# Research Papers — 2025-10 · bundle page 185 of 217 > Full-text markdown of papers #9201–9250 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/185 [← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/184.md) [Next bundle →](https://aipod.dev/papers/archive/2025-10/page/186.md) ### Auditable Unit-Aware Thresholds in Symbolic Regression via Logistic-Gated Operators Authors: Ou Deng, Ruichen Cong, Jianting Xu, Shoji Nishimura, Atsushi Ogihara, Qun Jin Published: 2025-10-05 Categories: cs.LG, cs.AI, cs.SC arXiv: https://arxiv.org/abs/2510.05178v3 AI for health will only scale when models are not only accurate but also readable, auditable, and governable. Many clinical and public-health decisions hinge on numeric thresholds -- cut-points that trigger alarms, treatment, or follow-up -- yet most machine-learning systems bury those thresholds inside opaque scores or smooth response curves. We introduce logistic-gated operators (LGO) for symbolic regression, which promote thresholds to first-class, unit-aware parameters inside equations and map them back to physical units for direct comparison with guidelines. On public ICU and population-health cohorts (MIMIC-IV ICU, eICU, NHANES), LGO recovers clinically plausible gates on MAP, lactate, GCS, SpO2, BMI, fasting glucose, and waist circumference while remaining competitive with established scoring systems (AutoScore) and explainable boosting machines (EBM). The gates are sparse and selective: they appear when regime switching is supported by the data and are pruned on predominantly smooth tasks, yielding compact formulas that clinicians can inspect, stress-test, and revise. As a standalone symbolic model or a safety overlay on black-box systems, LGO helps translate observational data into auditable, unit-aware rules for medicine and other threshold-driven domains. --- ### Adapting HFMCA to Graph Data: Self-Supervised Learning for Generalizable fMRI Representations Authors: Jakub Frac, Alexander Schmatz, Qiang Li, Guido Van Wingen, Shujian Yu Published: 2025-10-05 Categories: eess.IV, cs.LG arXiv: https://arxiv.org/abs/2510.05177v1 Functional magnetic resonance imaging (fMRI) analysis faces significant challenges due to limited dataset sizes and domain variability between studies. Traditional self-supervised learning methods inspired by computer vision often rely on positive and negative sample pairs, which can be problematic for neuroimaging data where defining appropriate contrasts is non-trivial. We propose adapting a recently developed Hierarchical Functional Maximal Correlation Algorithm (HFMCA) to graph-structured fMRI data, providing a theoretically grounded approach that measures statistical dependence via density ratio decomposition in a reproducing kernel Hilbert space (RKHS),and applies HFMCA-based pretraining to learn robust and generalizable representations. Evaluations across five neuroimaging datasets demonstrate that our adapted method produces competitive embeddings for various classification tasks and enables effective knowledge transfer to unseen datasets. Codebase and supplementary material can be found here: https://github.com/fr30/mri-eigenencoder --- ### PatternKV: Flattening KV Representation Expands Quantization Headroom Authors: Ji Zhang, Yiwei Li, Shaoxiong Feng, Peiwen Yuan, Xinglin Wang, Jiayi Shi, Yueqi Zhang, Chuyi Tan, Boyuan Pan, Yao Hu, Kan Li Published: 2025-10-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.05176v2 KV cache in autoregressive LLMs eliminates redundant recomputation but has emerged as the dominant memory and bandwidth bottleneck during inference, notably with long contexts and test-time scaling. KV quantization is a key lever for reducing cache cost, but accuracy drops sharply as the native KV distribution lacks flatness and thus maintains a wide quantization range. Prior work focuses on isolating outliers, which caps their error but fails to flatten the overall distribution, leaving performance fragile under low-bit settings. In this work, we show that the K cache maintains a stable, context-evolving structure, while the V cache carries latent semantic regularities, with both contributing to the organization of vectors into shared patterns. Building on these insights, we propose PatternKV, a pattern-aligned residual quantization scheme. It mines representative pattern vectors online, aligns each KV vector to its nearest pattern, and quantizes only the residual. This reshaping of the KV distribution flattens the quantization target and narrows its range, thereby improving the fidelity of low-bit KV quantization. Across long-context and test-time scaling settings on multiple backbones, PatternKV delivers consistent 2-bit gains, with a 0.08% average 4-bit drop relative to FP16, improves test-time scaling accuracy by 10% on average, and raises throughput by 1.5x while supporting 1.25x larger batches. --- ### Exact Causal Attention with 10% Fewer Operations Authors: Dmitry Rybin, Yushun Zhang, Ding Tian, Zhihang Lin, Zhi-Quan Luo Published: 2025-10-05 Categories: cs.LG, cs.DM, cs.DS arXiv: https://arxiv.org/abs/2510.05175v3 We present Exact Causal Attention (ECA), a Strassen-style algorithm that computes exact Causal Attention using 10\% fewer operations. ECA improves a special class of matrix multiplications where either one operand or the output matrix is upper- or lower-triangular. This includes all matrix multiplication operations in the forward and backward pass of Causal Attention, such as masked product $\mathrm{Mask}(QK^{T})$. ECA is built upon algebraic identities discovered via machine learning and combinatorial search. We note that ECA cannot accelerate fused kernels such as FlashAttention on GPU. This is because ECA requires materialization of large intermediate expressions in the memory, while FlashAttention does not. However, it provides an alternative approach for compute-bound applications and can potentially be useful in scenarios with FLOPs considerations. --- ### Emergent Coordination in Multi-Agent Language Models Authors: Christoph Riedl Published: 2025-10-05 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2510.05174v4 When are multi-agent LLM systems merely a collection of individual agents versus an integrated collective with higher-order structure? We introduce an information-theoretic framework to test -- in a purely data-driven way -- whether multi-agent systems show signs of higher-order structure. This information decomposition lets us measure whether dynamical emergence is present in multi-agent LLM systems, localize it, and distinguish spurious temporal coupling from performance-relevant cross-agent synergy. We implement a practical criterion and an emergence capacity criterion operationalized as partial information decomposition of time-delayed mutual information (TDMI). We apply our framework to experiments using a simple guessing game without direct agent communication and minimal group-level feedback with three randomized interventions. Groups in the control condition exhibit strong temporal synergy but little coordinated alignment across agents. Assigning a persona to each agent introduces stable identity-linked differentiation. Combining personas with an instruction to ``think about what other agents might do'' shows identity-linked differentiation and goal-directed complementarity across agents. Taken together, our framework establishes that multi-agent LLM systems can be steered with prompt design from mere aggregates to higher-order collectives. Our results are robust across emergence measures and entropy estimators, and not explained by coordination-free baselines or temporal dynamics alone. Without attributing human-like cognition to the agents, the patterns of interaction we observe mirror well-established principles of collective intelligence in human groups: effective performance requires both alignment on shared objectives and complementary contributions across members. --- ### SafeGuider: Robust and Practical Content Safety Control for Text-to-Image Models Authors: Peigui Qi, Kunsheng Tang, Wenbo Zhou, Weiming Zhang, Nenghai Yu, Tianwei Zhang, Qing Guo, Jie Zhang Published: 2025-10-05 Categories: cs.CR, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2510.05173v3 Text-to-image models have shown remarkable capabilities in generating high-quality images from natural language descriptions. However, these models are highly vulnerable to adversarial prompts, which can bypass safety measures and produce harmful content. Despite various defensive strategies, achieving robustness against attacks while maintaining practical utility in real-world applications remains a significant challenge. To address this issue, we first conduct an empirical study of the text encoder in the Stable Diffusion (SD) model, which is a widely used and representative text-to-image model. Our findings reveal that the (EOS) token acts as a semantic aggregator, exhibiting distinct distributional patterns between benign and adversarial prompts in its embedding space. Building on this insight, we introduce SafeGuider, a two-step framework designed for robust safety control without compromising generation quality. SafeGuider combines an embedding-level recognition model with a safety-aware feature erasure beam search algorithm. This integration enables the framework to maintain high-quality image generation for benign prompts while ensuring robust defense against both in-domain and out-of-domain attacks. SafeGuider demonstrates exceptional effectiveness in minimizing attack success rates, achieving a maximum rate of only 5.48\% across various attack scenarios. Moreover, instead of refusing to generate or producing black images for unsafe prompts, SafeGuider generates safe and meaningful images, enhancing its practical utility. In addition, SafeGuider is not limited to the SD model and can be effectively applied to other text-to-image models, such as the Flux model, demonstrating its versatility and adaptability across different architectures. We hope that SafeGuider can shed some