# Research Papers — 2025-10 · bundle page 196 of 217

> Full-text markdown of papers #9751–9800 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/196
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/195.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/197.md)

### PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning

Authors: Wanjia Zhao, Qinwei Ma, Jingzhe Shi, Shirley Wu, Jiaqi Han, Yijia Xiao, Si-Yuan Chen, Xiao Luo, Ludwig Schmidt, James Zou
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03185v2

Benchmarks for competition-style reasoning have advanced evaluation in mathematics and programming, yet physics remains comparatively explored. Most existing physics benchmarks evaluate only final answers, which fail to capture reasoning processes, while recent stepwise methods rely on heuristic LLM-as-judge scoring or restrictive linear assumptions, limiting reliability and diagnostic validity. We introduce PRISM-Physics, a process-level evaluation framework and benchmark for complex physics reasoning problems. Solutions are represented as directed acyclic graphs (DAGs) of formulas, explicitly encoding causal dependencies among intermediate steps to enable fine-grained, interpretable, and theoretically grounded scoring. We prove the optimality of the DAG representation and the corresponding scoring policy. Combining with a fully rule-based method for symbolic formula equivalence matching that we developed, we ensure consistent validation across diverse formulations without heuristic judgments. Results show that our evaluation framework is more aligned with human experts' scoring. Experiments on state-of-the-art LLMs reveal persistent reasoning failures in physics, while step-level scoring offers both diagnostic insight and rich signals for later training. By combining structural rigor, theoretical guarantees, and symbolic validation, PRISM-Physics provides a principled foundation for advancing process-level evaluation and guiding the development of models with deeper scientific reasoning capabilities.

---

### Simulation to Rules: A Dual-VLM Framework for Formal Visual Planning

Authors: Yilun Hao, Yongchao Chen, Chuchu Fan, Yang Zhang
Published: 2025-10-03
Categories: cs.RO, cs.AI, cs.CL, cs.SC
arXiv: https://arxiv.org/abs/2510.03182v2

Vision Language Models (VLMs) show strong potential for visual planning but struggle with precise spatial and long-horizon reasoning, while Planning Domain Definition Language (PDDL) planners excel at formal long-horizon planning but cannot interpret visual inputs. Recent works combine these complementary advantages by translating visual problems into PDDL. However, while VLMs can generate PDDL problem files satisfactorily, accurately generating PDDL domain files, which encode planning rules, remains challenging and typically requires human expertise or environment interaction. We propose VLMFP, a Dual-VLM-guided framework that autonomously generates both PDDL problem and domain files for formal visual planning. VLMFP combines a SimVLM that simulates action consequences with a GenVLM that generates and iteratively refines PDDL files by aligning symbolic execution with simulated outcomes, enabling multiple levels of generalization across unseen instances, visual appearances, and game rules. We evaluate VLMFP on 6 grid-world domains and demonstrate its generalization capability. On average, SimVLM achieves 87.3% and 86.0% scenario understanding and action simulation for seen and unseen appearances, respectively. With the guidance of SimVLM, VLMFP attains 70.0%, 54.1% planning success on unseen instances in seen and unseen appearances, respectively. We further demonstrate that VLMFP scales to complex long-horizon 3D planning tasks, including multi-robot collaboration and assembly scenarios with partial observability and diverse visual variations. Project page: https://sites.google.com/view/vlmfp.

---

### Q-Learning with Shift-Aware Upper Confidence Bound in Non-Stationary Reinforcement Learning

Authors: Ha Manh Bui, Felix Parker, Kimia Ghobadi, Anqi Liu
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03181v2

We study the Non-Stationary Reinforcement Learning (RL) under distribution shifts in both finite-horizon episodic and infinite-horizon discounted Markov Decision Processes (MDPs). In the finite-horizon case, the transition functions may suddenly change at a particular episode. In the infinite-horizon setting, such changes can occur at an arbitrary time step during the agent's interaction with the environment. While the Q-learning Upper Confidence Bound algorithm (QUCB) can discover a proper policy during learning, due to the distribution shifts, this policy can exploit sub-optimal rewards after the shift happens. To address this issue, we propose Density-QUCB (DQUCB), a shift-aware Q-learning UCB algorithm, which uses a transition density function to detect distribution shifts, then leverages its likelihood to enhance the uncertainty estimation quality of Q-learning UCB, resulting in a balance between exploration and exploitation. Theoretically, we prove that our oracle DQUCB achieves a better regret guarantee than QUCB. Empirically, our DQUCB enjoys the computational efficiency of model-free RL and outperforms QUCB baselines by having a lower regret across RL tasks, as well as a COVID-19 patient hospital allocation task using a Deep-Q-learning architecture.

---

### When Names Disappear: Revealing What LLMs Actually Understand About Code

Authors: Cuong Chi Le, Minh V. T. Pham, Cuong Duc Van, Hoang N. Phan, Huy N. Phan, Tien N. Nguyen
Published: 2025-10-03
Categories: cs.SE, cs.CL
arXiv: https://arxiv.org/abs/2510.03178v1

Large Language Models (LLMs) achieve strong results on code tasks, but how they derive program meaning remains unclear. We argue that code communicates through two channels: structural semantics, which define formal behavior, and human-interpretable naming, which conveys intent. Removing the naming channel severely degrades intent-level tasks such as summarization, where models regress to line-by-line descriptions. Surprisingly, we also observe consistent reductions on execution tasks that should depend only on structure, revealing that current benchmarks reward memorization of naming patterns rather than genuine semantic reasoning. To disentangle these effects, we introduce a suite of semantics-preserving obfuscations and show that they expose identifier leakage across both summarization and execution. Building on these insights, we release ClassEval-Obf, an obfuscation-enhanced benchmark that systematically suppresses naming cues while preserving behavior. Our results demonstrate that ClassEval-Obf reduces inflated performance gaps, weakens memorization shortcuts, and provides a more reliable basis for assessing LLMs' code understanding and generalization.

---

### LLM as Attention-Informed NTM and Topic Modeling as long-input Generation: Interpretability and long-Context Capability

Authors: Xuan Xu, Zhongliang Yang, Haolun Li, Beilin Chu, Rui Tian, Yu Li, Shaolin Tan, Linna Zhou
Published: 2025-10-03
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.03174v2

Topic modeling aims to produce interpretable topic representations and topic--document correspondences from corpora, but classical neural topic models (NTMs) remain constrained by limited representation assumptions and semantic abstraction ability. We study LLM-based topic modeling from both white-box and black-box perspectives. For white-box LLMs, we propose an attention-informed framework that recovers interpretable structures analogous to those in NTMs, including document-topic and topic-word distributions. This validates the view that LLM can serve as an attention-informed NTM. For black-box LLMs, we reformulate topic modeling as a structured long-input task and introduce a post-generation signal compensation method based on diversified topic cues and hybrid retrieval. Experiments show that recovered attention structures support effective topic assignment and keyword extraction, while black-box long-context LLMs achieve competitive or stronger performance than other baselines. These findings suggest a connection between LLMs and NTMs and highlight the promise of long-context LLMs for topic modeling.

---

### Improving Online-to-Nonconvex Conversion for Smooth Optimization via Double Optimism

Authors: Francisco Patitucci, Ruichen Jiang, Aryan Mokhtari
Published: 2025-10-03
Categories: math.OC, cs.LG
arXiv: https://arxiv.org/abs/2510.03167v3

A recent breakthrough in nonconvex optimization is the online-to-nonconvex conversion framework of (Cutkosky et al., 2023), which reformulates the task of finding an $\varepsilon$-first-order stationary point as an online learning problem. When both the gradient and the Hessian are Lipschitz continuous, instantiating this framework with two different online learners achieves a complexity of $O(\varepsilon^{-1.75}\log(1/\varepsilon))$ in the deterministic case and a complexity of $O(\varepsilon^{-3.5})$ in the stochastic case. However, this approach suffers from several limitations: (i) the deterministic method relies on a complex double-loop scheme that solves a fixed-point equation to construct hint vectors for an optimistic online learner, introducing an extra logarithmic factor; (ii) the stochastic method assumes a bounded second-order moment of the stochastic gradient, which is stronger than standard variance bounds; and (iii) different online learning algorithms are used in the two settings. In this paper, we address these issues by introducing an online optimistic gradient method based on a novel doubly optimistic hint function. Specifically, we use the gradient at an extrapolated point as the hint, motivated by two optimistic assumptions: that the difference between the hint and the target gradient remains near constant, and that consecutive update directions change slowly due to smoothness. Our method eliminates the need for a double loop and removes the logarithmic factor. Furthermore, by simply replacing full gradients with stochastic gradients and under the standard assumption that their variance is bounded by $σ^2$, we obtain a unified algorithm with complexity $O(\varepsilon^{-1.75} + σ^2 \varepsilon^{-3.5})$, smoothly interpolating between the best-known deterministic rate and the optimal stochastic rate.

