# Research Papers — 2025-10 · bundle page 134 of 217

> Full-text markdown of papers #6651–6700 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/134
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/133.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/135.md)

### Interactive Atmospheric Composition Emulation for Next-Generation Earth System Models

Authors: Seyed Mohammad Hassan Erfani, Kara Lamb, Susanne Bauer, Kostas Tsigaridis, Marcus van Lier-Walqui, Gavin Schmidt
Published: 2025-10-12
Categories: physics.ao-ph, cs.LG
arXiv: https://arxiv.org/abs/2510.10654v1

Interactive composition simulations in Earth System Models (ESMs) are computationally expensive as they transport numerous gaseous and aerosol tracers at each timestep. This limits higher-resolution transient climate simulations with current computational resources. ESMs like NASA GISS-ModelE3 (ModelE) often use pre-computed monthly-averaged atmospheric composition concentrations (Non-Interactive Tracers or NINT) to reduce computational costs. While NINT significantly cuts computations, it fails to capture real-time feedback between aerosols and other climate processes by relying on pre-calculated fields. We extended the ModelE NINT version using machine learning (ML) to create Smart NINT, which emulates interactive emissions. Smart NINT interactively calculates concentrations using ML with surface emissions and meteorological data as inputs, avoiding full physics parameterizations. Our approach utilizes a spatiotemporal architecture that possesses a well-matched inductive bias to effectively capture the spatial and temporal dependencies in tracer evolution. Input data processed through the first 20 vertical levels (from the surface up to 656 hPa) using the ModelE OMA scheme. This vertical range covers nearly the entire BCB concentration distribution in the troposphere, where significant variation on short time horizons due to surface-level emissions is observed. Our evaluation shows excellent model performance with R-squared values of 0.92 and Pearson-r of 0.96 at the first pressure level. This high performance continues through level 15 (808.5 hPa), then gradually decreases as BCB concentrations drop significantly. The model maintains acceptable performance even when tested on data from entirely different periods outside the training domain, which is a crucial capability for climate modeling applications requiring reliable long-term projections.

---

### A Machine Learning Perspective on Automated Driving Corner Cases

Authors: Sebastian Schmidt, Julius Körner, Stephan Günnemann
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10653v1

For high-stakes applications, like autonomous driving, a safe operation is necessary to prevent harm, accidents, and failures. Traditionally, difficult scenarios have been categorized into corner cases and addressed individually. However, this example-based categorization is not scalable and lacks a data coverage perspective, neglecting the generalization to training data of machine learning models. In our work, we propose a novel machine learning approach that takes the underlying data distribution into account. Based on our novel perspective, we present a framework for effective corner case recognition for perception on individual samples. In our evaluation, we show that our approach (i) unifies existing scenario-based corner case taxonomies under a distributional perspective, (ii) achieves strong performance on corner case detection tasks across standard benchmarks for which we extend established out-of-distribution detection benchmarks, and (iii) enables analysis of combined corner cases via a newly introduced fog-augmented Lost & Found dataset. These results provide a principled basis for corner case recognition, underlining our manual specification-free definition.

---

### DEMO: Disentangled Motion Latent Flow Matching for Fine-Grained Controllable Talking Portrait Synthesis

Authors: Peiyin Chen, Zhuowei Yang, Hui Feng, Sheng Jiang, Rui Yan
Published: 2025-10-12
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.10650v1

Audio-driven talking-head generation has advanced rapidly with diffusion-based generative models, yet producing temporally coherent videos with fine-grained motion control remains challenging. We propose DEMO, a flow-matching generative framework for audio-driven talking-portrait video synthesis that delivers disentangled, high-fidelity control of lip motion, head pose, and eye gaze. The core contribution is a motion auto-encoder that builds a structured latent space in which motion factors are independently represented and approximately orthogonalized. On this disentangled motion space, we apply optimal-transport-based flow matching with a transformer predictor to generate temporally smooth motion trajectories conditioned on audio. Extensive experiments across multiple benchmarks show that DEMO outperforms prior methods in video realism, lip-audio synchronization, and motion fidelity. These results demonstrate that combining fine-grained motion disentanglement with flow-based generative modeling provides a powerful new paradigm for controllable talking-head video synthesis.

---

### Unlocking Exploration in RLVR: Uncertainty-aware Advantage Shaping for Deeper Reasoning

Authors: Can Xie, Ruotong Pan, Xiangyu Wu, Yunfei Zhang, Jiayi Fu, Tingting Gao, Guorui Zhou
Published: 2025-10-12
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.10649v2