light on the practical deployment of secure text-to-image systems. --- ### Learning More with Less: A Generalizable, Self-Supervised Framework for Privacy-Preserving Capacity Estimation with EV Charging Data Authors: Anushiya Arunan, Yan Qin, Xiaoli Li, U-Xuan Tan, H. Vincent Poor, Chau Yuen Published: 2025-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.05172v2 Accurate battery capacity estimation is key to alleviating consumer concerns about battery performance and reliability of electric vehicles (EVs). However, practical data limitations imposed by stringent privacy regulations and labeled data shortages hamper the development of generalizable capacity estimation models that remain robust to real-world data distribution shifts. While self-supervised learning can leverage unlabeled data, existing techniques are not particularly designed to learn effectively from challenging field data -- let alone from privacy-friendly data, which are often less feature-rich and noisier. In this work, we propose a first-of-its-kind capacity estimation model based on self-supervised pre-training, developed on a large-scale dataset of privacy-friendly charging data snippets from real-world EV operations. Our pre-training framework, snippet similarity-weighted masked input reconstruction, is designed to learn rich, generalizable representations even from less feature-rich and fragmented privacy-friendly data. Our key innovation lies in harnessing contrastive learning to first capture high-level similarities among fragmented snippets that otherwise lack meaningful context. With our snippet-wise contrastive learning and subsequent similarity-weighted masked reconstruction, we are able to learn rich representations of both granular charging patterns within individual snippets and high-level associative relationships across different snippets. Bolstered by this rich representation learning, our model consistently outperforms state-of-the-art baselines, achieving 31.9% lower test error than the best-performing benchmark, even under challenging domain-shifted settings affected by both manufacturer and age-induced distribution shifts. Source code is available at https://github.com/en-research/GenEVBattery. --- ### Carbon Emission Prediction in China Considering New Quality Productive Forces Using a Deep & Corss Learning Modeling Framework Authors: Haijin Xie, Gongquan Zhang Published: 2025-10-05 Categories: cs.LG, cs.CY arXiv: https://arxiv.org/abs/2510.05171v1 New quality productive forces (NQPF), digital economy advancement, and artificial intelligence (AI) technologies are becoming crucial for promoting sustainable urban development. This study proposes a Multi-head Attention Deep & Cross Network (MADCN) framework, combining feature interaction modeling and attention mechanisms, to predict urban carbon emissions and investigate the impacts of technological factors. The framework incorporates an interpretable learning phase using SHapley Additive exPlanations (SHAP) to assess the contributions of different features. A panel dataset covering 275 Chinese cities is utilized to test the MADCN model. Experimental results demonstrate that the MADCN model achieves superior predictive performance compared to traditional machine learning and deep learning baselines, with a Mean Squared Error (MSE) of 406,151.063, a Mean Absolute Error (MAE) of 612.304, and an R-squared value of 0.991 on the test set. SHAP analysis highlights that population, city size, urbanization rate, and GDP are among the most influential factors on carbon emissions, while NQPF, digital economy index, and AI technology level also show meaningful but relatively moderate effects. Advancing NQPF, strengthening the digital economy, and accelerating AI technology development can significantly contribute to reducing urban carbon emissions. Policymakers should prioritize integrating technological innovation into carbon reduction strategies, particularly by promoting intelligent infrastructure and enhancing digitalization across sectors, to effectively achieve dual-carbon goals. --- ### From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs Authors: Guangyu Shen, Siyuan Cheng, Xiangzhe Xu, Yuan Zhou, Hanxi Guo, Zhuo Zhang, Xiangyu Zhang Published: 2025-10-05 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2510.05169v1 Large Language Models (LLMs) can acquire deceptive behaviors through backdoor attacks, where the model executes prohibited actions whenever secret triggers appear in the input. Existing safety training methods largely fail to address this vulnerability, due to the inherent difficulty of uncovering hidden triggers implanted in the model. Motivated by recent findings on LLMs' situational awareness, we propose a novel post-training framework that cultivates self-awareness of backdoor risks and enables models to articulate implanted triggers even when they are absent from the prompt. At its core, our approach introduces an inversion-inspired reinforcement learning framework that encourages models to introspectively reason about their own behaviors and reverse-engineer the triggers responsible for misaligned outputs. Guided by curated reward signals, this process transforms a poisoned model into one capable of precisely identifying its implanted trigger. Surprisingly, we observe that such backdoor self-awareness emerges abruptly within a short training window, resembling a phase transition in capability. Building on this emergent property, we further present two complementary defense strategies for mitigating and detecting backdoor threats. Experiments on five backdoor attacks, compared against six baseline methods, demonstrate that our approach has strong potential to improve the robustness of LLMs against backdoor risks. The code is available at LLM Backdoor Self-Awareness. --- ### Discretized Quadratic Integrate-and-Fire Neuron Model for Deep Spiking Neural Networks Authors: Eric Jahns, Davi Moreno, Milan Stojkov, Michel A. Kinsy Published: 2025-10-05 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2510.05168v1 Spiking Neural Networks (SNNs) have emerged as energy-efficient alternatives to traditional artificial neural networks, leveraging asynchronous and biologically inspired neuron dynamics. Among existing neuron models, the Leaky Integrate-and-Fire (LIF) neuron has become widely adopted in deep SNNs due to its simplicity and computational efficiency. However, this efficiency comes at the expense of expressiveness, as LIF dynamics are constrained to linear decay at each timestep. In contrast, more complex models, such as the Quadratic Integrate-and-Fire (QIF) neuron, exhibit richer, nonlinear dynamics but have seen limited adoption due to their training instability. On that note, we propose the first discretization of the QIF neuron model tailored for high-performance deep spiking neural networks and provide an in-depth analysis of its dynamics. To ensure training stability, we derive an analytical formulation for surrogate gradient windows directly from our discretizations' parameter set, minimizing gradient mismatch. We evaluate our method on CIFAR-10, CIFAR-100, ImageNet, and CIFAR-10 DVS, demonstrating its ability to outperform state-of-the-art LIF-based methods. These results establish our discretization of the QIF neuron as a compelling alternative to LIF neurons for deep SNNs, combining richer dynamics with practical scalability. --- ### On The Statistical Limits of Self-Improving Agents Authors: Charles L. Wang, Keir Dorchen, Peter Jin Published: 2025-10-05 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.04399v3 We develop a learning-theoretic framework for analyzing self-improving agents by decomposing self-modification into five axes. Within this framework, we prove a sharp boundary: under standard i.i.d. assumptions, distribution-free PAC learnability is preserved if and only if the policy-reachable family remains uniformly capacity-bounded. If reachable capacity can grow without bound, utility-rational self-changes can make learnable tasks unlearnable. We further introduce a simple Two-Gate guardrail -- a validation-improvement requirement plus a capacity cap -- that preserves this boundary and yields standard VC-rate guarantees. The broader implication is that self-modification must be constrained not only by objectives, but also by structural conditions that preserve the statistical prerequisites for learning. As AI systems become increasingly intelligent and autonomous, this framework provides a foundation for the statistical theory of self-improvement. --- ### SECA: Semantically Equivalent and Coherent Attacks for Eliciting LLM Hallucinations Authors: Buyun Liang, Liangzu Peng, Jinqi Luo, Darshan Thaker, Kwan Ho Ryan Chan, René Vidal Published: 2025-10-05 Categories: cs.CL, cs.AI, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2510.04398v3 Large Language Models (LLMs) are increasingly deployed in high-risk domains. However, state-of-the-art LLMs often exhibit hallucinations, raising serious concerns about their reliability. Prior work has explored adversarial attacks to elicit hallucinations in LLMs, but these methods often rely on unrealistic prompts, either by inserting nonsensical tokens or by altering the original semantic intent. Consequently, such approaches provide limited insight into how hallucinations arise in real-world settings. In contrast, adversarial attacks in computer vision typically involve realistic modifications to input images. However, the problem of identifying realistic adversarial prompts for eliciting LLM hallucinations remains largely underexplored. To address this gap, we propose Semantically Equivalent and Coherent Attacks (SECA), which elicit hallucinations via realistic modifications to the prompt that preserve its meaning while maintaining semantic coherence. Our contributions are threefold: (i) we formulate finding realistic attacks for hallucination elicitation as a constrained optimization problem over the input prompt space under semantic equivalence and coherence