---

### FTTE: Enabling Federated and Resource-Constrained Deep Edge Intelligence

Authors: Irene Tenison, Anna Murphy, Charles Beauville, Lalana Kagal
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03165v2

Federated learning (FL) enables collaborative model training across distributed devices while preserving data privacy, but deployment on resource-constrained edge nodes remains challenging due to limited memory, energy, and communication bandwidth. Traditional synchronous and asynchronous FL approaches further suffer from straggler induced delays and slow convergence in heterogeneous, large scale networks. We present FTTE (Federated Tiny Training Engine),a novel semi-asynchronous FL framework that uniquely employs sparse parameter updates and a staleness-weighted aggregation based on both age and variance of client updates. Extensive experiments across diverse models and data distributions - including up to 500 clients and 90% stragglers - demonstrate that FTTE not only achieves 81% faster convergence, 80% lower on-device memory usage, and 69% communication payload reduction than synchronous FL (eg.FedAVG), but also consistently reaches comparable or higher target accuracy than semi-asynchronous (eg.FedBuff) in challenging regimes. These results establish FTTE as the first practical and scalable solution for real-world FL deployments on heterogeneous and predominantly resource-constrained edge devices.

---

### Why Do We Need Warm-up? A Theoretical Perspective

Authors: Foivos Alimisis, Rustem Islamov, Aurelien Lucchi
Published: 2025-10-03
Categories: cs.LG, math.OC, stat.ML
arXiv: https://arxiv.org/abs/2510.03164v2

Learning rate warm-up -- increasing the learning rate at the beginning of training -- has become a ubiquitous heuristic in modern deep learning, yet its theoretical foundations remain poorly understood. In this work, we provide a principled explanation for why warm-up improves training. We rely on a generalization of the $(L_0, L_1)$-smoothness condition, which bounds local curvature as a linear function of the loss suboptimality and exhibits desirable closure properties. We show -- both theoretically and empirically -- that this condition is satisfied by common neural architectures and accurately captures the curvature of the optimization landscape early in training. Adapting the learning rate in response to this curvature condition naturally induces a warm-up-like schedule, and we show that this choice yields provably faster convergence guarantees than using a fixed learning rate. Experiments on language and vision models show that the resulting one-parameter warm-up schedule can match tuned linear warm-up and improve over no warm-up.

---

### ROGR: Relightable 3D Objects using Generative Relighting

Authors: Jiapeng Tang, Matthew Levine, Dor Verbin, Stephan J. Garbin, Matthias Nießner, Ricardo Martin Brualla, Pratul P. Srinivasan, Philipp Henzler
Published: 2025-10-03
Categories: cs.CV, cs.GR
arXiv: https://arxiv.org/abs/2510.03163v3

We introduce ROGR, a novel approach that reconstructs a relightable 3D model of an object captured from multiple views, driven by a generative relighting model that simulates the effects of placing the object under novel environment illuminations. Our method samples the appearance of the object under multiple lighting environments, creating a dataset that is used to train a lighting-conditioned Neural Radiance Field (NeRF) that outputs the object's appearance under any input environmental lighting. The lighting-conditioned NeRF uses a novel dual-branch architecture to encode the general lighting effects and specularities separately. The optimized lighting-conditioned NeRF enables efficient feed-forward relighting under arbitrary environment maps without requiring per-illumination optimization or light transport simulation. We evaluate our approach on the established TensoIR and Stanford-ORB datasets, where it improves upon the state-of-the-art on most metrics, and showcase our approach on real-world object captures.

---

### Calibrated Uncertainty Sampling for Active Learning

Authors: Ha Manh Bui, Iliana Maifeld-Carucci, Anqi Liu
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03162v1

We study the problem of actively learning a classifier with a low calibration error. One of the most popular Acquisition Functions (AFs) in pool-based Active Learning (AL) is querying by the model's uncertainty. However, we recognize that an uncalibrated uncertainty model on the unlabeled pool may significantly affect the AF effectiveness, leading to sub-optimal generalization and high calibration error on unseen data. Deep Neural Networks (DNNs) make it even worse as the model uncertainty from DNN is usually uncalibrated. Therefore, we propose a new AF by estimating calibration errors and query samples with the highest calibration error before leveraging DNN uncertainty. Specifically, we utilize a kernel calibration error estimator under the covariate shift and formally show that AL with this AF eventually leads to a bounded calibration error on the unlabeled pool and unseen test data. Empirically, our proposed method surpasses other AF baselines by having a lower calibration and generalization error across pool-based AL settings.

---

### UniShield: An Adaptive Multi-Agent Framework for Unified Forgery Image Detection and Localization

Authors: Qing Huang, Zhipei Xu, Xuanyu Zhang, Xiangyu Yu, Jian Zhang
Published: 2025-10-03
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.03161v4

With the rapid advancements in image generation, synthetic images have become increasingly realistic, posing significant societal risks, such as misinformation and fraud. Forgery Image Detection and Localization (FIDL) thus emerges as essential for maintaining information integrity and societal security. Despite impressive performances by existing domain-specific detection methods, their practical applicability remains limited, primarily due to their narrow specialization, poor cross-domain generalization, and the absence of an integrated adaptive framework. To address these issues, we propose UniShield, the novel multi-agent-based unified system capable of detecting and localizing image forgeries across diverse domains, including image manipulation, document manipulation, DeepFake, and AI-generated images. UniShield innovatively integrates a perception agent with a detection agent. The perception agent intelligently analyzes image features to dynamically select suitable detection models, while the detection agent consolidates various expert detectors into a unified framework and generates interpretable reports. Extensive experiments show that UniShield achieves state-of-the-art results, surpassing both existing unified approaches and domain-specific detectors, highlighting its superior practicality, adaptiveness, and scalability.

---

### SpineBench: A Clinically Salient, Level-Aware Benchmark Powered by the SpineMed-450k Corpus

Authors: Ming Zhao, Wenhui Dong, Yang Zhang, Xiang Zheng, Zhonghao Zhang, Zian Zhou, Yunzhi Guan, Liukun Xu, Wei Peng, Zhaoyang Gong, Zhicheng Zhang, Dachuan Li et al.
Published: 2025-10-03
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.03160v3

Spine disorders affect 619 million people globally and are a leading cause of disability, yet AI-assisted diagnosis remains limited by the lack of level-aware, multimodal datasets. Clinical decision-making for spine disorders requires sophisticated reasoning across X-ray, CT, and MRI at specific vertebral levels. However, progress has been constrained by the absence of traceable, clinically-grounded instruction data and standardized, spine-specific benchmarks. To address this, we introduce SpineMed, an ecosystem co-designed with practicing spine surgeons. It features SpineMed-450k, the first large-scale dataset explicitly designed for vertebral-level reasoning across imaging modalities with over 450,000 instruction instances, and SpineBench, a clinically-grounded evaluation framework. SpineMed-450k is curated from diverse sources, including textbooks, guidelines, open datasets, and ~1,000 de-identified hospital cases, using a clinician-in-the-loop pipeline with a two-stage LLM generation method (draft and revision) to ensure high-quality, traceable data for question-answering, multi-turn consultations, and report generation. SpineBench evaluates models on clinically salient axes, including level identification, pathology assessment, and surgical planning. Our comprehensive evaluation of several recently advanced large vision-language models (LVLMs) on SpineBench reveals systematic weaknesses in fine-grained, level-specific reasoning. In contrast, our model fine-tuned on SpineMed-450k demonstrates consistent and significant improvements across all tasks. Clinician assessments confirm the diagnostic clarity and practical utility of our model's outputs.