Reinforcement Learning with Verifiable Rewards (RLVR) has shown significant promise for enhancing the reasoning capabilities of large language models (LLMs). However, prevailing algorithms like GRPO broadcast a uniform advantage signal across all tokens in a sequence. This coarse-grained approach overlooks the pivotal role of uncertain, high-stakes decisions during reasoning, leading to inefficient exploration and the well-documented problem of entropy collapse. To address this, we introduce UnCertainty-aware Advantage Shaping (UCAS), a model-free method that refines credit assignment by leveraging the model's internal uncertainty signals. UCAS operates in two stages: it first modulates the response-level advantage using a logit-space self-confidence proxy, and then applies an asymmetric token-level penalty based on raw logit certainty. This dual mechanism encourages exploration of high-uncertainty paths that yield correct answers while penalizing overconfident yet erroneous reasoning, effectively balancing the exploration-exploitation trade-off. Extensive experiments on five mathematical reasoning benchmarks show that UCAS significantly outperforms strong RLVR baselines across multiple model scales, including 1.5B and 7B. Our analysis confirms that UCAS not only achieves higher rewards but also promotes greater reasoning diversity and successfully mitigates entropy collapse. Code is available at https://github.com/xvolcano02/UCAS.

---

### JND-Guided Light-Weight Neural Pre-Filter for Perceptual Image Coding

Authors: Chenlong He, Zhijian Hao, Leilei Huang, Xiaoyang Zeng, Yibo Fan
Published: 2025-10-12
Categories: eess.IV, cs.CV, cs.MM
arXiv: https://arxiv.org/abs/2510.10648v2

Just Noticeable Distortion (JND)-guided pre-filter is a promising technique for improving the perceptual compression efficiency of image coding. However, existing methods are often computationally expensive, and the field lacks standardized benchmarks for fair comparison. To address these challenges, this paper introduces a twofold contribution. First, we develop and open-source FJNDF-Pytorch, a unified benchmark for frequency-domain JND-Guided pre-filters. Second, leveraging this platform, we propose a complete learning framework for a novel, lightweight Convolutional Neural Network (CNN). Experimental results demonstrate that our proposed method achieves state-of-the-art compression efficiency, consistently outperforming competitors across multiple datasets and encoders. In terms of computational cost, our model is exceptionally lightweight, requiring only 7.15 GFLOPs to process a 1080p image, which is merely 14.1% of the cost of recent lightweight network. Our work presents a robust, state-of-the-art solution that excels in both performance and efficiency, supported by a reproducible research platform. The open-source implementation is available at https://github.com/viplab-fudan/FJNDF-Pytorch.

---

### Trustworthy Retrosynthesis: Eliminating Hallucinations with a Diverse Ensemble of Reaction Scorers

Authors: Michal Sadowski, Tadija Radusinović, Maria Wyrzykowska, Lukasz Sztukiewicz, Jan Rzymkowski, Paweł Włodarczyk-Pruszyński, Mikołaj Sacha, Piotr Kozakowski, Ruard van Workum, Stanislaw Kamil Jastrzebski
Published: 2025-10-12
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.10645v3

Retrosynthesis is one of the domains transformed by the rise of generative models, and it is one where the problem of nonsensical or erroneous outputs (hallucinations) is particularly insidious: reliable assessment of synthetic plans is time-consuming, with automatic methods lacking. In this work, we present RetroTrim, a retrosynthesis system that successfully avoids nonsensical plans on a set of challenging drug-like targets. Compared to common baselines in the field, our system is not only the sole method that succeeds in filtering out hallucinated reactions, but it also results in the highest number of high-quality paths overall. The key insight behind RetroTrim is the combination of diverse reaction scoring strategies, based on machine learning models and existing chemical databases. We show that our scoring strategies capture different classes of hallucinations by analyzing them on a dataset of labeled retrosynthetic intermediates. This approach formed the basis of our winning solution to the Standard Industries \$1 million Retrosynthesis Challenge. To measure the performance of retrosynthesis systems, we propose a novel evaluation protocol for reactions and synthetic paths based on a structured review by expert chemists. Using this protocol, we compare systems on a set of 32 novel targets, curated to reflect recent trends in drug structures. While the insights behind our methodology are broadly applicable to retrosynthesis, our focus is on targets in the drug-like domain. By releasing our benchmark targets and the details of our evaluation protocol, we hope to inspire further research into reliable retrosynthesis.

---

### Hierarchical Optimization via LLM-Guided Objective Evolution for Mobility-on-Demand Systems

Authors: Yi Zhang, Yushen Long, Yun Ni, Liping Huang, Xiaohong Wang, Jun Liu
Published: 2025-10-12
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.10644v2

Online ride-hailing platforms aim to deliver efficient mobility-on-demand services, often facing challenges in balancing dynamic and spatially heterogeneous supply and demand. Existing methods typically fall into two categories: reinforcement learning (RL) approaches, which suffer from data inefficiency, oversimplified modeling of real-world dynamics, and difficulty enforcing operational constraints; or decomposed online optimization methods, which rely on manually designed high-level objectives that lack awareness of low-level routing dynamics. To address this issue, we propose a novel hybrid framework that integrates large language model (LLM) with mathematical optimization in a dynamic hierarchical system: (1) it is training-free, removing the need for large-scale interaction data as in RL, and (2) it leverages LLM to bridge cognitive limitations caused by problem decomposition by adaptively generating high-level objectives. Within this framework, LLM serves as a meta-optimizer, producing semantic heuristics that guide a low-level optimizer responsible for constraint enforcement and real-time decision execution. These heuristics are refined through a closed-loop evolutionary process, driven by harmony search, which iteratively adapts the LLM prompts based on feasibility and performance feedback from the optimization layer. Extensive experiments based on scenarios derived from both the New York and Chicago taxi datasets demonstrate the effectiveness of our approach, achieving an average improvement of 16% compared to state-of-the-art baselines.