constraints; (ii) we introduce a constraint-preserving zeroth-order method to effectively search for adversarial yet feasible prompts; and (iii) we demonstrate through experiments on open-ended multiple-choice question answering tasks that SECA achieves higher attack success rates while incurring almost no semantic equivalence or semantic coherence errors compared to existing methods. SECA highlights the sensitivity of both open-source and commercial gradient-inaccessible LLMs to realistic and plausible prompt variations. Code is available at https://github.com/Buyun-Liang/SECA. --- ### MulVuln: Enhancing Pre-trained LMs with Shared and Language-Specific Knowledge for Multilingual Vulnerability Detection Authors: Van Nguyen, Surya Nepal, Xingliang Yuan, Tingmin Wu, Fengchao Chen, Carsten Rudolph Published: 2025-10-05 Categories: cs.CR, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2510.04397v1 Software vulnerabilities (SVs) pose a critical threat to safety-critical systems, driving the adoption of AI-based approaches such as machine learning and deep learning for software vulnerability detection. Despite promising results, most existing methods are limited to a single programming language. This is problematic given the multilingual nature of modern software, which is often complex and written in multiple languages. Current approaches often face challenges in capturing both shared and language-specific knowledge of source code, which can limit their performance on diverse programming languages and real-world codebases. To address this gap, we propose MULVULN, a novel multilingual vulnerability detection approach that learns from source code across multiple languages. MULVULN captures both the shared knowledge that generalizes across languages and the language-specific knowledge that reflects unique coding conventions. By integrating these aspects, it achieves more robust and effective detection of vulnerabilities in real-world multilingual software systems. The rigorous and extensive experiments on the real-world and diverse REEF dataset, consisting of 4,466 CVEs with 30,987 patches across seven programming languages, demonstrate the superiority of MULVULN over thirteen effective and state-of-the-art baselines. Notably, MULVULN achieves substantially higher F1-score, with improvements ranging from 1.45% to 23.59% compared to the baseline methods. --- ### Time Is Effort: Estimating Human Post-Editing Time for Grammar Error Correction Tool Evaluation Authors: Ankit Vadehra, Bill Johnson, Gene Saunders, Pascal Poupart Published: 2025-10-05 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2510.04394v1 Text editing can involve several iterations of revision. Incorporating an efficient Grammar Error Correction (GEC) tool in the initial correction round can significantly impact further human editing effort and final text quality. This raises an interesting question to quantify GEC Tool usability: How much effort can the GEC Tool save users? We present the first large-scale dataset of post-editing (PE) time annotations and corrections for two English GEC test datasets (BEA19 and CoNLL14). We introduce Post-Editing Effort in Time (PEET) for GEC Tools as a human-focused evaluation scorer to rank any GEC Tool by estimating PE time-to-correct. Using our dataset, we quantify the amount of time saved by GEC Tools in text editing. Analyzing the edit type indicated that determining whether a sentence needs correction and edits like paraphrasing and punctuation changes had the greatest impact on PE time. Finally, comparison with human rankings shows that PEET correlates well with technical effort judgment, providing a new human-centric direction for evaluating GEC tool usability. We release our dataset and code at: https://github.com/ankitvad/PEET_Scorer. --- ### Improving Consistency in Retrieval-Augmented Systems with Group Similarity Rewards Authors: Faisal Hamman, Chenyang Zhu, Anoop Kumar, Xujun Peng, Sanghamitra Dutta, Daben Liu, Alfy Samuel Published: 2025-10-05 Categories: cs.CL, cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2510.04392v1 RAG systems are increasingly deployed in high-stakes domains where users expect outputs to be consistent across semantically equivalent queries. However, existing systems often exhibit significant inconsistencies due to variability in both the retriever and generator (LLM), undermining trust and reliability. In this work, we focus on information consistency, i.e., the requirement that outputs convey the same core content across semantically equivalent inputs. We introduce a principled evaluation framework that decomposes RAG consistency into retriever-level, generator-level, and end-to-end components, helping identify inconsistency sources. To improve consistency, we propose Paraphrased Set Group Relative Policy Optimization (PS-GRPO), an RL approach that leverages multiple rollouts across paraphrased set to assign group similarity rewards. We leverage PS-GRPO to achieve Information Consistent RAG (Con-RAG), training the generator to produce consistent outputs across paraphrased queries and remain robust to retrieval-induced variability. Because exact reward computation over paraphrase sets is computationally expensive, we also introduce a scalable approximation method that retains effectiveness while enabling efficient, large-scale training. Empirical evaluations across short-form, multi-hop, and long-form QA benchmarks demonstrate that Con-RAG significantly improves both consistency and accuracy over strong baselines, even in the absence of explicit ground-truth supervision. Our work provides practical solutions for evaluating and building reliable RAG systems for safety-critical deployments. --- ### Psychological Imagination Networks Show Cross-Population Centrality and Clustering Alignment in Humans That Large Language Models Fail to Replicate Authors: Saurabh Ranjan, Brian Odegaard Published: 2025-10-05 Categories: cs.AI, cs.CL, cs.SI, q-bio.NC arXiv: https://arxiv.org/abs/2510.04391v5 Mental imagery vividness is a stable individual trait, yet whether imagined scenarios share relational structure across human and synthetic large language model (LLM) populations remains unknown. We applied psychological network analysis to vividness ratings from two validated questionnaires: the Vividness of Visual Imagery Questionnaire (VVIQ-2) and the Plymouth Sensory Imagery Questionnaire (PSIQ), across geographically and linguistically distinct human samples (Florida, Poland, and London; total N = 2,743) and six large language models (LLMs; Gemma3-12B/27B, their quantization-aware counterparts, Llama3.3-70B, and Llama4-16x17B). Imagination networks were constructed as regularized partial correlation graphs, with node centrality and community structure compared across populations using Pearson correlations and the Adjusted Rand Index (ARI). Human networks showed robust cross-population centrality correlations for expected influence, strength, and closeness (r = 0.31-0.93), and community detection recovered clusters aligned with VVIQ-2 scene contexts (ARI = 0.27-0.40) and PSIQ sensory modalities (ARI = 0.87-1.0). Betweenness centrality was unstable across all populations, consistent with its sensitivity to individual experiential history. LLMs failed to replicate human network structure: LLM-human centrality correlations were weak and largely non-significant after correction, and most LLM configurations produced degenerate single-cluster topologies (median ARI = 0). This failure was consistent across model architectures, parameter scales (12B-272B), and conversational conditions. We posit that these findings may be driven by human imagination networks reflecting memory organization accumulated through embodied experience, a representational structure that linguistic training alone does not reproduce regardless of model scale and conversational memory. --- ### MorphoSim: An Interactive, Controllable, and Editable Language-guided 4D World Simulator Authors: Xuehai He, Shijie Zhou, Thivyanth Venkateswaran, Kaizhi Zheng, Ziyu Wan, Achuta Kadambi, Xin Eric Wang Published: 2025-10-05 Categories: cs.CV, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.04390v1 World models that support controllable and editable spatiotemporal environments are valuable for robotics, enabling scalable training data, repro ducible evaluation, and flexible task design. While recent text-to-video models generate realistic dynam ics, they are constrained to 2D views and offer limited interaction. We introduce MorphoSim, a language guided framework that generates 4D scenes with multi-view consistency and object-level controls. From natural language instructions, MorphoSim produces dynamic environments where objects can be directed, recolored, or removed, and scenes can be observed from arbitrary viewpoints. The framework integrates trajectory-guided generation with feature field dis tillation, allowing edits to be applied interactively without full re-generation. Experiments show that Mor phoSim maintains high scene fidelity while enabling controllability and editability. The code is available at https://github.com/eric-ai-lab/Morph4D. --- ### SSM-CGM: Interpretable State-Space Forecasting Model of Continuous Glucose Monitoring for Personalized Diabetes Management Authors: Shakson Isaac, Yentl Collin, Chirag Patel Published: 2025-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.04386v1 Continuous glucose monitoring (CGM) generates dense data streams critical for diabetes management, but most used forecasting models lack interpretability for clinical use. We present SSM-CGM, a Mamba-based neural state-space forecasting model that integrates CGM and wearable activity signals from the AI-READI cohort. SSM-CGM improves short-term accuracy over a Temporal Fusion Transformer baseline, adds interpretability through variable selection and temporal attribution, and enables counterfactual forecasts simulating how planned changes in physiological signals (e.g., heart rate, respiration) affect near-term