---

### Neural Correlates of Language Models Are Specific to Human Language

Authors: Iñigo Parra
Published: 2025-10-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.03156v2

Previous work has shown correlations between the hidden states of large language models and fMRI brain responses, on language tasks. These correlations have been taken as evidence of the representational similarity of these models and brain states. This study tests whether these previous results are robust to several possible concerns. Specifically this study shows: (i) that the previous results are still found after dimensionality reduction, and thus are not attributable to the curse of dimensionality; (ii) that previous results are confirmed when using new measures of similarity; (iii) that correlations between brain representations and those from models are specific to models trained on human language; and (iv) that the results are dependent on the presence of positional encoding in the models. These results confirm and strengthen the results of previous research and contribute to the debate on the biological plausibility and interpretability of state-of-the-art large language models.

---

### Stimulus-Voltage-Based Prediction of Action Potential Onset Timing: Classical vs. Quantum-Inspired Approaches

Authors: Stevens Johnson, Varun Puram, Johnson Thomas, Acsah Konuparamban, Ashwin Kannan
Published: 2025-10-03
Categories: q-bio.NC, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.03155v1

Accurate modeling of neuronal action potential (AP) onset timing is crucial for understanding neural coding of danger signals. Traditional leaky integrate-and-fire (LIF) models, while widely used, exhibit high relative error in predicting AP onset latency, especially under strong or rapidly changing stimuli. Inspired by recent experimental findings and quantum theory, we present a quantum-inspired leaky integrate-and-fire (QI-LIF) model that treats AP onset as a probabilistic event, represented by a Gaussian wave packet in time. This approach captures the biological variability and uncertainty inherent in neuronal firing. We systematically compare the relative error of AP onset predictions between the classical LIF and QI-LIF models using synthetic data from hippocampal and sensory neurons subjected to varying stimulus amplitudes. Our results demonstrate that the QI-LIF model significantly reduces prediction error, particularly for high-intensity stimuli, aligning closely with observed biological responses. This work highlights the potential of quantum-inspired computational frameworks in advancing the accuracy of neural modeling and has implications for quantum engineering approaches to brain-inspired computing.

---

### EditLens: Quantifying the Extent of AI Editing in Text

Authors: Katherine Thai, Bradley Emi, Elyas Masrour, Mohit Iyyer
Published: 2025-10-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.03154v1

A significant proportion of queries to large language models ask them to edit user-provided text, rather than generate new text from scratch. While previous work focuses on detecting fully AI-generated text, we demonstrate that AI-edited text is distinguishable from human-written and AI-generated text. First, we propose using lightweight similarity metrics to quantify the magnitude of AI editing present in a text given the original human-written text and validate these metrics with human annotators. Using these similarity metrics as intermediate supervision, we then train EditLens, a regression model that predicts the amount of AI editing present within a text. Our model achieves state-of-the-art performance on both binary (F1=94.7%) and ternary (F1=90.4%) classification tasks in distinguishing human, AI, and mixed writing. Not only do we show that AI-edited text can be detected, but also that the degree of change made by AI to human writing can be detected, which has implications for authorship attribution, education, and policy. Finally, as a case study, we use our model to analyze the effects of AI-edits applied by Grammarly, a popular writing assistance tool. To encourage further research, we commit to publicly releasing our models and dataset.

---

### Improving Cooperation in Collaborative Embodied AI

Authors: Hima Jacob Leven Suprabha, Laxmi Nag Laxminarayan Nagesh, Ajith Nair, Alvin Reuben Amal Selvaster, Ayan Khan, Raghuram Damarla, Sanju Hannah Samuel, Sreenithi Saravana Perumal, Titouan Puech, Venkataramireddy Marella, Vishal Sonar, Alessandro Suglia et al.
Published: 2025-10-03
Categories: cs.AI, cs.MA, cs.RO
arXiv: https://arxiv.org/abs/2510.03153v1

The integration of Large Language Models (LLMs) into multiagent systems has opened new possibilities for collaborative reasoning and cooperation with AI agents. This paper explores different prompting methods and evaluates their effectiveness in enhancing agent collaborative behaviour and decision-making. We enhance CoELA, a framework designed for building Collaborative Embodied Agents that leverage LLMs for multi-agent communication, reasoning, and task coordination in shared virtual spaces. Through systematic experimentation, we examine different LLMs and prompt engineering strategies to identify optimised combinations that maximise collaboration performance. Furthermore, we extend our research by integrating speech capabilities, enabling seamless collaborative voice-based interactions. Our findings highlight the effectiveness of prompt optimisation in enhancing collaborative agent performance; for example, our best combination improved the efficiency of the system running with Gemma3 by 22% compared to the original CoELA system. In addition, the speech integration provides a more engaging user interface for iterative system development and demonstrations.

---

### Markovian Reeb Graphs for Simulating Spatiotemporal Patterns of Life

Authors: Anantajit Subrahmanya, Chandrakanth Gudavalli, Connor Levenson, B. S. Manjunath
Published: 2025-10-03
Categories: cs.CV, cs.CE, cs.LG, cs.SI
arXiv: https://arxiv.org/abs/2510.03152v3

Accurately modeling human mobility is critical for urban planning, epidemiology, and traffic management. In this work, we introduce Markovian Reeb Graphs, a novel framework that transforms Reeb graphs from a descriptive analysis tool into a generative model for spatiotemporal trajectories. Our approach captures individual and population-level Patterns of Life (PoLs) and generates realistic trajectories that preserve baseline behaviors while incorporating stochastic variability by embedding probabilistic transitions within the Reeb graph structure. We present two variants: Sequential Reeb Graphs (SRGs) for individual agents and Hybrid Reeb Graphs (HRGs) that combine individual with population PoLs, evaluated on the Urban Anomalies and Geolife datasets using five mobility statistics. Results demonstrate that HRGs achieve strong fidelity across metrics while requiring modest trajectory datasets without specialized side information. This work establishes Markovian Reeb Graphs as a promising framework for trajectory simulation with broad applicability across urban environments.

---

### Mixture of Many Zero-Compute Experts: A High-Rate Quantization Theory Perspective

Authors: Yehuda Dar
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03151v2

This paper uses classical high-rate quantization theory to provide new insights into mixture-of-experts (MoE) models for regression tasks. Our MoE is defined by a segmentation of the input space to regions, each with a single-parameter expert that acts as a constant predictor with zero-compute at inference. Motivated by high-rate quantization theory assumptions, we assume that the number of experts is sufficiently large to make their input-space regions very small. This lets us study the approximation error of our MoE model class: (i) for one-dimensional inputs, we formulate the test error and its minimizing segmentation and experts; (ii) for multidimensional inputs, we formulate an upper bound for the test error and study its minimization. Moreover, we consider the learning of the expert parameters from a training dataset, given an input-space segmentation, and formulate their statistical learning properties. This leads us to theoretically and empirically show how the tradeoff between approximation and estimation errors in MoE learning depends on the number of experts.

---

### Taming Imperfect Process Verifiers: A Sampling Perspective on Backtracking

Authors: Dhruv Rohatgi, Abhishek Shetty, Donya Saless, Yuchen Li, Ankur Moitra, Andrej Risteski, Dylan J. Foster
Published: 2025-10-03
Categories: cs.LG, cs.DS
arXiv: https://arxiv.org/abs/2510.03149v1

Test-time algorithms that combine the generative power of language models with process verifiers that assess the quality of partial generations offer a promising lever for eliciting new reasoning capabilities, but the algorithmic design space and computational scaling properties of such approaches are still opaque, and their benefits are far from apparent when one accounts for the cost of learning a high-quality verifier. Our starting point is the observation that seemingly benign errors in a learned verifier can lead to catastrophic failures for standard decoding techniques due to error amplification during the course of generation. We then ask: can this be improved with more sophisticated decoding strategies? We introduce a new process-guided test-time sampling algorithm, VGB, which uses theoretically grounded backtracking to achieve provably better robustness to verifier errors. VGB interprets autoregressive generation as a random walk on a tree of partial generations, with transition probabilities guided by the process verifier and base model; crucially, backtracking occurs probabilistically. This process generalizes the seminal Sinclair-Jerrum random walk (Sinclair & Jerrum, 1989) from the literature on approximate counting and sampling in theoretical computer science, and a conceptual contribution of our work is to highlight parallels with this literature. Empirically, we demonstrate on both synthetic and real language modeling tasks that VGB outperforms baselines on a variety of metrics.

