first project features from each modality into a unified BEV space and enhance them using a window-based attention mechanism. Subsequently, an adaptive gated fusion module based on cross-modal attention is designed to integrate these features into reliable BEV representations robust to challenging environments. Furthermore, we construct a new dataset named Excavator3D (E3D) focusing on challenging excavator operation scenarios to benchmark performance in complex conditions. Our method not only achieves competitive performance on the standard KITTI dataset with 93.92% accuracy, but also significantly outperforms the baseline by 24.88% on the challenging E3D dataset, demonstrating superior robustness to unreliable modal information in complex industrial scenes.

---

### Revisiting the Structure of Trend Premia: When Diversification Hides Redundancy

Authors: Alban Etienne, Jean-Jacques Ohana, Eric Benhamou, Béatrice Guez, Ethan Setrouk, Thomas Jacquot
Published: 2025-10-27
Categories: q-fin.PR, q-fin.PM, q-fin.RM, q-fin.TR, stat.ML
arXiv: https://arxiv.org/abs/2510.23150v2

Recent work has emphasized the diversification benefits of combining trend signals across multiple horizons, with the medium-term window-typically six months to one year-long viewed as the "sweet spot" of trend-following. This paper revisits this conventional view by reallocating exposure dynamically across horizons using a Bayesian optimization framework designed to learn the optimal weights assigned to each trend horizon at the asset level. The common practice of equal weighting implicitly assumes that all assets benefit equally from all horizons; we show that this assumption is both theoretically and empirically suboptimal. We first optimize the horizon-level weights at the asset level to maximize the informativeness of trend signals before applying Bayesian graphical models-with sparsity and turnover control-to allocate dynamically across assets. The key finding is that the medium-term band contributes little incremental performance or diversification once short- and long-term components are included. Removing the 125-day layer improves Sharpe ratios and drawdown efficiency while maintaining benchmark correlation. We then rationalize this outcome through a minimum-variance formulation, showing that the medium-term horizon largely overlaps with its neighboring horizons. The resulting "barbell" structure-combining short- and long-term trends-captures most of the performance while reducing model complexity. This result challenges the common belief that more horizons always improve diversification and suggests that some forms of time-scale diversification may conceal unnecessary redundancy in trend premia.

---

### Complexity Dependent Error Rates for Physics-informed Statistical Learning via the Small-ball Method

Authors: Diego Marcondes
Published: 2025-10-27
Categories: stat.ML, cs.LG, math.ST
arXiv: https://arxiv.org/abs/2510.23149v2

Physics-informed statistical learning (PISL) integrates empirical data with physical knowledge to enhance the statistical performance of estimators. While PISL methods are widely used in practice, a comprehensive theoretical understanding of how informed regularization affects statistical properties is still missing. Specifically, two fundamental questions have yet to be fully addressed: (A) what is the trade-off between soft penalties and hard constraints, and (B) what is the statistical gain of incorporating physical knowledge compared to purely data-driven empirical error minimisation. In this paper, we address these questions for PISL in classes of functions whose convex hull satisfies a small-ball condition, in particular convex classes, under physical knowledge expressed as linear equations by developing appropriate complexity dependent error rates based on the small-ball method. We specialise them to linear classes of finite dimension $q$ in which the space of solutions of the equation has dimension $k < q$, and show that (A) the physics-informed estimator attains the error bound of the hard-constrained empirical error minimiser whenever the penalty parameter is above an explicit threshold, and that (B) the penalty improves the error rate from $\sqrt{q/n}$ to $\sqrt{k/n}$ under a second moment condition on the noise only, having the effect of a dimensionality reduction. We also quantify how these conclusions degrade when the equation is misspecified, in terms of the excess residual of the target function. This work establishes a theoretical framework for evaluating the statistical properties of physics-informed estimators in such classes, contributing to closing the gap between statistical theory and practical PISL.