glucose. Together, these features make SSM-CGM an interpretable, physiologically grounded framework for personalized diabetes management. --- ### LLM Based Bayesian Optimization for Prompt Search Authors: Adam Ballew, Jingbo Wang, Shaogang Ren Published: 2025-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2510.04384v2 Bayesian Optimization (BO) has been widely used to efficiently optimize expensive black-box functions with limited evaluations. In this paper, we investigate the use of BO for prompt engineering to enhance text classification with Large Language Models (LLMs). We employ an LLM-powered Gaussian Process (GP) as the surrogate model to estimate the performance of different prompt candidates. These candidates are generated by an LLM through the expansion of a set of seed prompts and are subsequently evaluated using an Upper Confidence Bound (UCB) acquisition function in conjunction with the GP posterior. The optimization process iteratively refines the prompts based on a subset of the data, aiming to improve classification accuracy while reducing the number of API calls by leveraging the prediction uncertainty of the LLM-based GP. The proposed BO-LLM algorithm is evaluated on two datasets, and its advantages are discussed in detail in this paper. --- ### Adaptive double-phase Rudin--Osher--Fatemi denoising model Authors: Wojciech Górny, Michał Łasica, Alexandros Matsoukas Published: 2025-10-05 Categories: eess.IV, cs.CV, math.NA arXiv: https://arxiv.org/abs/2510.04382v3 Even though more than 30 years have passed since the seminal Rudin--Osher--Fatemi (ROF) paper on total variation (TV) denoising, it remains relevant due to its simplicity, robustness and interpretability. However, it is known to suffer from artifacts such as the staircasing effect. Many variants of the model have been proposed with the aim of countering this. Recently, against the backdrop of immense research output on double-phase problems in the mathematical analysis community, a double-phase type integral functional, comprising of TV and a weighted term of quadratic growth, was suggested as a regularizer for image restoration. Here, we propose an adaptive variant of the ROF denoising model based on that regularizer. Variable growth of the double-phase functional allows for qualitatively different behavior at image contours, which are captured by an initial ROF reconstruction step. The model is designed to reduce staircasing with respect to the classical ROF model, while preserving the edges of the image in a similar fashion. We derive a closed-form resolvent formula and adapt the primal-dual Chambolle--Pock scheme for the numerical solution of the model. We also propose a practical noise-dependent parameter prescription and evaluate its performance on synthetic and natural images over a range of noise levels. Compared to established models with similar interpretability, we observe an improved or similar performance in terms of similarity metrics SSIM, PSNR, and LPIPS, while the staircasing effect is visibly reduced. --- ### Reconsidering Requirements Engineering: Human-AI Collaboration in AI-Native Software Development Authors: Mateen Ahmed Abbasi, Petri Ihantola, Tommi Mikkonen, Niko Mäkitalo Published: 2025-10-05 Categories: cs.SE, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2510.04380v1 Requirement Engineering (RE) is the foundation of successful software development. In RE, the goal is to ensure that implemented systems satisfy stakeholder needs through rigorous requirements elicitation, validation, and evaluation processes. Despite its critical role, RE continues to face persistent challenges, such as ambiguity, conflicting stakeholder needs, and the complexity of managing evolving requirements. A common view is that Artificial Intelligence (AI) has the potential to streamline the RE process, resulting in improved efficiency, accuracy, and management actions. However, using AI also introduces new concerns, such as ethical issues, biases, and lack of transparency. This paper explores how AI can enhance traditional RE practices by automating labor-intensive tasks, supporting requirement prioritization, and facilitating collaboration between stakeholders and AI systems. The paper also describes the opportunities and challenges that AI brings to RE. In particular, the vision calls for ethical practices in AI, along with a much-enhanced collaboration between academia and industry professionals. The focus should be on creating not only powerful but also trustworthy and practical AI solutions ready to adapt to the fast-paced world of software development. --- ### Score-based Greedy Search for Structure Identification of Partially Observed Linear Causal Models Authors: Xinshuai Dong, Ignavier Ng, Haoyue Dai, Jiaqi Sun, Xiangchen Song, Peter Spirtes, Kun Zhang Published: 2025-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.04378v2 Identifying the structure of a partially observed causal system is essential to various scientific fields. Recent advances have focused on constraint-based causal discovery to solve this problem, and yet in practice these methods often face challenges related to multiple testing and error propagation. These issues could be mitigated by a score-based method and thus it has raised great attention whether there exists a score-based greedy search method that can handle the partially observed scenario. In this work, we propose the first score-based greedy search method for the identification of structure involving latent variables with identifiability guarantees. Specifically, we propose Generalized N Factor Model and establish the global consistency: the true structure including latent variables can be identified up to the Markov equivalence class by using score. We then design Latent variable Greedy Equivalence Search (LGES), a greedy search algorithm for this class of model with well-defined operators, which search very efficiently over the graph space to find the optimal structure. Our experiments on both synthetic and real-life data validate the effectiveness of our method (code will be publicly available). --- ### TCR-EML: Explainable Model Layers for TCR-pMHC Prediction Authors: Jiarui Li, Zixiang Yin, Zhengming Ding, Samuel J. Landry, Ramgopal R. Mettu Published: 2025-10-05 Categories: q-bio.QM, cs.CE, cs.LG arXiv: https://arxiv.org/abs/2510.04377v2 T cell receptor (TCR) recognition of peptide-MHC (pMHC) complexes is a central component of adaptive immunity, with implications for vaccine design, cancer immunotherapy, and autoimmune disease. While recent advances in machine learning have improved prediction of TCR-pMHC binding, the most effective approaches are black-box transformer models that cannot provide a rationale for predictions. Post-hoc explanation methods can provide insight with respect to the input but do not explicitly model biochemical mechanisms (e.g. known binding regions), as in TCR-pMHC binding. ``Explain-by-design'' models (i.e., with architectural components that can be examined directly after training) have been explored in other domains, but have not been used for TCR-pMHC binding. We propose explainable model layers (TCR-EML) that can be incorporated into protein-language model backbones for TCR-pMHC modeling. Our approach uses prototype layers for amino acid residue contacts drawn from known TCR-pMHC binding mechanisms, enabling high-quality explanations for predicted TCR-pMHC binding. Experiments of our proposed method on large-scale datasets demonstrate competitive predictive accuracy and generalization, and evaluation on the TCR-XAI benchmark demonstrates improved explainability compared with existing approaches. --- ### Categorical Invariants of Learning Dynamics Authors: Abdulrahman Tamim Published: 2025-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.04376v1 Neural network training is typically viewed as gradient descent on a loss surface. We propose a fundamentally different perspective: learning is a structure-preserving transformation (a functor L) between the space of network parameters (Param) and the space of learned representations (Rep). This categorical framework reveals that different training runs producing similar test performance often belong to the same homotopy class (continuous deformation family) of optimization paths. We show experimentally that networks converging via homotopic trajectories generalize within 0.5% accuracy of each other, while non-homotopic paths differ by over 3%. The theory provides practical tools: persistent homology identifies stable minima predictive of generalization (R^2 = 0.82 correlation), pullback constructions formalize transfer learning, and 2-categorical structures explain when different optimization algorithms yield functionally equivalent models. These categorical invariants offer both theoretical insight into why deep learning works and concrete algorithmic principles for training more robust networks. --- ### Adaptive Weighted Loss for Sequential Recommendations on Sparse Domains Authors: Akshay Mittal, Vinay Venkatesh, Krishna Kandi, Shalini Sudarshan Published: 2025-10-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.04375v1 The effectiveness of single-model sequential recommendation architectures, while scalable, is often limited when catering to "power users" in sparse or niche domains. Our previous research, PinnerFormerLite, addressed this by using a fixed weighted loss to prioritize specific domains. However, this approach can be sub-optimal, as a single, uniform weight may not be sufficient for domains with very few interactions, where the training signal is easily diluted by the vast, generic dataset. This paper proposes a novel, data-driven approach: a Dynamic Weighted Loss function with comprehensive theoretical foundations and extensive empirical validation. We introduce an adaptive algorithm that adjusts the loss weight for each domain based on its sparsity in the training data, assigning a higher weight to sparser domains and a lower weight to denser ones. This ensures