---

### The Computational Complexity of Almost Stable Clustering with Penalties

Authors: Kamyar Khodamoradi, Farnam Mansouri, Sandra Zilles
Published: 2025-10-03
Categories: cs.CC, cs.DS, cs.LG
arXiv: https://arxiv.org/abs/2510.03143v1

We investigate the complexity of stable (or perturbation-resilient) instances of $\mathrm{k-M\small{EANS}}$ and $\mathrm{k-M\small{EDIAN}}$ clustering problems in metrics with small doubling dimension. While these problems have been extensively studied under multiplicative perturbation resilience in low-dimensional Euclidean spaces (e.g., (Friggstad et al., 2019; Cohen-Addad and Schwiegelshohn, 2017)), we adopt a more general notion of stability, termed ``almost stable'', which is closer to the notion of $(α, \varepsilon)$-perturbation resilience introduced by Balcan and Liang (2016). Additionally, we extend our results to $\mathrm{k-M\small{EANS}}$/$\mathrm{k-M\small{EDIAN}}$ with penalties, where each data point is either assigned to a cluster centre or incurs a penalty. We show that certain special cases of almost stable $\mathrm{k-M\small{EANS}}$/$\mathrm{k-M\small{EDIAN}}$ (with penalties) are solvable in polynomial time. To complement this, we also examine the hardness of almost stable instances and $(1 + \frac{1}{poly(n)})$-stable instances of $\mathrm{k-M\small{EANS}}$/$\mathrm{k-M\small{EDIAN}}$ (with penalties), proving super-polynomial lower bounds on the runtime of any exact algorithm under the widely believed Exponential Time Hypothesis (ETH).

---

### MM-Nav: Multi-View VLA Model for Robust Visual Navigation via Multi-Expert Learning

Authors: Tianyu Xu, Jiawei Chen, Jiazhao Zhang, Wenyao Zhang, Zekun Qi, Minghan Li, Zhizheng Zhang, He Wang
Published: 2025-10-03
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2510.03142v2

Visual navigation policy is widely regarded as a promising direction, as it mimics humans by using egocentric visual observations for navigation. However, optical information of visual observations is difficult to be explicitly modeled like LiDAR point clouds or depth maps, which subsequently requires intelligent models and large-scale data. To this end, we propose to leverage the intelligence of the Vision-Language-Action (VLA) model to learn diverse navigation capabilities from synthetic expert data in a teacher-student manner. Specifically, we implement the VLA model, MM-Nav, as a multi-view VLA (with 360 observations) based on pretrained large language models and visual foundation models. For large-scale navigation data, we collect expert data from three reinforcement learning (RL) experts trained with privileged depth information in three challenging tailor-made environments for different navigation capabilities: reaching, squeezing, and avoiding. We iteratively train our VLA model using data collected online from RL experts, where the training ratio is dynamically balanced based on performance on individual capabilities. Through extensive experiments in synthetic environments, we demonstrate that our model achieves strong generalization capability. Moreover, we find that our student VLA model outperforms the RL teachers, demonstrating the synergistic effect of integrating multiple capabilities. Extensive real-world experiments further confirm the effectiveness of our method.

---

### Beyond the Final Layer: Intermediate Representations for Better Multilingual Calibration in Large Language Models

Authors: Ej Zhou, Caiqi Zhang, Tiancheng Hu, Chengzu Li, Nigel Collier, Ivan Vulić, Anna Korhonen
Published: 2025-10-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.03136v2

Confidence calibration, the alignment of a model's predicted confidence with its actual accuracy, is crucial for the reliable deployment of Large Language Models (LLMs). However, this critical property remains largely under-explored in multilingual contexts. In this work, we conduct the first large-scale, systematic studies of multilingual calibration across six model families and over 100 languages, revealing that non-English languages suffer from systematically worse calibration. To diagnose this, we investigate the model's internal representations and find that the final layer, biased by English-centric training, provides a poor signal for multilingual confidence. In contrast, our layer-wise analysis uncovers a key insight that late-intermediate layers consistently offer a more reliable and better-calibrated signal. Building on this, we introduce a suite of training-free methods, including Language-Aware Confidence Ensemble (LACE), which adaptively selects an optimal ensemble of layers for each specific language. Our study highlights the hidden costs of English-centric alignment and offer a new path toward building more globally equitable and trustworthy LLMs by looking beyond the final layer.

---

### Mask2IV: Interaction-Centric Video Generation via Mask Trajectories

Authors: Gen Li, Bo Zhao, Jianfei Yang, Laura Sevilla-Lara
Published: 2025-10-03
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2510.03135v2

Generating interaction-centric videos, such as those depicting humans or robots interacting with objects, is crucial for embodied intelligence, as they provide rich and diverse visual priors for robot learning, manipulation policy training, and affordance reasoning. However, existing methods often struggle to model such complex and dynamic interactions. While recent studies show that masks can serve as effective control signals and enhance generation quality, obtaining dense and precise mask annotations remains a major challenge for real-world use. To overcome this limitation, we introduce Mask2IV, a novel framework specifically designed for interaction-centric video generation. It adopts a decoupled two-stage pipeline that first predicts plausible motion trajectories for both actor and object, then generates a video conditioned on these trajectories. This design eliminates the need for dense mask inputs from users while preserving the flexibility to manipulate the interaction process. Furthermore, Mask2IV supports versatile and intuitive control, allowing users to specify the target object of interaction and guide the motion trajectory through action descriptions or spatial position cues. To support systematic training and evaluation, we curate two benchmarks covering diverse action and object categories across both human-object interaction and robotic manipulation scenarios. Extensive experiments demonstrate that our method achieves superior visual realism and controllability compared to existing baselines.

---

### Enhancing XAI Narratives through Multi-Narrative Refinement and Knowledge Distillation

Authors: Flavio Giorgi, Matteo Silvestri, Cesare Campagnano, Fabrizio Silvestri, Gabriele Tolomei
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03134v2

Explainable Artificial Intelligence has become a crucial area of research, aiming to demystify the decision-making processes of deep learning models. Among various explainability techniques, counterfactual explanations have been proven particularly promising, as they offer insights into model behavior by highlighting minimal changes that would alter a prediction. Despite their potential, these explanations are often complex and technical, making them difficult for non-experts to interpret. To address this challenge, we propose a novel pipeline that leverages Language Models, large and small, to compose narratives for counterfactual explanations. We employ knowledge distillation techniques along with a refining mechanism to enable Small Language Models to perform comparably to their larger counterparts while maintaining robust reasoning abilities. In addition, we introduce a simple but effective evaluation method to assess natural language narratives, designed to verify whether the models' responses are in line with the factual, counterfactual ground truth. As a result, our proposed pipeline enhances both the reasoning capabilities and practical performance of student models, making them more suitable for real-world use cases.

---

### Total robustness in Bayesian Nonlinear Regression

Authors: Mengqi Chen, Charita Dellaporta, Thomas B. Berrett, Theodoros Damoulas
Published: 2025-10-03
Categories: stat.ME, stat.ML
arXiv: https://arxiv.org/abs/2510.03131v3

Modern regression analyses are often undermined by covariate measurement error, misspecification of the regression model, and misspecification of the measurement error distribution. We present, to the best of our knowledge, the first Bayesian nonparametric learning framework targeting total robustness to all three challenges in general nonlinear regression. Our framework places a joint Dirichlet process prior on the latent covariate--response distribution and updates it with posterior pseudo-samples of the latent covariates, so that inference is calibrated to the joint law. This yields estimators defined by minimizing the discrepancy between posterior realizations of the joint Dirichlet process and the model-implied joint distribution. We establish generalization bounds and provide a first proof of convergence and consistency of the resulting estimators under non-degenerate measurement error. A gradient-based implementation enables efficient computation; simulations and two real-data studies show improved stability to misspecification under increasing measurement error relative to recent Bayesian and frequentist alternatives.