---

### Adapting Interleaved Encoders with PPO for Language-Guided Reinforcement Learning in BabyAI

Authors: Aryan Mathur, Asaduddin Ahmed
Published: 2025-10-27
Categories: cs.LG, cs.AI, eess.IV
arXiv: https://arxiv.org/abs/2510.23148v1

Deep reinforcement learning agents often struggle when tasks require understanding both vision and language. Conventional architectures typically isolate perception (for example, CNN-based visual encoders) from decision-making (policy networks). This separation can be inefficient, since the policy's failures do not directly help the perception module learn what is important. To address this, we implement the Perception-Decision Interleaving Transformer (PDiT) architecture introduced by Mao et al. (2023), a model that alternates between perception and decision layers within a single transformer. This interleaving allows feedback from decision-making to refine perceptual features dynamically. In addition, we integrate a contrastive loss inspired by CLIP to align textual mission embeddings with visual scene features. We evaluate the PDiT encoders on the BabyAI GoToLocal environment and find that the approach achieves more stable rewards and stronger alignment compared to a standard PPO baseline. The results suggest that interleaved transformer encoders are a promising direction for developing more integrated autonomous agents.

---

### Implicit Modeling for Transferability Estimation of Vision Foundation Models

Authors: Yaoyan Zheng, Huiqun Wang, Nan Zhou, Di Huang
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23145v1

Transferability estimation identifies the best pre-trained models for downstream tasks without incurring the high computational cost of full fine-tuning. This capability facilitates deployment and advances the pre-training and fine-tuning paradigm. However, existing methods often struggle to accurately assess transferability for emerging pre-trained models with diverse architectures, training strategies, and task alignments. In this work, we propose Implicit Transferability Modeling (ITM), a novel framework that implicitly models each model's intrinsic transferability, coupled with a Divide-and-Conquer Variational Approximation (DVA) strategy to efficiently approximate embedding space evolution. This design enables generalization across a broader range of models and downstream tasks. Extensive experiments on a comprehensive benchmark--spanning extensive training regimes and a wider variety of model types--demonstrate that ITM consistently outperforms existing methods in terms of stability, effectiveness, and efficiency.

---

### DQ3D: Depth-guided Query for Transformer-Based 3D Object Detection in Traffic Scenarios

Authors: Ziyu Wang, Wenhao Li, Ji Wu
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23144v1

3D object detection from multi-view images in traffic scenarios has garnered significant attention in recent years. Many existing approaches rely on object queries that are generated from 3D reference points to localize objects. However, a limitation of these methods is that some reference points are often far from the target object, which can lead to false positive detections. In this paper, we propose a depth-guided query generator for 3D object detection (DQ3D) that leverages depth information and 2D detections to ensure that reference points are sampled from the surface or interior of the object. Furthermore, to address partially occluded objects in current frame, we introduce a hybrid attention mechanism that fuses historical detection results with depth-guided queries, thereby forming hybrid queries. Evaluation on the nuScenes dataset demonstrates that our method outperforms the baseline by 6.3\% in terms of mean Average Precision (mAP) and 4.3\% in the NuScenes Detection Score (NDS).

---

### Rethinking GSPO: The Perplexity-Entropy Equivalence

Authors: Chi Liu
Published: 2025-10-27
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.23142v1

We provide a new perspective on GSPO's length-normalized importance ratios by establishing their connection to information-theoretic quantities. We show that GSPO's sequence-level weight $s(θ) = (π_θ/π_{θ_{\text{old}}})^{1/|y|}$ can be equivalently expressed as the inverse perplexity ratio $\text{PPL}_{θ_{\text{old}}}/\text{PPL}_θ$ and as the exponential cross-entropy change $\exp(ΔH)$. While the perplexity-entropy relationship follows from standard definitions, this observation provides a useful lens for understanding GSPO: the algorithm weights policy gradient updates by perplexity ratios, offering an information-theoretic interpretation of the importance weights. This perspective helps explain GSPO's empirical properties, including log-domain variance reduction through geometric averaging and stability in training mixture-of-experts models. We validate the mathematical equivalences and variance predictions through controlled experiments on mathematical reasoning tasks.

---

### Treble10: A high-quality dataset for far-field speech recognition, dereverberation, and enhancement

Authors: Sarabeth S. Mullins, Georg Götz, Eric Bezzam, Steven Zheng, Daniel Gert Nielsen
Published: 2025-10-27
Categories: eess.AS, cs.LG
arXiv: https://arxiv.org/abs/2510.23141v1