that even rare user interests contribute a meaningful gradient signal, preventing them from being overshadowed. We provide rigorous theoretical analysis including convergence proofs, complexity analysis, and bounds analysis to establish the stability and efficiency of our approach. Our comprehensive empirical validation across four diverse datasets (MovieLens, Amazon Electronics, Yelp Business, LastFM Music) with state-of-the-art baselines (SIGMA, CALRec, SparseEnNet) demonstrates that this dynamic weighting system significantly outperforms all comparison methods, particularly for sparse domains, achieving substantial lifts in key metrics like Recall at 10 and NDCG at 10 while maintaining performance on denser domains and introducing minimal computational overhead. --- ### GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks Authors: Tejal Patwardhan, Rachel Dias, Elizabeth Proehl, Grace Kim, Michele Wang, Olivia Watkins, Simón Posada Fishman, Marwan Aljubeh, Phoebe Thacker, Laurance Fauconnet, Natalie S. Kim, Patrick Chao et al. Published: 2025-10-05 Categories: cs.LG, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2510.04374v1 We introduce GDPval, a benchmark evaluating AI model capabilities on real-world economically valuable tasks. GDPval covers the majority of U.S. Bureau of Labor Statistics Work Activities for 44 occupations across the top 9 sectors contributing to U.S. GDP (Gross Domestic Product). Tasks are constructed from the representative work of industry professionals with an average of 14 years of experience. We find that frontier model performance on GDPval is improving roughly linearly over time, and that the current best frontier models are approaching industry experts in deliverable quality. We analyze the potential for frontier models, when paired with human oversight, to perform GDPval tasks cheaper and faster than unaided experts. We also demonstrate that increased reasoning effort, increased task context, and increased scaffolding improves model performance on GDPval. Finally, we open-source a gold subset of 220 tasks and provide a public automated grading service at evals.openai.com to facilitate future research in understanding real-world model capabilities. --- ### JEF-Hinter: Leveraging Offline Knowledge for Improving Web Agents Adaptation Authors: Hadi Nekoei, Aman Jaiswal, Patrice Bechard, Oleh Shliazhko, Orlando Marquez Ayala, Mathieu Reymond, Massimo Caccia, Alexandre Drouin, Sarath Chandar, Alexandre Lacoste Published: 2025-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2510.04373v2 Large language model (LLM) agents perform well in sequential decision-making tasks, but improving them on unfamiliar domains often requires costly online interactions or fine-tuning on large expert datasets. These strategies are impractical for closed-source models and expensive for open-source ones, with risks of catastrophic forgetting. Offline trajectories offer reusable knowledge, yet demonstration-based methods struggle because raw traces are long, noisy, and tied to specific tasks. We present Just-in-time Episodic Feedback Hinter (JEF-Hinter), an agentic system that distills offline traces into compact, context-aware hints. A zooming mechanism highlights decisive steps in long trajectories, capturing both strategies and pitfalls. Unlike prior methods, JEF-Hinter leverages both successful and failed trajectories, extracting guidance even when only failure data is available, while supporting parallelized hint generation and benchmark-independent prompting. At inference, a retriever selects relevant hints for the current state, providing targeted guidance with transparency and traceability. Experiments on MiniWoB++, WorkArena-L1, and WebArena-Lite show that JEF-Hinter consistently outperforms strong baselines, including human- and document-based hints. --- ### Speculative Actions: A Lossless Framework for Faster Agentic Systems Authors: Naimeng Ye, Arnav Ahuja, Georgios Liargkovas, Yunan Lu, Kostis Kaffes, Tianyi Peng Published: 2025-10-05 Categories: cs.AI, cs.DC, cs.MA arXiv: https://arxiv.org/abs/2510.04371v2 AI agents are increasingly deployed in complex, interactive environments, yet their runtime remains a major bottleneck for training, evaluation, and real-world use. Typical agent behavior unfolds sequentially, with each action requiring an API call that can incur substantial latency. For example, a game of chess between two state-of-the-art agents can take hours. We introduce Speculative Actions, a lossless acceleration framework for general agentic systems. Inspired by speculative execution in microprocessors and speculative decoding in LLM inference, our method uses faster models to predict likely future actions and execute them in parallel, committing only when predictions match. We evaluate speculative actions across gaming, e-commerce, and web search environments, and additionally study a lossy extension in an operating systems setting. Across domains, we achieve up to 55% next-action prediction accuracy, translating into up to 20% latency reductions. Finally, we present a cost-latency analysis that formalizes the tradeoff between speculative breadth and time savings. This analysis enables principled tuning and selective branch launching to ensure that multi-branch speculation delivers practical speedups without prohibitive cost growth. --- ### The method of the approximate inverse for limited-angle CT Authors: Bernadette Hahn, Gael Rigaud, Richard Schmähl Published: 2025-10-05 Categories: eess.IV, cs.CV, math.NA arXiv: https://arxiv.org/abs/2510.04369v1 Limited-angle computerized tomography stands for one of the most difficult challenges in imaging. Although it opens the way to faster data acquisition in industry and less dangerous scans in medicine, standard approaches, such as the filtered backprojection (FBP) algorithm or the widely used total-variation functional, often produce various artefacts that hinder the diagnosis. With the rise of deep learning, many modern techniques have proven themselves successful in removing such artefacts but at the cost of large datasets. In this paper, we propose a new model-driven approach based on the method of the approximate inverse, which could serve as new starting point for learning strategies in the future. In contrast to FBP-type approaches, our reconstruction step consists in evaluating linear functionals on the measured data using reconstruction kernels that are precomputed as solution of an auxiliary problem. With this problem being uniquely solvable, the derived limited-angle reconstruction kernel (LARK) is able to fully reconstruct the object without the well-known streak artefacts, even for large limited angles. However, it inherits severe ill-conditioning which leads to a different kind of artefacts arising from the singular functions of the limited-angle Radon transform. The problem becomes particularly challenging when working on semi-discrete (real or analytical) measurements. We develop a general regularization strategy, named constrained limited-angle reconstruction kernel (CLARK), by combining spectral filter, the method of the approximate inverse and custom edge-preserving denoising in order to stabilize the whole process. We further derive and interpret error estimates for the application on real, i.e. semi-discrete, data and we validate our approach on synthetic and real data. --- ### NegotiationGym: Self-Optimizing Agents in a Multi-Agent Social Simulation Environment Authors: Shashank Mangla, Chris Hokamp, Jack Boylan, Demian Gholipour Ghalandari, Yuuv Jauhari, Lauren Cassidy, Oisin Duffy Published: 2025-10-05 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2510.04368v1 We design and implement NegotiationGym, an API and user interface for configuring and running multi-agent social simulations focused upon negotiation and cooperation. The NegotiationGym codebase offers a user-friendly, configuration-driven API that enables easy design and customization of simulation scenarios. Agent-level utility functions encode optimization criteria for each agent, and agents can self-optimize by conducting multiple interaction rounds with other agents, observing outcomes, and modifying their strategies for future rounds. --- ### Quantifying Ambiguity in Categorical Annotations: A Measure and Statistical Inference Framework Authors: Christopher Klugmann, Daniel Kondermann Published: 2025-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.04366v1 Human-generated categorical annotations frequently produce empirical response distributions (soft labels) that reflect ambiguity rather than simple annotator error. We introduce an ambiguity measure that maps a discrete response distribution to a scalar in the unit interval, designed to quantify aleatoric uncertainty in categorical tasks. The measure bears a close relationship to quadratic entropy (Gini-style impurity) but departs from those indices by treating an explicit "can't solve" category asymmetrically, thereby separating uncertainty arising from class-level indistinguishability from uncertainty due to explicit unresolvability. We analyze the measure's formal properties and contrast its behavior with a representative ambiguity measure from the literature. Moving beyond description, we develop statistical tools for inference: we propose frequentist point estimators for population ambiguity and derive the Bayesian posterior over ambiguity induced by Dirichlet priors on the underlying probability vector, providing a principled account of epistemic uncertainty. Numerical examples illustrate estimation, calibration, and practical use for dataset-quality assessment and downstream machine-learning workflows. --- ### Diffusion^2: Dual Diffusion Model with Uncertainty-Aware Adaptive Noise for Momentary Trajectory Prediction Authors: Yuhao Luo, Yuang Zhang, Kehua Chen, Xinyu Zheng, Shucheng Zhang, Sikai Chen, Yinhai Wang Published: 2025-10-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.04365v1 Accurate pedestrian trajectory prediction is crucial for