---

### Signature-Informed Transformer for Asset Allocation

Authors: Yoontae Hwang, Stefan Zohren
Published: 2025-10-03
Categories: cs.LG, cs.AI, q-fin.PM
arXiv: https://arxiv.org/abs/2510.03129v3

Modern deep learning for asset allocation typically separates forecasting from optimization. We argue this creates a fundamental mismatch where minimizing prediction errors fails to yield robust portfolios. We propose the Signature Informed Transformer to address this by unifying feature extraction and decision making into a single policy. Our model employs path signatures to encode complex path dependencies and introduces a specialized attention mechanism that targets geometric asset relationships. By directly minimizing the Conditional Value at Risk we ensure the training objective aligns with financial goals. We prove that our attention module rigorously amplifies signature derived signals. Experiments across diverse equity universes show our approach significantly outperforms both traditional strategies and advanced forecasting baselines. The code is available at: https://anonymous.4open.science/r/Signature-Informed-Transformer-For-Asset-Allocation-DB88

---

### A Study of Rule Omission in Raven's Progressive Matrices

Authors: Binze Li
Published: 2025-10-03
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.03127v1

Analogical reasoning lies at the core of human cognition and remains a fundamental challenge for artificial intelligence. Raven's Progressive Matrices (RPM) serve as a widely used benchmark to assess abstract reasoning by requiring the inference of underlying structural rules. While many vision-based and language-based models have achieved success on RPM tasks, it remains unclear whether their performance reflects genuine reasoning ability or reliance on statistical shortcuts. This study investigates the generalization capacity of modern AI systems under conditions of incomplete training by deliberately omitting several structural rules during training. Both sequence-to-sequence transformer models and vision-based architectures such as CoPINet and the Dual-Contrast Network are evaluated on the Impartial-RAVEN (I-RAVEN) dataset. Experiments reveal that although transformers demonstrate strong performance on familiar rules, their accuracy declines sharply when faced with novel or omitted rules. Moreover, the gap between token-level accuracy and complete answer accuracy highlights fundamental limitations in current approaches. These findings provide new insights into the reasoning mechanisms underlying deep learning models and underscore the need for architectures that move beyond pattern recognition toward robust abstract reasoning.

---

### HAVIR: HierArchical Vision to Image Reconstruction using CLIP-Guided Versatile Diffusion

Authors: Shiyi Zhang, Dong Liang, Hairong Zheng, Yihang Zhou
Published: 2025-10-03
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.03122v2

The reconstruction of visual information from brain activity fosters interdisciplinary integration between neuroscience and computer vision. However, existing methods still face challenges in accurately recovering highly complex visual stimuli. This difficulty stems from the characteristics of natural scenes: low-level features exhibit heterogeneity, while high-level features show semantic entanglement due to contextual overlaps. Inspired by the hierarchical representation theory of the visual cortex, we propose the HAVIR model, which separates the visual cortex into two hierarchical regions and extracts distinct features from each. Specifically, the Structural Generator extracts structural information from spatial processing voxels and converts it into latent diffusion priors, while the Semantic Extractor converts semantic processing voxels into CLIP embeddings. These components are integrated via the Versatile Diffusion model to synthesize the final image. Experimental results demonstrate that HAVIR enhances both the structural and semantic quality of reconstructions, even in complex scenes, and outperforms existing models.

---

### Real Time Headway Predictions in Urban Rail Systems and Implications for Service Control: A Deep Learning Approach

Authors: Muhammad Usama, Haris Koutsopoulos
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03121v1

Efficient real-time dispatching in urban metro systems is essential for ensuring service reliability, maximizing resource utilization, and improving passenger satisfaction. This study presents a novel deep learning framework centered on a Convolutional Long Short-Term Memory (ConvLSTM) model designed to predict the complex spatiotemporal propagation of train headways across an entire metro line. By directly incorporating planned terminal headways as a critical input alongside historical headway data, the proposed model accurately forecasts future headway dynamics, effectively capturing both their temporal evolution and spatial dependencies across all stations. This capability empowers dispatchers to evaluate the impact of various terminal headway control decisions without resorting to computationally intensive simulations. We introduce a flexible methodology to simulate diverse dispatcher strategies, ranging from maintaining even headways to implementing custom patterns derived from observed terminal departures. In contrast to existing research primarily focused on passenger load predictioning or atypical disruption scenarios, our approach emphasizes proactive operational control. Evaluated on a large-scale dataset from an urban metro line, the proposed ConvLSTM model demonstrates promising headway predictions, offering actionable insights for real-time decision-making. This framework provides rail operators with a powerful, computationally efficient tool to optimize dispatching strategies, thereby significantly improving service consistency and passenger satisfaction.

---

### SurveyBench: Can LLM(-Agents) Write Academic Surveys that Align with Reader Needs?

Authors: Zhaojun Sun, Xuzhou Zhu, Xuanhe Zhou, Xin Tong, Shuo Wang, Jie Fu, Guoliang Li, Zhiyuan Liu, Fan Wu
Published: 2025-10-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.03120v2

Academic survey writing, which distills vast literature into a coherent and insightful narrative, remains a labor-intensive and intellectually demanding task. While recent approaches, such as general DeepResearch agents and survey-specialized methods, can generate surveys automatically (a.k.a. LLM4Survey), their outputs often fall short of human standards and there lacks a rigorous, reader-aligned benchmark for thoroughly revealing their deficiencies. To fill the gap, we propose a fine-grained, quiz-driven evaluation framework SurveyBench, featuring (1) typical survey topics source from recent 11,343 arXiv papers and corresponding 4,947 high-quality surveys; (2) a multifaceted metric hierarchy that assesses the outline quality (e.g., coverage breadth, logical coherence), content quality (e.g., synthesis granularity, clarity of insights), and non-textual richness; and (3) a dual-mode evaluation protocol that includes content-based and quiz-based answerability tests, explicitly aligned with readers' informational needs. Results show SurveyBench effectively challenges existing LLM4Survey approaches (e.g., on average 21% lower than human in content-based evaluation).

---

### Taming Text-to-Sounding Video Generation via Advanced Modality Condition and Interaction

Authors: Kaisi Guan, Xihua Wang, Zhengfeng Lai, Xin Cheng, Peng Zhang, XiaoJiang Liu, Ruihua Song, Meng Cao
Published: 2025-10-03
Categories: cs.CV, cs.SD
arXiv: https://arxiv.org/abs/2510.03117v2

This study focuses on a challenging yet promising task, Text-to-Sounding-Video (T2SV) generation, which aims to generate a video with synchronized audio from text conditions, meanwhile ensuring both modalities are aligned with text. Despite progress in joint audio-video training, two critical challenges still remain unaddressed: (1) a single, shared text caption where the text for video is equal to the text for audio often creates modal interference, confusing the pretrained backbones, and (2) the optimal mechanism for cross-modal feature interaction remains unclear. To address these challenges, we first propose the Hierarchical Visual-Grounded Captioning (HVGC) framework that generates pairs of disentangled captions, a video caption, and an audio caption, eliminating interference at the conditioning stage. Based on HVGC, we further introduce BridgeDiT, a novel dual-tower diffusion transformer, which employs a Dual CrossAttention (DCA) mechanism that acts as a robust ``bridge" to enable a symmetric, bidirectional exchange of information, achieving both semantic and temporal synchronization. Extensive experiments on three benchmark datasets, supported by human evaluations, demonstrate that our method achieves state-of-the-art results on most metrics. Comprehensive ablation studies further validate the effectiveness of our contributions, offering key insights for the future T2SV task. All the codes and checkpoints will be publicly released.