Accurate far-field speech datasets are critical for tasks such as automatic speech recognition (ASR), dereverberation, speech enhancement, and source separation. However, current datasets are limited by the trade-off between acoustic realism and scalability. Measured corpora provide faithful physics but are expensive, low-coverage, and rarely include paired clean and reverberant data. In contrast, most simulation-based datasets rely on simplified geometrical acoustics, thus failing to reproduce key physical phenomena like diffraction, scattering, and interference that govern sound propagation in complex environments. We introduce Treble10, a large-scale, physically accurate room-acoustic dataset. Treble10 contains over 3000 broadband room impulse responses (RIRs) simulated in 10 fully furnished real-world rooms, using a hybrid simulation paradigm implemented in the Treble SDK that combines a wave-based and geometrical acoustics solver. The dataset provides six complementary subsets, spanning mono, 8th-order Ambisonics, and 6-channel device RIRs, as well as pre-convolved reverberant speech scenes paired with LibriSpeech utterances. All signals are simulated at 32 kHz, accurately modelling low-frequency wave effects and high-frequency reflections. Treble10 bridges the realism gap between measurement and simulation, enabling reproducible, physically grounded evaluation and large-scale data augmentation for far-field speech tasks. The dataset is openly available via the Hugging Face Hub, and is intended as both a benchmark and a template for next-generation simulation-driven audio research.

---

### Fast Voxel-Wise Kinetic Modeling in Dynamic PET using a Physics-Informed CycleGAN

Authors: Christian Salomonsen, Samuel Kuttner, Michael Kampffmeyer, Robert Jenssen, Kristoffer Wickstrøm, Jong Chul Ye, Elisabeth Wetzer
Published: 2025-10-27
Categories: cs.CV, q-bio.OT
arXiv: https://arxiv.org/abs/2510.23140v1

Tracer kinetic modeling serves a vital role in diagnosis, treatment planning, tracer development and oncology, but burdens practitioners with complex and invasive arterial input function estimation (AIF). We adopt a physics-informed CycleGAN showing promise in DCE-MRI quantification to dynamic PET quantification. Our experiments demonstrate sound AIF predictions and parameter maps closely resembling the reference.

---

### Note on the Construction of Structure Tensor

Authors: Josef Bigun, Fernado Alonso-Fernandez
Published: 2025-10-27
Categories: cs.CV, math.SP
arXiv: https://arxiv.org/abs/2510.23137v1

This note presents a theoretical discussion of two structure tensor constructions: one proposed by Bigun and Granlund 1987, and the other by Granlund and Knutsson 1995. At first glance, these approaches may appear quite different--the former is implemented by averaging outer products of gradient filter responses, while the latter constructs the tensor from weighted outer products of tune-in frequency vectors of quadrature filters. We argue that when both constructions are viewed through the common lens of Total Least Squares (TLS) line fitting to the power spectrum, they can be reconciled to a large extent, and additional benefits emerge. From this perspective, the correction term introduced in Granlund and Knutsson 1995 becomes unnecessary. Omitting it ensures that the resulting tensor remains positive semi-definite, thereby simplifying the interpretation of its eigenvalues. Furthermore, this interpretation allows fitting more than a single 0rientation to the input by reinterpreting quadrature filter responses without relying on a structure tensor. It also removes the constraint that responses must originate strictly from quadrature filters, allowing the use of alternative filter types and non-angular tessellations. These alternatives include Gabor filters--which, although not strictly quadrature, are still suitable for structure tensor construction--even when they tessellate the spectrum in a Cartesian fashion, provided they are sufficiently concentrated.

---

### A method for outlier detection based on cluster analysis and visual expert criteria

Authors: Juan A. Lara, David Lizcano, Víctor Rampérez, Javier Soriano
Published: 2025-10-27
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.23136v1

Outlier detection is an important problem occurring in a wide range of areas. Outliers are the outcome of fraudulent behaviour, mechanical faults, human error, or simply natural deviations. Many data mining applications perform outlier detection, often as a preliminary step in order to filter out outliers and build more representative models. In this paper, we propose an outlier detection method based on a clustering process. The aim behind the proposal outlined in this paper is to overcome the specificity of many existing outlier detection techniques that fail to take into account the inherent dispersion of domain objects. The outlier detection method is based on four criteria designed to represent how human beings (experts in each domain) visually identify outliers within a set of objects after analysing the clusters. This has an advantage over other clustering-based outlier detection techniques that are founded on a purely numerical analysis of clusters. Our proposal has been evaluated, with satisfactory results, on data (particularly time series) from two different domains: stabilometry, a branch of medicine studying balance-related functions in human beings and electroencephalography (EEG), a neurological exploration used to diagnose nervous system disorders. To validate the proposed method, we studied method outlier detection and efficiency in terms of runtime. The results of regression analyses confirm that our proposal is useful for detecting outlier data in different domains, with a false positive rate of less than 2% and a reliability greater than 99%.