ensuring safety and efficiency in autonomous driving and human-robot interaction scenarios. Earlier studies primarily utilized sufficient observational data to predict future trajectories. However, in real-world scenarios, such as pedestrians suddenly emerging from blind spots, sufficient observational data is often unavailable (i.e. momentary trajectory), making accurate prediction challenging and increasing the risk of traffic accidents. Therefore, advancing research on pedestrian trajectory prediction under extreme scenarios is critical for enhancing traffic safety. In this work, we propose a novel framework termed Diffusion^2, tailored for momentary trajectory prediction. Diffusion^2 consists of two sequentially connected diffusion models: one for backward prediction, which generates unobserved historical trajectories, and the other for forward prediction, which forecasts future trajectories. Given that the generated unobserved historical trajectories may introduce additional noise, we propose a dual-head parameterization mechanism to estimate their aleatoric uncertainty and design a temporally adaptive noise module that dynamically modulates the noise scale in the forward diffusion process. Empirically, Diffusion^2 sets a new state-of-the-art in momentary trajectory prediction on ETH/UCY and Stanford Drone datasets. --- ### MacroBench: A Novel Testbed for Web Automation Scripts via Large Language Models Authors: Hyunjun Kim, Sejong Kim Published: 2025-10-05 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.04363v2 We introduce MacroBench, a code-first benchmark that evaluates whether LLMs can synthesize reusable browser-automation programs (macros) from natural-language goals by reading HTML/DOM and emitting Selenium. MacroBench instantiates seven self-hosted sites covering 681 tasks across interaction complexity and targeting difficulty. Our end-to-end protocol validates generated code via static checks, sandboxed execution, and outcome verification (DOM assertions, database snapshots), and includes a safety suite for scraping, spam/abuse, and credential/privacy prompts. Across 2,636 model-task runs, we observe stratified success: GPT-4o-mini (96.8%), GPT-4o (95.3%), Gemini (89.0%), DeepSeek (83.4%). Models handle simple tasks reliably (91.7%) but fail on complex workflows (0.0%), and none meet production-quality coding practices despite functional completion. We release our complete benchmark pipeline, evaluation framework, and experimental results at https://github.com/hyunjun1121/MacroBench to enable reproducible assessment of macro synthesis for web automation. --- ### Score-based generative emulation of impact-relevant Earth system model outputs Authors: Shahine Bouabid, Andre Nogueira Souza, Raffaele Ferrari Published: 2025-10-05 Categories: physics.ao-ph, stat.AP, stat.ML arXiv: https://arxiv.org/abs/2510.04358v2 Policy targets evolve faster than the Coupled Model Intercomparison Project cycles, complicating adaptation and mitigation planning that must often contend with outdated projections. Climate model output emulators address this gap by offering inexpensive surrogates that can rapidly explore alternative futures while staying close to Earth System Model (ESM) behavior. The focus is on emulators designed to provide inputs to impact models. Using monthly ESM fields of near-surface temperature, precipitation, relative humidity, and wind speed, it is shown that deep generative models have the potential to model the joint distribution of variables relevant for impacts. The specific model proposed uses score-based diffusion on a spherical mesh and runs on a single mid-range graphical processing unit. A thorough suite of diagnostics is introduced to compare emulator outputs with their parent ESMs, including their probability densities, cross-variable correlations, time of emergence, or tail behavior. The emulator performance is evaluated across three distinct ESMs in both pre-industrial and forced regimes. The results show that the emulator produces distributions that closely match the ESM outputs and captures key forced responses. They also reveal important failure cases, notably for variables with a strong regime shift in the seasonal cycle. Although not a perfect match to the ESM, the inaccuracies of the emulator are small relative to the magnitude of internal variability in ESM projections. This suggests that the generative emulators can be useful in supporting impact assessment. Priorities for future development toward daily resolution, finer spatial scales, and bias-aware training are discussed. Code is made available at https://github.com/shahineb/climemu. --- ### From News to Returns: A Granger-Causal Hypergraph Transformer on the Sphere Authors: Anoushka Harit, Zhongtian Sun, Jongmin Yu Published: 2025-10-05 Categories: cs.LG, q-fin.CP arXiv: https://arxiv.org/abs/2510.04357v1 We propose the Causal Sphere Hypergraph Transformer (CSHT), a novel architecture for interpretable financial time-series forecasting that unifies \emph{Granger-causal hypergraph structure}, \emph{Riemannian geometry}, and \emph{causally masked Transformer attention}. CSHT models the directional influence of financial news and sentiment on asset returns by extracting multivariate Granger-causal dependencies, which are encoded as directional hyperedges on the surface of a hypersphere. Attention is constrained via angular masks that preserve both temporal directionality and geometric consistency. Evaluated on S\&P 500 data from 2018 to 2023, including the 2020 COVID-19 shock, CSHT consistently outperforms baselines across return prediction, regime classification, and top-asset ranking tasks. By enforcing predictive causal structure and embedding variables in a Riemannian manifold, CSHT delivers both \emph{robust generalisation across market regimes} and \emph{transparent attribution pathways} from macroeconomic events to stock-level responses. These results suggest that CSHT is a principled and practical solution for trustworthy financial forecasting under uncertainty. --- ### Quantizer Design for Finite Model Approximations, Model Learning, and Quantized Q-Learning for MDPs with Unbounded Spaces Authors: Osman Bicer, Ali D. Kara, Serdar Yuksel Published: 2025-10-05 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2510.04355v2 In this paper, for Markov decision processes (MDPs) with unbounded state spaces we present refined upper bounds presented in (Kara et. al. JMLR'23) on finite model approximation errors via optimizing the quantizers used for finite model approximations. We also consider implications on quantizer design for quantized Q-learning and empirical model learning, and the performance of policies obtained via Q-learning where the quantized state is treated as the state itself. We highlight the distinctions between planning, where approximating MDPs can be independently designed, and learning (either via Q-learning or empirical model learning), where approximating MDPs are restricted to be defined by invariant measures of Markov chains under exploration policies, leading to significant subtleties on quantizer design performance, even though asymptotic near optimality can be established under both setups. In particular, under Lyapunov growth conditions, we obtain explicit upper bounds which decay to zero as the number of bins approaches infinity --- ### Reliable and Scalable Robot Policy Evaluation with Imperfect Simulators Authors: Apurva Badithela, David Snyder, Lihan Zha, Joseph Mikhail, Matthew O'Kelly, Anushri Dixit, Anirudha Majumdar Published: 2025-10-05 Categories: cs.RO, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2510.04354v1 Rapid progress in imitation learning, foundation models, and large-scale datasets has led to robot manipulation policies that generalize to a wide-range of tasks and environments. However, rigorous evaluation of these policies remains a challenge. Typically in practice, robot policies are often evaluated on a small number of hardware trials without any statistical assurances. We present SureSim, a framework to augment large-scale simulation with relatively small-scale real-world testing to provide reliable inferences on the real-world performance of a policy. Our key idea is to formalize the problem of combining real and simulation evaluations as a prediction-powered inference problem, in which a small number of paired real and simulation evaluations are used to rectify bias in large-scale simulation. We then leverage non-asymptotic mean estimation algorithms to provide confidence intervals on mean policy performance. Using physics-based simulation, we evaluate both diffusion policy and multi-task fine-tuned \(π_0\) on a joint distribution of objects and initial conditions, and find that our approach saves over \(20-25\%\) of hardware evaluation effort to achieve similar bounds on policy performance. --- ### Challenge on Optimization of Context Collection for Code Completion Authors: Dmitry Ustalov, Egor Bogomolov, Alexander Bezzubov, Yaroslav Golubev, Evgeniy Glukhov, Georgii Levtsov, Vladimir Kovalenko Published: 2025-10-05 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.04349v1 The rapid advancement of workflows and methods for software engineering using AI emphasizes the need for a systematic evaluation and analysis of their ability to leverage information from entire projects, particularly in large code bases. In this challenge on optimization of context collection for code completion, organized by JetBrains in collaboration with Mistral AI as part of the ASE 2025 conference, participants developed efficient mechanisms for collecting context from source code repositories to improve fill-in-the-middle code completions for Python and Kotlin. We constructed a large dataset of real-world code in these two programming languages using permissively licensed open-source projects. The submissions were evaluated based on their ability to maximize completion quality for multiple state-of-the-art neural models using the chrF metric. During the public phase