---

### Listening or Reading? Evaluating Speech Awareness in Chain-of-Thought Speech-to-Text Translation

Authors: Jacobo Romero-Díaz, Gerard I. Gállego, Oriol Pareras, Federico Costa, Javier Hernando, Cristina España-Bonet
Published: 2025-10-03
Categories: cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2510.03115v2

Speech-to-Text Translation (S2TT) systems built from Automatic Speech Recognition (ASR) and Text-to-Text Translation (T2TT) modules face two major limitations: error propagation and the inability to exploit prosodic or other acoustic cues. Chain-of-Thought (CoT) prompting has recently been introduced, with the expectation that jointly accessing speech and transcription will overcome these issues. Analyzing CoT through attribution methods, robustness evaluations with corrupted transcripts, and prosody-awareness, we find that it largely mirrors cascaded behavior, relying mainly on transcripts while barely leveraging speech. Simple training interventions, such as adding Direct S2TT data or noisy transcript injection, enhance robustness and increase speech attribution. These findings challenge the assumed advantages of CoT and highlight the need for architectures that explicitly integrate acoustic information into translation.

---

### GeoComplete: Geometry-Aware Diffusion for Reference-Driven Image Completion

Authors: Beibei Lin, Tingting Chen, Robby T. Tan
Published: 2025-10-03
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.03110v1

Reference-driven image completion, which restores missing regions in a target view using additional images, is particularly challenging when the target view differs significantly from the references. Existing generative methods rely solely on diffusion priors and, without geometric cues such as camera pose or depth, often produce misaligned or implausible content. We propose GeoComplete, a novel framework that incorporates explicit 3D structural guidance to enforce geometric consistency in the completed regions, setting it apart from prior image-only approaches. GeoComplete introduces two key ideas: conditioning the diffusion process on projected point clouds to infuse geometric information, and applying target-aware masking to guide the model toward relevant reference cues. The framework features a dual-branch diffusion architecture. One branch synthesizes the missing regions from the masked target, while the other extracts geometric features from the projected point cloud. Joint self-attention across branches ensures coherent and accurate completion. To address regions visible in references but absent in the target, we project the target view into each reference to detect occluded areas, which are then masked during training. This target-aware masking directs the model to focus on useful cues, enhancing performance in difficult scenarios. By integrating a geometry-aware dual-branch diffusion architecture with a target-aware masking strategy, GeoComplete offers a unified and robust solution for geometry-conditioned image completion. Experiments show that GeoComplete achieves a 17.1 PSNR improvement over state-of-the-art methods, significantly boosting geometric accuracy while maintaining high visual quality.

---

### Rates of Convergence of Generalised Variational Inference Posteriors under Prior Misspecification

Authors: Terje Mildner, Paris Giampouras, Theodoros Damoulas
Published: 2025-10-03
Categories: math.ST, stat.ML
arXiv: https://arxiv.org/abs/2510.03109v2

We prove rates of convergence and robustness to prior misspecification within a Generalised Variational Inference (GVI) framework with bounded divergences. This addresses a significant open challenge for GVI and Federated GVI that employ a different divergence to the Kullback-Leibler under prior misspecification, operate within a subset of possible probability measures, and result in intractable posteriors. Our theoretical contributions extend to misspecified priors that lead to inconsistent Bayes posteriors. In particular, we are able to establish sufficient conditions for existence and uniqueness of GVI posteriors on arbitrary Polish spaces, prove that the GVI posterior measure concentrates on a neighbourhood of loss minimisers, and extend this to rates of convergence regardless of the prior measure.

---

### Geometry Meets Vision: Revisiting Pretrained Semantics in Distilled Fields

Authors: Zhiting Mei, Ola Shorinwa, Anirudha Majumdar
Published: 2025-10-03
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2510.03104v1

Semantic distillation in radiance fields has spurred significant advances in open-vocabulary robot policies, e.g., in manipulation and navigation, founded on pretrained semantics from large vision models. While prior work has demonstrated the effectiveness of visual-only semantic features (e.g., DINO and CLIP) in Gaussian Splatting and neural radiance fields, the potential benefit of geometry-grounding in distilled fields remains an open question. In principle, visual-geometry features seem very promising for spatial tasks such as pose estimation, prompting the question: Do geometry-grounded semantic features offer an edge in distilled fields? Specifically, we ask three critical questions: First, does spatial-grounding produce higher-fidelity geometry-aware semantic features? We find that image features from geometry-grounded backbones contain finer structural details compared to their counterparts. Secondly, does geometry-grounding improve semantic object localization? We observe no significant difference in this task. Thirdly, does geometry-grounding enable higher-accuracy radiance field inversion? Given the limitations of prior work and their lack of semantics integration, we propose a novel framework SPINE for inverting radiance fields without an initial guess, consisting of two core components: coarse inversion using distilled semantics, and fine inversion using photometric-based optimization. Surprisingly, we find that the pose estimation accuracy decreases with geometry-grounded features. Our results suggest that visual-only features offer greater versatility for a broader range of downstream tasks, although geometry-grounded features contain more geometric detail. Notably, our findings underscore the necessity of future research on effective strategies for geometry-grounding that augment the versatility and performance of pretrained semantic features.

---

### Semantic Similarity in Radiology Reports via LLMs and NER

Authors: Beth Pearson, Ahmed Adnan, Zahraa S. Abdallah
Published: 2025-10-03
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.03102v2

Radiology report evaluation is a crucial part of radiologists' training and plays a key role in ensuring diagnostic accuracy. As part of the standard reporting workflow, a junior radiologist typically prepares a preliminary report, which is then reviewed and edited by a senior radiologist to produce the final report. Identifying semantic differences between preliminary and final reports is essential for junior doctors, both as a training tool and to help uncover gaps in clinical knowledge. While AI in radiology is a rapidly growing field, the application of large language models (LLMs) remains challenging due to the need for specialised domain knowledge. In this paper, we explore the ability of LLMs to provide explainable and accurate comparisons of reports in the radiology domain. We begin by comparing the performance of several LLMs in comparing radiology reports. We then assess a more traditional approach based on Named-Entity-Recognition (NER). However, both approaches exhibit limitations in delivering accurate feedback on semantic similarity. To address this, we propose Llama-EntScore, a semantic similarity scoring method using a combination of Llama 3.1 and NER with tunable weights to emphasise or de-emphasise specific types of differences. Our approach generates a quantitative similarity score for tracking progress and also gives an interpretation of the score that aims to offer valuable guidance in reviewing and refining their reporting. We find our method achieves 67% exact-match accuracy and 93% accuracy within +/- 1 when compared to radiologist-provided ground truth scores - outperforming both LLMs and NER used independently. Code is available at: https://github.com/otmive/llama_reports

---

### AdaBet: Gradient-free Layer Selection for Efficient Training of Deep Neural Networks

Authors: Irene Tenison, Soumyajit Chatterjee, Fahim Kawsar, Mohammad Malekzadeh
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03101v2

To utilize pre-trained neural networks on edge and mobile devices, we often require efficient adaptation to user-specific runtime data distributions while operating under limited compute and memory resources. On-device retraining with a target dataset can facilitate such adaptations; however, it remains impractical due to the increasing depth of modern neural nets, as well as the computational overhead associated with gradient-based optimization across all layers. Current approaches reduce training cost by selecting a subset of layers for retraining, however, they rely on labeled data, at least one full-model backpropagation, or server-side meta-training; limiting their suitability for constrained devices. We introduce AdaBet, a gradient-free layer selection approach to rank important layers by analyzing topological features of their activation spaces through Betti Numbers and using forward passes alone. AdaBet allows selecting layers with high learning capacity, which are important for retraining and adaptation, without requiring labels or gradients. Evaluating AdaBet on sixteen pairs of benchmark models and datasets, shows AdaBet achieves an average gain of 2.5% more classification accuracy over gradient-based baselines while reducing average peak memory consumption by 40%.