---

### Corpus Frequencies in Morphological Inflection: Do They Matter?

Authors: Tomáš Sourada, Jana Straková
Published: 2025-10-27
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.23131v1

The traditional approach to morphological inflection (the task of modifying a base word (lemma) to express grammatical categories) has been, for decades, to consider lexical entries of lemma-tag-form triples uniformly, lacking any information about their frequency distribution. However, in production deployment, one might expect the user inputs to reflect a real-world distribution of frequencies in natural texts. With future deployment in mind, we explore the incorporation of corpus frequency information into the task of morphological inflection along three key dimensions during system development: (i) for train-dev-test split, we combine a lemma-disjoint approach, which evaluates the model's generalization capabilities, with a frequency-weighted strategy to better reflect the realistic distribution of items across different frequency bands in training and test sets; (ii) for evaluation, we complement the standard type accuracy (often referred to simply as accuracy), which treats all items equally regardless of frequency, with token accuracy, which assigns greater weight to frequent words and better approximates performance on running text; (iii) for training data sampling, we introduce a method novel in the context of inflection, frequency-aware training, which explicitly incorporates word frequency into the sampling process. We show that frequency-aware training outperforms uniform sampling in 26 out of 43 languages.

---

### Lost in Tokenization: Context as the Key to Unlocking Biomolecular Understanding in Scientific LLMs

Authors: Kai Zhuang, Jiawei Zhang, Yumou Liu, Hanqun Cao, Chunbin Gu, Mengdi Liu, Zhangyang Gao, Zitong Jerry Wang, Xuanhe Zhou, Pheng-Ann Heng, Lijun Wu, Conghui He et al.
Published: 2025-10-27
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.23127v2

Scientific Large Language Models (Sci-LLMs) have emerged as a promising frontier for accelerating biological discovery. However, these models face a fundamental challenge when processing raw biomolecular sequences: the tokenization dilemma. Whether treating sequences as a specialized language, risking the loss of functional motif information, or as a separate modality, introducing formidable alignment challenges, current strategies fundamentally limit their reasoning capacity. We challenge this sequence-centric paradigm by positing that a more effective strategy is to provide Sci-LLMs with high-level structured context derived from established bioinformatics tools, thereby bypassing the need to interpret low-level noisy sequence data directly. Through a systematic comparison of leading Sci-LLMs on biological reasoning tasks, we tested three input modes: sequence-only, context-only, and a combination of both. Our findings are striking: the context-only approach consistently and substantially outperforms all other modes. Even more revealing, the inclusion of the raw sequence alongside its high-level context consistently degrades performance, indicating that raw sequences act as informational noise, even for models with specialized tokenization schemes. These results suggest that the primary strength of existing Sci-LLMs lies not in their nascent ability to interpret biomolecular syntax from scratch, but in their profound capacity for reasoning over structured, human-readable knowledge. Therefore, we argue for reframing Sci-LLMs not as sequence decoders, but as powerful reasoning engines over expert knowledge. This work lays the foundation for a new class of hybrid scientific AI agents, repositioning the developmental focus from direct sequence interpretation towards high-level knowledge synthesis. The code is available at https://github.com/opendatalab-raiser/CoKE.

---

### DeepSalt: Bridging Laboratory and Satellite Spectra through Domain Adaptation and Knowledge Distillation for Large-Scale Soil Salinity Estimation

Authors: Rupasree Dey, Abdul Matin, Everett Lewark, Tanjim Bin Faruk, Andrei Bachinin, Sam Leuthold, M. Francesca Cotrufo, Shrideep Pallickara, Sangmi Lee Pallickara
Published: 2025-10-27
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.23124v1