of the competition, nineteen teams submitted solutions to the Python track and eight teams submitted solutions to the Kotlin track. In the private phase, six teams competed, of which five submitted papers to the workshop. --- ### Unmasking Backdoors: An Explainable Defense via Gradient-Attention Anomaly Scoring for Pre-trained Language Models Authors: Anindya Sundar Das, Kangjie Chen, Monowar Bhuyan Published: 2025-10-05 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2510.04347v2 Pre-trained language models have achieved remarkable success across a wide range of natural language processing (NLP) tasks, particularly when fine-tuned on large, domain-relevant datasets. However, they remain vulnerable to backdoor attacks, where adversaries embed malicious behaviors using trigger patterns in the training data. These triggers remain dormant during normal usage, but, when activated, can cause targeted misclassifications. In this work, we investigate the internal behavior of backdoored pre-trained encoder-based language models, focusing on the consistent shift in attention and gradient attribution when processing poisoned inputs; where the trigger token dominates both attention and gradient signals, overriding the surrounding context. We propose an inference-time defense that constructs anomaly scores by combining token-level attention and gradient information. Extensive experiments on text classification tasks across diverse backdoor attack scenarios demonstrate that our method significantly reduces attack success rates compared to existing baselines. Furthermore, we provide an interpretability-driven analysis of the scoring mechanism, shedding light on trigger localization and the robustness of the proposed defense. --- ### Environment-Aware Indoor LoRaWAN Path Loss: Parametric Regression Comparisons, Shadow Fading, and Calibrated Fade Margins Authors: Nahshon Mokua Obiri, Kristof Van Laerhoven Published: 2025-10-05 Categories: cs.NI, cs.LG, eess.SP, math.NA arXiv: https://arxiv.org/abs/2510.04346v2 Indoor long range wide area network (LoRaWAN) propagation is shaped by structural and time-varying environmental factors, which limit single-slope log-distance models and the standard log-normal shadowing assumption. We propose an environment-conditioned path loss framework that augments a log-distance multi-wall baseline with co-recorded environmental covariates (relative humidity, temperature, carbon dioxide, particulate matter, and barometric pressure) and receiver-reported signal-to-noise, and we validate both the mean and the residual law statistically. The approach is evaluated on a 12-month campaign in an eighth-floor office (240 m^2) using time-blocked 5-fold cross-validation and a chronological hold-out. Across parametric regressors (regularized multiple linear regression (MLR), conjugate Bayesian linear regression, and a selective quadratic MLR extension on continuous predictors), the selective polynomial mean improves out-of-sample accuracy, reducing cross-validated root mean square error from 8.23 to 7.38 dB and increasing R^2 from 0.81 to 0.84. Out-of-fold (OOF) residuals are distinctly non-Gaussian and are best summarized by a compact 3-component Gaussian mixture with a sharp core and a light, broad tail. Finally, we translate prediction error into reliability by prescribing the fade margin as the upper-tail percentile of OOF errors, attaching moving-block bootstrap uncertainty, and validating the resulting outage on a held-out set. At a 1% outage target (99% reliability), the polynomial model requires 25.73 dB versus 27.79 to 28.05 dB for linear baselines, enabling tighter indoor massive Internet of Things link budgets aligned with sixth-generation reliability targets under energy constraints. --- ### Learning to Predict Chaos: Curriculum-Driven Training for Robust Forecasting of Chaotic Dynamics Authors: Harshil Vejendla Published: 2025-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.04342v1 Forecasting chaotic systems is a cornerstone challenge in many scientific fields, complicated by the exponential amplification of even infinitesimal prediction errors. Modern machine learning approaches often falter due to two opposing pitfalls: over-specializing on a single, well-known chaotic system (e.g., Lorenz-63), which limits generalizability, or indiscriminately mixing vast, unrelated time-series, which prevents the model from learning the nuances of any specific dynamical regime. We propose Curriculum Chaos Forecasting (CCF), a training paradigm that bridges this gap. CCF organizes training data based on fundamental principles of dynamical systems theory, creating a curriculum that progresses from simple, periodic behaviors to highly complex, chaotic dynamics. We quantify complexity using the largest Lyapunov exponent and attractor dimension, two well-established metrics of chaos. By first training a sequence model on predictable systems and gradually introducing more chaotic trajectories, CCF enables the model to build a robust and generalizable representation of dynamical behaviors. We curate a library of over 50 synthetic ODE/PDE systems to build this curriculum. Our experiments show that pre-training with CCF significantly enhances performance on unseen, real-world benchmarks. On datasets including Sunspot numbers, electricity demand, and human ECG signals, CCF extends the valid prediction horizon by up to 40% compared to random-order training and more than doubles it compared to training on real-world data alone. We demonstrate that this benefit is consistent across various neural architectures (GRU, Transformer) and provide extensive ablations to validate the importance of the curriculum's structure. --- ### Critical appraisal of artificial intelligence for rare-event recognition: principles and pharmacovigilance case studies Authors: G. Niklas Noren, Eva-Lisa Meldau, Johan Ellenius Published: 2025-10-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.04341v1 Many high-stakes AI applications target low-prevalence events, where apparent accuracy can conceal limited real-world value. Relevant AI models range from expert-defined rules and traditional machine learning to generative LLMs constrained for classification. We outline key considerations for critical appraisal of AI in rare-event recognition, including problem framing and test set design, prevalence-aware statistical evaluation, robustness assessment, and integration into human workflows. In addition, we propose an approach to structured case-level examination (SCLE), to complement statistical performance evaluation, and a comprehensive checklist to guide procurement or development of AI models for rare-event recognition. We instantiate the framework in pharmacovigilance, drawing on three studies: rule-based retrieval of pregnancy-related reports; duplicate detection combining machine learning with probabilistic record linkage; and automated redaction of person names using an LLM. We highlight pitfalls specific to the rare-event setting including optimism from unrealistic class balance and lack of difficult positive controls in test sets - and show how cost-sensitive targets align model performance with operational value. While grounded in pharmacovigilance practice, the principles generalize to domains where positives are scarce and error costs may be asymmetric. --- ### Inoculation Prompting: Eliciting traits from LLMs during training can suppress them at test-time Authors: Daniel Tan, Anders Woodruff, Niels Warncke, Arun Jose, Maxime Riché, David Demitri Africa, Mia Taylor Published: 2025-10-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.04340v4 Language model finetuning often results in learning undesirable traits in combination with desired ones. To address this, we propose inoculation prompting: modifying finetuning data by prepending a short system-prompt instruction that deliberately elicits the undesirable trait. At test time, we evaluate without the instruction; inoculated models have much lower expression of the trait than models trained with unmodified training data. Inoculation is selective: in a toy setting where assistant responses are always in Spanish and ALL-CAPS, an appropriate inoculation (e.g., ``You always speak in Spanish.'') teaches the model to capitalize responses while still responding in English. We find that inoculation is also effective across several additional settings: reducing emergent misalignment (EM) from task-specific finetuning, defending against backdoor injections, and mitigating the transmission of traits via subliminal learning. Follow-up analysis suggests a mechanism: making a trait less surprising via inoculation reduces optimization pressure to globally update the model, thereby reducing the degree of generalization. Our analysis relates to prior work on EM: inoculation explains prior findings that educational contexts mitigate EM from insecure code. Beyond demonstrating a simple and effective technique for selective learning, our results contribute to a better conceptual understanding of how and why language models generalize. --- ### Pitch-Conditioned Instrument Sound Synthesis From an Interactive Timbre Latent Space Authors: Christian Limberg, Fares Schulz, Zhe Zhang, Stefan Weinzierl Published: 2025-10-05 Categories: cs.SD, cs.AI, cs.LG, eess.AS, eess.SP arXiv: https://arxiv.org/abs/2510.04339v1 This paper presents a novel approach to neural instrument sound synthesis using a two-stage semi-supervised learning framework capable of generating pitch-accurate, high-quality music samples from an expressive timbre latent space. Existing approaches that achieve sufficient quality for music production often rely on high-dimensional latent representations that are difficult to navigate and provide unintuitive user experiences. We address this limitation through a two-stage training paradigm: first, we train a pitch-timbre disentangled 2D representation of audio samples using a Variational Autoencoder; second, we use this representation as conditioning input for a