---

### Adaptive Node Feature Selection For Graph Neural Networks

Authors: Madeline Navarro, Ali Azizpour, Santiago Segarra
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03096v3

We propose an adaptive node feature selection approach for graph neural networks (GNNs) that identifies and removes unnecessary features during training. The ability to measure how features contribute to model output is key for interpreting decisions and reducing dimensionality by eliminating unhelpful variables. However, graph-structured data introduces complex dependencies that may be unsuited to classical feature importance metrics. Inspired by this, we present a data-, model-, and task-agnostic method that determines relevant features during training based on changes in validation performance upon permuting feature values. We theoretically motivate our approach by characterizing how the relationships between node data and graph structure influences GNN performance. Empirically, we show that (i) our highly general approach rivals the performance of tailored feature selection approaches that exploit prior assumptions; (ii) we return meaningful feature importance scores well before the GNN is fully trained; and (iii) our scores demonstrably extract relevant properties that inform feature importance for various graph learning settings.

---

### SPID: Distilled Protein Backbone Generation

Authors: Liyang Xie, Haoran Zhang, Zhendong Wang, Wesley Tansey, Mingyuan Zhou
Published: 2025-10-03
Categories: cs.LG, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2510.03095v4

Diffusion- and flow-based generative models have recently demonstrated strong performance in protein backbone generation tasks, offering unprecedented capabilities for de novo protein design. However, despite their generation quality, these models are constrained by slow sampling, often requiring hundreds of iterative steps. This computational bottleneck limits their practical utility in large-scale protein discovery, where thousands to millions of candidate structures are needed. To address this challenge, we explore the techniques of score distillation, which has shown great success in reducing the number of sampling steps in the vision domain while maintaining high generation quality. However, a straightforward adaptation of these methods results in unacceptably low designability. We introduce Score Protein identity Distillation (SPID), which resolves this incompatibility by combining few-step generation with inference-time noise scaling. SPID adapts the Score identity Distillation (SiD) framework to both diffusion- and flow-based models without requiring access to pretraining data. Applied to the Proteina flow-matching model, our 16-step generator achieves 94.4% designability, matching the 400-step teacher, while delivering more than a 20-fold reduction in effective backbone-generation time and maintaining comparable diversity and novelty. SPID generalizes across unconditional generation, fold-class conditional generation, and motif scaffolding, and extends to equivariant diffusion architectures, achieving significant reduction in generation time with comparable generation quality to the teacher in all tasks. The resulting reduction in inference cost could facilitate large-scale in silico protein design, thereby advancing diffusion-based models toward real-world protein engineering applications. The PyTorch implementation is available at https://github.com/LY-Xie/SiD_Protein

---

### Revisiting Direct Speech-to-Text Translation with Speech LLMs: Better Scaling than CoT Prompting?

Authors: Oriol Pareras, Gerard I. Gállego, Federico Costa, Cristina España-Bonet, Javier Hernando
Published: 2025-10-03
Categories: cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2510.03093v2

Recent work on Speech-to-Text Translation (S2TT) has focused on LLM-based models, introducing the increasingly adopted Chain-of-Thought (CoT) prompting, where the model is guided to first transcribe the speech and then translate it. CoT typically outperforms direct prompting primarily because it can exploit abundant Automatic Speech Recognition (ASR) and Text-to-Text Translation (T2TT) datasets to explicitly model its steps. In this paper, we systematically compare CoT and Direct prompting under increasing amounts of S2TT data. To this end, we pseudo-label an ASR corpus by translating its transcriptions into six European languages, and train LLM-based S2TT systems with both prompting strategies at different data scales. Our results show that Direct improves more consistently as the amount of data increases, suggesting that it may become a more effective approach as larger S2TT resources are created.

---

### Latent Diffusion Unlearning: Protecting Against Unauthorized Personalization Through Trajectory Shifted Perturbations

Authors: Naresh Kumar Devulapally, Shruti Agarwal, Tejas Gokhale, Vishnu Suresh Lokhande
Published: 2025-10-03
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.03089v1

Text-to-image diffusion models have demonstrated remarkable effectiveness in rapid and high-fidelity personalization, even when provided with only a few user images. However, the effectiveness of personalization techniques has lead to concerns regarding data privacy, intellectual property protection, and unauthorized usage. To mitigate such unauthorized usage and model replication, the idea of generating ``unlearnable'' training samples utilizing image poisoning techniques has emerged. Existing methods for this have limited imperceptibility as they operate in the pixel space which results in images with noise and artifacts. In this work, we propose a novel model-based perturbation strategy that operates within the latent space of diffusion models. Our method alternates between denoising and inversion while modifying the starting point of the denoising trajectory: of diffusion models. This trajectory-shifted sampling ensures that the perturbed images maintain high visual fidelity to the original inputs while being resistant to inversion and personalization by downstream generative models. This approach integrates unlearnability into the framework of Latent Diffusion Models (LDMs), enabling a practical and imperceptible defense against unauthorized model adaptation. We validate our approach on four benchmark datasets to demonstrate robustness against state-of-the-art inversion attacks. Results demonstrate that our method achieves significant improvements in imperceptibility ($\sim 8 \% -10\%$ on perceptual metrics including PSNR, SSIM, and FID) and robustness ( $\sim 10\%$ on average across five adversarial settings), highlighting its effectiveness in safeguarding sensitive data.

---

### Chaining 2-FWL GNNs for Combinatorial Graph Alignment

Authors: Marc Lelarge
Published: 2025-10-03
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.03086v2

For the combinatorial graph alignment problem (GAP) -- finding the node correspondence that maximizes the number of common edges (nce) between two unlabeled graphs -- properly initialized FAQ remains a strong classical baseline, while existing GNN approaches struggle in the purely structural setting. We introduce a chaining procedure: a sequence of Folklore-type (2-FWL) GNNs in which each network is trained with cross-entropy after decoding the previous network's similarity matrix and ranking nodes by their current alignment quality. This non-differentiable ranking step injects discrete combinatorial feedback at every link; at inference, we iterate the final network and keep the candidate with highest observed nce. On sparse Erdos-Renyi graphs at noise level 0.25, chained FGNNs with FAQ post-processing reach 85% accuracy versus 13% for FAQ initialized from the convex relaxation, and essentially 0% for prior GNN methods. On correlated regular graphs, where MPNNs with constant features produce identical node embeddings (1-WL fails to refine) and FAQ's convex initialization is degenerate, chaining is the only method we know that recovers a non-trivial alignment. On three real-world benchmarks (yeast PPI, coauthorship, and road networks), we show that recent comparisons underestimate FAQ by initializing it from a uniform doubly stochastic matrix; once FAQ is initialized from the convex relaxation it already surpasses prior reported numbers, and dataset-specific chained FGNNs further improve on this strengthened baseline.

---

### From Facts to Foils: Designing and Evaluating Counterfactual Explanations for Smart Environments

Authors: Anna Trapp, Mersedeh Sadeghi, Andreas Vogelsang
Published: 2025-10-03
Categories: cs.AI, cs.SE
arXiv: https://arxiv.org/abs/2510.03078v1

Explainability is increasingly seen as an essential feature of rule-based smart environments. While counterfactual explanations, which describe what could have been done differently to achieve a desired outcome, are a powerful tool in eXplainable AI (XAI), no established methods exist for generating them in these rule-based domains. In this paper, we present the first formalization and implementation of counterfactual explanations tailored to this domain. It is implemented as a plugin that extends an existing explanation engine for smart environments. We conducted a user study (N=17) to evaluate our generated counterfactuals against traditional causal explanations. The results show that user preference is highly contextual: causal explanations are favored for their linguistic simplicity and in time-pressured situations, while counterfactuals are preferred for their actionable content, particularly when a user wants to resolve a problem. Our work contributes a practical framework for a new type of explanation in smart environments and provides empirical evidence to guide the choice of when each explanation type is most effective.