Soil salinization poses a significant threat to both ecosystems and agriculture because it limits plants' ability to absorb water and, in doing so, reduces crop productivity. This phenomenon alters the soil's spectral properties, creating a measurable relationship between salinity and light reflectance that enables remote monitoring. While laboratory spectroscopy provides precise measurements, its reliance on in-situ sampling limits scalability to regional or global levels. Conversely, hyperspectral satellite imagery enables wide-area observation but lacks the fine-grained interpretability of laboratory instruments. To bridge this gap, we introduce DeepSalt, a deep-learning-based spectral transfer framework that leverages knowledge distillation and a novel Spectral Adaptation Unit to transfer high-resolution spectral insights from laboratory-based spectroscopy to satellite-based hyperspectral sensing. Our approach eliminates the need for extensive ground sampling while enabling accurate, large-scale salinity estimation, as demonstrated through comprehensive empirical benchmarks. DeepSalt achieves significant performance gains over methods without explicit domain adaptation, underscoring the impact of the proposed Spectral Adaptation Unit and the knowledge distillation strategy. The model also effectively generalized to unseen geographic regions, explaining a substantial portion of the salinity variance.

---

### Beyond Higher Rank: Token-wise Input-Output Projections for Efficient Low-Rank Adaptation

Authors: Shiwei Li, Xiandi Luo, Haozhao Wang, Xing Tang, Ziqiang Cui, Dugang Liu, Yuhua Li, Xiuqiang He, Ruixuan Li
Published: 2025-10-27
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.23123v1

Low-rank adaptation (LoRA) is a parameter-efficient fine-tuning (PEFT) method widely used in large language models (LLMs). LoRA essentially describes the projection of an input space into a low-dimensional output space, with the dimensionality determined by the LoRA rank. In standard LoRA, all input tokens share the same weights and undergo an identical input-output projection. This limits LoRA's ability to capture token-specific information due to the inherent semantic differences among tokens. To address this limitation, we propose Token-wise Projected Low-Rank Adaptation (TopLoRA), which dynamically adjusts LoRA weights according to the input token, thereby learning token-wise input-output projections in an end-to-end manner. Formally, the weights of TopLoRA can be expressed as $BΣ_X A$, where $A$ and $B$ are low-rank matrices (as in standard LoRA), and $Σ_X$ is a diagonal matrix generated from each input token $X$. Notably, TopLoRA does not increase the rank of LoRA weights but achieves more granular adaptation by learning token-wise LoRA weights (i.e., token-wise input-output projections). Extensive experiments across multiple models and datasets demonstrate that TopLoRA consistently outperforms LoRA and its variants. The code is available at https://github.com/Leopold1423/toplora-neurips25.

---

### Quantizing Space and Time: Fusing Time Series and Images for Earth Observation

Authors: Gianfranco Basile, Johannes Jakubik, Benedikt Blumenstiel, Thomas Brunschwiler, Juan Bernabe Moreno
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23118v4

We propose a task-agnostic framework for multimodal fusion of time series and single timestamp images, enabling cross-modal generation and robust downstream performance. Our approach explores deterministic and learned strategies for time series quantization and then leverages a masked correlation learning objective, aligning discrete image and time series tokens in a unified representation space. Instantiated in the Earth observation domain, the pretrained model generates consistent global temperature profiles from satellite imagery and is validated through counterfactual experiments. Across downstream tasks, our task-agnostic pretraining outperforms task-specific fusion by 6% in R^2 and 2% in RMSE on average, and exceeds baseline methods by 50% in R^2 and 12% in RMSE. Finally, we analyze gradient sensitivity across modalities, providing insights into model robustness. Code, data, and weights will be released under a permissive license.

---

### Seeing Structural Failure Before it Happens: An Image-Based Physics-Informed Neural Network (PINN) for Spaghetti Bridge Load Prediction

Authors: Omer Jauhar Khan, Sudais Khan, Hafeez Anwar, Shahzeb Khan, Shams Ul Arifeen, Farman Ullah
Published: 2025-10-27
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2510.23117v4

Physics Informed Neural Networks (PINNs) are gaining attention for their ability to embed physical laws into deep learning models, which is particularly useful in structural engineering tasks with limited data. This paper aims to explore the use of PINNs to predict the weight of small scale spaghetti bridges, a task relevant to understanding load limits and potential failure modes in simplified structural models. Our proposed framework incorporates physics-based constraints to the prediction model for improved performance. In addition to standard PINNs, we introduce a novel architecture named Physics Informed Kolmogorov Arnold Network (PIKAN), which blends universal function approximation theory with physical insights. The structural parameters provided as input to the model are collected either manually or through computer vision methods. Our dataset includes 15 real bridges, augmented to 100 samples, and our best model achieves an $R^2$ score of 0.9603 and a mean absolute error (MAE) of 10.50 units. From applied perspective, we also provide a web based interface for parameter entry and prediction. These results show that PINNs can offer reliable estimates of structural weight, even with limited data, and may help inform early stage failure analysis in lightweight bridge designs. The complete data and code are available at https://github.com/OmerJauhar/PINNS-For-Spaghetti-Bridges.