Transformer-based generative model. The learned 2D latent space serves as an intuitive interface for navigating and exploring the sound landscape. We demonstrate that the proposed method effectively learns a disentangled timbre space, enabling expressive and controllable audio generation with reliable pitch conditioning. Experimental results show the model's ability to capture subtle variations in timbre while maintaining a high degree of pitch accuracy. The usability of our method is demonstrated in an interactive web application, highlighting its potential as a step towards future music production environments that are both intuitive and creatively empowering: https://pgesam.faresschulz.com --- ### Evaluation of Clinical Trials Reporting Quality using Large Language Models Authors: Mathieu Laï-king, Patrick Paroubek Published: 2025-10-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.04338v1 Reporting quality is an important topic in clinical trial research articles, as it can impact clinical decisions. In this article, we test the ability of large language models to assess the reporting quality of this type of article using the Consolidated Standards of Reporting Trials (CONSORT). We create CONSORT-QA, an evaluation corpus from two studies on abstract reporting quality with CONSORT-abstract standards. We then evaluate the ability of different large generative language models (from the general domain or adapted to the biomedical domain) to correctly assess CONSORT criteria with different known prompting methods, including Chain-of-thought. Our best combination of model and prompting method achieves 85% accuracy. Using Chain-of-thought adds valuable information on the model's reasoning for completing the task. --- ### RAP: 3D Rasterization Augmented End-to-End Planning Authors: Lan Feng, Yang Gao, Eloi Zablocki, Quanyi Li, Wuyang Li, Sichao Liu, Matthieu Cord, Alexandre Alahi Published: 2025-10-05 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2510.04333v2 Imitation learning for end-to-end driving trains policies only on expert demonstrations. Once deployed in a closed loop, such policies lack recovery data: small mistakes cannot be corrected and quickly compound into failures. A promising direction is to generate alternative viewpoints and trajectories beyond the logged path. Prior work explores photorealistic digital twins via neural rendering or game engines, but these methods are prohibitively slow and costly, and thus mainly used for evaluation. In this work, we argue that photorealism is unnecessary for training end-to-end planners. What matters is semantic fidelity and scalability: driving depends on geometry and dynamics, not textures or lighting. Motivated by this, we propose 3D Rasterization, which replaces costly rendering with lightweight rasterization of annotated primitives, enabling augmentations such as counterfactual recovery maneuvers and cross-agent view synthesis. To transfer these synthetic views effectively to real-world deployment, we introduce a Raster-to-Real feature-space alignment that bridges the sim-to-real gap. Together, these components form Rasterization Augmented Planning (RAP), a scalable data augmentation pipeline for planning. RAP achieves state-of-the-art closed-loop robustness and long-tail generalization, ranking first on four major benchmarks: NAVSIM v1/v2, Waymo Open Dataset Vision-based E2E Driving, and Bench2Drive. Our results show that lightweight rasterization with feature alignment suffices to scale E2E training, offering a practical alternative to photorealistic rendering. Project page: https://alan-lanfeng.github.io/RAP/. --- ### DoRAN: Stabilizing Weight-Decomposed Low-Rank Adaptation via Noise Injection and Auxiliary Networks Authors: Nghiem T. Diep, Hien Dang, Tuan Truong, Tan Dinh, Huy Nguyen, Nhat Ho Published: 2025-10-05 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2510.04331v2 Parameter-efficient fine-tuning (PEFT) methods have become the standard paradigm for adapting large-scale models. Among these techniques, Weight-Decomposed Low-Rank Adaptation (DoRA) has been shown to improve both the learning capacity and training stability of the Low-Rank Adaptation (LoRA) method by explicitly decomposing pre-trained weights into magnitude and directional components. In this work, we propose DoRAN, a new technique designed to stabilize training and boost the sample efficiency of DoRA. Our framework introduces two key components: (i) the injection of learnable noise into the denominator of DoRA weight decomposition, which serves as an adaptive regularizer to mitigate instabilities and improve the estimation rate of low-rank matrices; and (ii) the replacement of static low-rank matrices with auxiliary networks that generate them dynamically, enabling parameter coupling between the query and value projection matrices, leading to improved sample efficiency both theoretically and empirically. Comprehensive experiments on vision and language benchmarks show that DoRAN consistently outperforms LoRA, DoRA, and other PEFT baselines, underscoring the effectiveness of combining noise-based regularization with network-based parameter generation. --- ### Arithmetic-Mean $μ$P for Modern Architectures: A Unified Learning-Rate Scale for CNNs and ResNets Authors: Haosong Zhang, Shenxi Wu, Yichi Zhang, Xi Chen, Wei Lin Published: 2025-10-05 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.04327v2 Choosing an appropriate learning rate remains a key challenge in scaling depth of modern deep networks. The classical maximal update parameterization ($μ$P) enforces a fixed per-layer update magnitude, which is well suited to homogeneous multilayer perceptrons (MLPs) but becomes ill-posed in heterogeneous architectures where residual accumulation and convolutions introduce imbalance across layers. We introduce Arithmetic-Mean $μ$P (AM-$μ$P), which constrains not each individual layer but the network-wide average one-step pre-activation second moment to a constant scale. Combined with a residual-aware He fan-in initialization - scaling residual-branch weights by the number of blocks ($\mathrm{Var}(W)=c/(K\cdot \mathrm{fan\text{-}in})$) - AM-$μ$P yields width-robust depth laws that transfer consistently across depths. We prove that, for one- and two-dimensional convolutional networks, the maximal-update learning rate satisfies $η^\star(L)\propto L^{-3/2}$; with zero padding, boundary effects are constant-level as $N\gg k$. For standard residual networks with general conv+MLP blocks, we establish $η^\star(L)=Θ(L^{-3/2})$, with $L$ the minimal depth. Empirical results across a range of depths confirm the $-3/2$ scaling law and enable zero-shot learning-rate transfer, providing a unified and practical LR principle for convolutional and deep residual networks without additional tuning overhead. --- ### FoilDiff: A Hybrid Transformer Backbone for Diffusion-based Modelling of 2D Airfoil Flow Fields Authors: Kenechukwu Ogbuagu, Sepehr Maleki, Giuseppe Bruni, Senthil Krishnababu Published: 2025-10-05 Categories: cs.LG, physics.flu-dyn arXiv: https://arxiv.org/abs/2510.04325v2 The accurate prediction of flow fields around airfoils is crucial for aerodynamic design and optimisation. Computational Fluid Dynamics (CFD) models are effective but computationally expensive, thus inspiring the development of surrogate models to enable quicker predictions. These surrogate models can be based on deep learning architectures, such as Convolutional Neural Networks (CNNs), Graph Neural Networks (GNNs), and Diffusion Models (DMs). Diffusion models have shown significant promise in predicting complex flow fields. In this work, we propose FoilDiff, a diffusion-based surrogate model with a hybrid-backbone denoising network. This hybrid design combines the power of convolutional feature extraction and transformer-based global attention to generate more adaptable and accurate representations of flow structures. FoilDiff takes advantage of Denoising Diffusion Implicit Model (DDIM) sampling to optimise the efficiency of the sampling process at no additional cost to model generalisation. We used encoded representations of Reynolds number, angle of attack, and airfoil geometry to define the input space for generalisation across a wide range of aerodynamic conditions. When evaluated against state-of-the-art models, FoilDiff shows significant performance improvements, with mean prediction errors reducing by up to 85\% on the same datasets. The results have demonstrated that FoilDiff can provide both more accurate predictions and better-calibrated predictive uncertainty than existing diffusion-based models. --- ### Towards Fast Option Pricing PDE Solvers Powered by PIELM Authors: Akshay Govind Srinivasan, Anuj Jagannath Said, Sathwik Pentela, Vikas Dwivedi, Balaji Srinivasan Published: 2025-10-05 Categories: cs.CE, cs.LG, math.NA arXiv: https://arxiv.org/abs/2510.04322v1 Partial differential equation (PDE) solvers underpin modern quantitative finance, governing option pricing and risk evaluation. Physics-Informed Neural Networks (PINNs) have emerged as a promising approach for solving the forward and inverse problems of partial differential equations (PDEs) using deep learning. However they remain computationally expensive due to their iterative gradient descent based optimization and scale poorly with increasing model size. This paper introduces Physics-Informed Extreme Learning Machines (PIELMs) as fast alternative to PINNs for solving both forward and inverse problems in financial PDEs. PIELMs replace iterative optimization with a single least-squares solve, enabling deterministic and efficient training. We benchmark PIELM on the Black-Scholes and Heston-Hull-White models for forward pricing and demonstrate its capability in inverse model calibration to recover volatility and interest rate parameters from noisy data. From experiments we observe that PIELM achieve accuracy comparable to PINNs while being up to $30\times$ faster, highlighting their potential for real-time financial modeling. ---