---

### What Drives Compositional Generalization in Visual Generative Models? The Importance of Continuous Training Objectives

Authors: Karim Farid, Rajat Sahay, Yumna Ali Alnaggar, Simon Schrodi, Volker Fischer, Cordelia Schmid, Thomas Brox
Published: 2025-10-03
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.03075v4

Compositional generalization, the ability to generate novel combinations of known concepts, is a key ingredient for visual generative models. Yet, not all mechanisms that enable or inhibit it are fully understood. In this work, we conduct a systematic study of which design choices critically determine compositional generalization in image and video generation. By isolating independent design axes, we identify two key factors strongly associated with compositional success: (i) whether the training objective operates on a discrete or continuous distribution, and (ii) the completeness of conditioning information about constituent factors during training. We also show that relaxing the discrete loss with an auxiliary continuous latent objective can partially recover compositional performance in discrete models like MaskGIT. Our findings, corroborated by diverse compositional tasks and preliminary evidence in world models and LLMs, motivate a shift toward continuous objectives for compositional generalization.

---

### Neural Posterior Estimation with Autoregressive Tiling for Detecting Objects in Astronomical Images

Authors: Jeffrey Regier
Published: 2025-10-03
Categories: stat.AP, astro-ph.IM, cs.CV
arXiv: https://arxiv.org/abs/2510.03074v1

Upcoming astronomical surveys will produce petabytes of high-resolution images of the night sky, providing information about billions of stars and galaxies. Detecting and characterizing the astronomical objects in these images is a fundamental task in astronomy -- and a challenging one, as most of these objects are faint and many visually overlap with other objects. We propose an amortized variational inference procedure to solve this instance of small-object detection. Our key innovation is a family of spatially autoregressive variational distributions that partition and order the latent space according to a $K$-color checkerboard pattern. By construction, the conditional independencies of this variational family mirror those of the posterior distribution. We fit the variational distribution, which is parameterized by a convolutional neural network, using neural posterior estimation (NPE) to minimize an expectation of the forward KL divergence. Using images from the Sloan Digital Sky Survey, our method achieves state-of-the-art performance. We further demonstrate that the proposed autoregressive structure greatly improves posterior calibration.

---

### A Study of Neural Polar Decoders for Communication

Authors: Rom Hirsch, Ziv Aharoni, Henry D. Pfister, Haim H. Permuter
Published: 2025-10-03
Categories: eess.SP, cs.AI
arXiv: https://arxiv.org/abs/2510.03069v1

In this paper, we adapt and analyze Neural Polar Decoders (NPDs) for end-to-end communication systems. While prior work demonstrated the effectiveness of NPDs on synthetic channels, this study extends the NPD to real-world communication systems. The NPD was adapted to complete OFDM and single-carrier communication systems. To satisfy practical system requirements, the NPD is extended to support any code length via rate matching, higher-order modulations, and robustness across diverse channel conditions. The NPD operates directly on channels with memory, exploiting their structure to achieve higher data rates without requiring pilots and a cyclic prefix. Although NPD entails higher computational complexity than the standard 5G polar decoder, its neural network architecture enables an efficient representation of channel statistics, resulting in manageable complexity suitable for practical systems. Experimental results over 5G channels demonstrate that the NPD consistently outperforms the 5G polar decoder in terms of BER, BLER, and throughput. These improvements are particularly significant for low-rate and short-block configurations, which are prevalent in 5G control channels. Furthermore, NPDs applied to single-carrier systems offer performance comparable to OFDM with lower PAPR, enabling effective single-carrier transmission over 5G channels. These results position the NPD as a high-performance, pilotless, and robust decoding solution.

---

### InsideOut: An EfficientNetV2-S Based Deep Learning Framework for Robust Multi-Class Facial Emotion Recognition

Authors: Ahsan Farabi, Israt Khandaker, Ibrahim Khalil Shanto, Md Abdul Ahad Minhaz, Tanisha Zaman
Published: 2025-10-03
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.03066v1

Facial Emotion Recognition (FER) is a key task in affective computing, enabling applications in human-computer interaction, e-learning, healthcare, and safety systems. Despite advances in deep learning, FER remains challenging due to occlusions, illumination and pose variations, subtle intra-class differences, and dataset imbalance that hinders recognition of minority emotions. We present InsideOut, a reproducible FER framework built on EfficientNetV2-S with transfer learning, strong data augmentation, and imbalance-aware optimization. The approach standardizes FER2013 images, applies stratified splitting and augmentation, and fine-tunes a lightweight classification head with class-weighted loss to address skewed distributions. InsideOut achieves 62.8% accuracy with a macro averaged F1 of 0.590 on FER2013, showing competitive results compared to conventional CNN baselines. The novelty lies in demonstrating that efficient architectures, combined with tailored imbalance handling, can provide practical, transparent, and reproducible FER solutions.

---

### A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman Problem

Authors: Mingfeng Fan, Jiaqi Cheng, Yaoxin Wu, Yifeng Zhang, Yibin Yang, Guohua Wu, Guillaume Sartoretti
Published: 2025-10-03
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.03065v1

In recent years, deep reinforcement learning (DRL) has gained traction for solving the NP-hard traveling salesman problem (TSP). However, limited attention has been given to the close-enough TSP (CETSP), primarily due to the challenge introduced by its neighborhood-based visitation criterion, wherein a node is considered visited if the agent enters a compact neighborhood around it. In this work, we formulate a Markov decision process (MDP) for CETSP using a discretization scheme and propose a novel unified dual-decoder DRL (UD3RL) framework that separates decision-making into node selection and waypoint determination. Specifically, an adapted encoder is employed for effective feature extraction, followed by a node-decoder and a loc-decoder to handle the two sub-tasks, respectively. A k-nearest neighbors subgraph interaction strategy is further introduced to enhance spatial reasoning during location decoding. Furthermore, we customize the REINFORCE algorithm to train UD3RL as a unified model capable of generalizing across different problem sizes and varying neighborhood radius types (i.e., constant and random radii). Experimental results show that UD3RL outperforms conventional methods in both solution quality and runtime, while exhibiting strong generalization across problem scales, spatial distributions, and radius ranges, as well as robustness to dynamic environments.

---

### Comparative Analysis of Parameterized Action Actor-Critic Reinforcement Learning Algorithms for Web Search Match Plan Generation

Authors: Ubayd Bapoo, Clement N Nyirenda
Published: 2025-10-03
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.03064v1

This study evaluates the performance of Soft Actor Critic (SAC), Greedy Actor Critic (GAC), and Truncated Quantile Critics (TQC) in high-dimensional decision-making tasks using fully observable environments. The focus is on parametrized action (PA) spaces, eliminating the need for recurrent networks, with benchmarks Platform-v0 and Goal-v0 testing discrete actions linked to continuous action-parameter spaces. Hyperparameter optimization was performed with Microsoft NNI, ensuring reproducibility by modifying the codebase for GAC and TQC. Results show that Parameterized Action Greedy Actor-Critic (PAGAC) outperformed other algorithms, achieving the fastest training times and highest returns across benchmarks, completing 5,000 episodes in 41:24 for the Platform game and 24:04 for the Robot Soccer Goal game. Its speed and stability provide clear advantages in complex action spaces. Compared to PASAC and PATQC, PAGAC demonstrated superior efficiency and reliability, making it ideal for tasks requiring rapid convergence and robust performance. Future work could explore hybrid strategies combining entropy-regularization with truncation-based methods to enhance stability and expand investigations into generalizability.

---

### Semantic Differentiation in Speech Emotion Recognition: Insights from Descriptive and Expressive Speech Roles

Authors: Rongchen Guo, Vincent Francoeur, Isar Nejadgholi, Sylvain Gagnon, Miodrag Bolic
Published: 2025-10-03
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.03060v1

Speech Emotion Recognition (SER) is essential for improving human-computer interaction, yet its accuracy remains constrained by the complexity of emotional nuances in speech. In this study, we distinguish between descriptive semantics, which represents the contextual content of speech, and expressive semantics, which reflects the speaker's emotional state. After watching emotionally charged movie segments, we recorded audio clips of participants describing their experiences, along with the intended emotion tags for each clip, participants' self-rated emotional responses, and their valence/arousal scores. Through experiments, we show that descriptive semantics align with intended emotions, while expressive semantics correlate with evoked emotions. Our findings inform SER applications in human-AI interaction and pave the way for more context-aware AI systems.

---