---

### Residual Diffusion Bridge Model for Image Restoration

Authors: Hebaixu Wang, Jing Zhang, Haoyang Chen, Haonan Guo, Di Wang, Jiayi Ma, Bo Du
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23116v4

Diffusion bridge models establish probabilistic paths between arbitrary paired distributions and exhibit great potential for universal image restoration. Most existing methods merely treat them as simple variants of stochastic interpolants, lacking a unified analytical perspective. Besides, they indiscriminately reconstruct images through global noise injection and removal, inevitably distorting undegraded regions due to imperfect reconstruction. To address these challenges, we propose the Residual Diffusion Bridge Model (RDBM). Specifically, we theoretically reformulate the stochastic differential equations of generalized diffusion bridge and derive the analytical formulas of its forward and reverse processes. Crucially, we leverage the residuals from given distributions to modulate the noise injection and removal, enabling adaptive restoration of degraded regions while preserving intact others. Moreover, we unravel the fundamental mathematical essence of existing bridge models, all of which are special cases of RDBM and empirically demonstrate the optimality of our proposed models. Extensive experiments are conducted to demonstrate the state-of-the-art performance of our method both qualitatively and quantitatively across diverse image restoration tasks. Code is publicly available at https://github.com/MiliLab/RDBM.

---

### Flexing in 73 Languages: A Single Small Model for Multilingual Inflection

Authors: Tomáš Sourada, Jana Straková
Published: 2025-10-27
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.23114v1

We present a compact, single-model approach to multilingual inflection, the task of generating inflected word forms from base lemmas to express grammatical categories. Our model, trained jointly on data from 73 languages, is lightweight, robust to unseen words, and outperforms monolingual baselines in most languages. This demonstrates the effectiveness of multilingual modeling for inflection and highlights its practical benefits: simplifying deployment by eliminating the need to manage and retrain dozens of separate monolingual models. In addition to the standard SIGMORPHON shared task benchmarks, we evaluate our monolingual and multilingual models on 73 Universal Dependencies (UD) treebanks, extracting lemma-tag-form triples and their frequency counts. To ensure realistic data splits, we introduce a novel frequency-weighted, lemma-disjoint train-dev-test resampling procedure. Our work addresses the lack of an open-source, general-purpose, multilingual morphological inflection system capable of handling unseen words across a wide range of languages, including Czech. All code is publicly released at: https://github.com/tomsouri/multilingual-inflection.

---

### GroupSHAP-Guided Integration of Financial News Keywords and Technical Indicators for Stock Price Prediction

Authors: Minjoo Kim, Jinwoong Kim, Sangjin Park
Published: 2025-10-27
Categories: cs.CE, cs.AI
arXiv: https://arxiv.org/abs/2510.23112v3

Recent advances in finance-specific language models such as FinBERT have enabled the quantification of public sentiment into index-based measures, yet compressing diverse linguistic signals into single metrics overlooks contextual nuances and limits interpretability. To address this limitation, explainable AI techniques, particularly SHAP (SHapley Additive Explanations), have been employed to identify influential features. However, SHAP's computational cost grows exponentially with input features, making it impractical for large-scale text-based financial data. This study introduces a GRU-based forecasting framework enhanced with GroupSHAP, which quantifies contributions of semantically related keyword groups rather than individual tokens, substantially reducing computational burden while preserving interpretability. We employed FinBERT to embed news articles from 2015 to 2024, clustered them into coherent semantic groups, and applied GroupSHAP to measure each group's contribution to stock price movements. The resulting group-level SHAP variables across multiple topics were used as input features for the prediction model. Empirical results from one-day-ahead forecasting of the S&P 500 index throughout 2024 demonstrate that our approach achieves a 32.2% reduction in MAE and a 40.5% reduction in RMSE compared with benchmark models without the GroupSHAP mechanism. This research presents the first application of GroupSHAP in news-driven financial forecasting, showing that grouped sentiment representations simultaneously enhance interpretability and predictive performance.

---

### Neural Emulator Superiority: When Machine Learning for PDEs Surpasses its Training Data

Authors: Felix Koehler, Nils Thuerey
Published: 2025-10-27
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.23111v2

Neural operators or emulators for PDEs trained on data from numerical solvers are conventionally assumed to be limited by their training data's fidelity. We challenge this assumption by identifying "emulator superiority," where neural networks trained purely on low-fidelity solver data can achieve higher accuracy than those solvers when evaluated against a higher-fidelity reference. Our theoretical analysis reveals how the interplay between emulator inductive biases, training objectives, and numerical error characteristics enables superior performance during multi-step rollouts. We empirically validate this finding across different PDEs using standard neural architectures, demonstrating that emulators can implicitly learn dynamics that are more regularized or exhibit more favorable error accumulation properties than their training data, potentially surpassing training data limitations and mitigating numerical artifacts. This work prompts a re-evaluation of emulator benchmarking, suggesting neural emulators might achieve greater physical fidelity than their training source within specific operational regimes. Project Page: https://tum-pbs.github.io/emulator-superiority

---

### Sampling from Energy distributions with Target Concrete Score Identity

Authors: Sergei Kholkin, Francisco Vargas, Alexander Korotin
Published: 2025-10-27
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.23106v1

We introduce the Target Concrete Score Identity Sampler (TCSIS), a method for sampling from unnormalized densities on discrete state spaces by learning the reverse dynamics of a Continuous-Time Markov Chain (CTMC). Our approach builds on a forward in time CTMC with a uniform noising kernel and relies on the proposed Target Concrete Score Identity, which relates the concrete score, the ratio of marginal probabilities of two states, to a ratio of expectations of Boltzmann factors under the forward uniform diffusion kernel. This formulation enables Monte Carlo estimation of the concrete score without requiring samples from the target distribution or computation of the partition function. We approximate the concrete score with a neural network and propose two algorithms: Self-Normalized TCSIS and Unbiased TCSIS. Finally, we demonstrate the effectiveness of TCSIS on problems from statistical physics.

---

### Leveraging Hierarchical Organization for Medical Multi-document Summarization

Authors: Yi-Li Hsu, Katelyn X. Mei, Lucy Lu Wang
Published: 2025-10-27
Categories: cs.CL, cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2510.23104v2

Medical multi-document summarization (MDS) is a complex task that requires effectively managing cross-document relationships. This paper investigates whether incorporating hierarchical structures in the inputs of MDS can improve a model's ability to organize and contextualize information across documents compared to traditional flat summarization methods. We investigate two ways of incorporating hierarchical organization across three large language models (LLMs), and conduct comprehensive evaluations of the resulting summaries using automated metrics, model-based metrics, and domain expert evaluation of preference, understandability, clarity, complexity, relevance, coverage, factuality, and coherence. Our results show that human experts prefer model-generated summaries over human-written summaries. Hierarchical approaches generally preserve factuality, coverage, and coherence of information, while also increasing human preference for summaries. Additionally, we examine whether simulated judgments from GPT-4 align with human judgments, finding higher agreement along more objective evaluation facets. Our findings demonstrate that hierarchical structures can improve the clarity of medical summaries generated by models while maintaining content coverage, providing a practical way to improve human preference for generated summaries.

---

### Revisiting Multimodal Positional Encoding in Vision-Language Models

Authors: Jie Huang, Xuejing Liu, Sibo Song, Ruibing Hou, Hong Chang, Junyang Lin, Shuai Bai
Published: 2025-10-27
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.23095v3

Multimodal position encoding is essential for vision-language models, yet there has been little systematic investigation into multimodal position encoding. We conduct a comprehensive analysis of multimodal Rotary Positional Embedding (RoPE) by examining its two core components: position design and frequency allocation. Through extensive experiments, we identify three key guidelines: positional coherence, full frequency utilization, and preservation of textual priors-ensuring unambiguous layout, rich representation, and faithful transfer from the pre-trained LLM. Based on these insights, we propose Multi-Head RoPE (MHRoPE) and MRoPE-Interleave (MRoPE-I), two simple and plug-and-play variants that require no architectural changes. Our methods consistently outperform existing approaches across diverse benchmarks, with significant improvements in both general and fine-grained multimodal understanding. Code will be avaliable at https://github.com/JJJYmmm/Multimodal-RoPEs.

---