---

### UniJEPA: Enhancing Robot Policy via Unified Continuous and Discrete Representation Learning

Authors: Jianke Zhang, Yucheng Hu, Yanjiang Guo, Xiaoyu Chen, Yichen Liu, Wenna Chen, Chaochao Lu, Jianyu Chen
Published: 2025-10-12
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2510.10642v3

Building generalist robot policies that can handle diverse tasks in open-ended environments is a central challenge in robotics. To leverage knowledge from large-scale pretraining, prior work (VLA) has typically built generalist policies either on top of vision-language understanding models (VLMs) or generative models. However, both semantic understanding from vision-language pretraining and visual dynamics modeling from visual-generation pretraining are crucial for embodied robots. Recent unified models of generation and understanding have demonstrated strong capabilities in both comprehension and generation through large-scale pretraining. We posit that robotic policy learning can likewise benefit from the combined strengths of understanding, planning, and continuous future representation learning. Building on this insight, we introduce UniJEPA, which acquires the ability to dynamically model high-dimensional visual features through pretraining on over 1M internet-scale instructional manipulation videos. Subsequently, UniJEPA is fine-tuned on data collected from the robot embodiment, enabling the learning of mappings from predictive representations to action tokens. Extensive experiments show our approach consistently outperforms baseline methods in terms of 9\% and 12\% across simulation environments and real-world out-of-distribution tasks.

---

### Equity-Aware Geospatial AI for Forecasting Demand-Driven Hospital Locations in Germany

Authors: Piyush Pant, Marcellius William Suntoro, Ayesha Siddiqua, Muhammad Shehryaar Sharif, Daniyal Ahmed
Published: 2025-10-12
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.10640v1

This paper presents EA-GeoAI, an integrated framework for demand forecasting and equitable hospital planning in Germany through 2030. We combine district-level demographic shifts, aging population density, and infrastructure balances into a unified Equity Index. An interpretable Agentic AI optimizer then allocates beds and identifies new facility sites to minimize unmet need under budget and travel-time constraints. This approach bridges GeoAI, long-term forecasting, and equity measurement to deliver actionable recommendations for policymakers.

---

### Automatic Piecewise Linear Regression for Predicting Student Learning Satisfaction

Authors: Haemin Choi, Gayathri Nadarajan
Published: 2025-10-12
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.10639v1

Although student learning satisfaction has been widely studied, modern techniques such as interpretable machine learning and neural networks have not been sufficiently explored. This study demonstrates that a recent model that combines boosting with interpretability, automatic piecewise linear regression(APLR), offers the best fit for predicting learning satisfaction among several state-of-the-art approaches. Through the analysis of APLR's numerical and visual interpretations, students' time management and concentration abilities, perceived helpfulness to classmates, and participation in offline courses have the most significant positive impact on learning satisfaction. Surprisingly, involvement in creative activities did not positively affect learning satisfaction. Moreover, the contributing factors can be interpreted on an individual level, allowing educators to customize instructions according to student profiles.

---

### ProteinAE: Protein Diffusion Autoencoders for Structure Encoding

Authors: Shaoning Li, Le Zhuo, Yusong Wang, Mingyu Li, Xinheng He, Fandi Wu, Hongsheng Li, Pheng-Ann Heng
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10634v1

Developing effective representations of protein structures is essential for advancing protein science, particularly for protein generative modeling. Current approaches often grapple with the complexities of the SE(3) manifold, rely on discrete tokenization, or the need for multiple training objectives, all of which can hinder the model optimization and generalization. We introduce ProteinAE, a novel and streamlined protein diffusion autoencoder designed to overcome these challenges by directly mapping protein backbone coordinates from E(3) into a continuous, compact latent space. ProteinAE employs a non-equivariant Diffusion Transformer with a bottleneck design for efficient compression and is trained end-to-end with a single flow matching objective, substantially simplifying the optimization pipeline. We demonstrate that ProteinAE achieves state-of-the-art reconstruction quality, outperforming existing autoencoders. The resulting latent space serves as a powerful foundation for a latent diffusion model that bypasses the need for explicit equivariance. This enables efficient, high-quality structure generation that is competitive with leading structure-based approaches and significantly outperforms prior latent-based methods. Code is available at https://github.com/OnlyLoveKFC/ProteinAE_v1.

---

### Collaborative Text-to-Image Generation via Multi-Agent Reinforcement Learning and Semantic Fusion

Authors: Jiabao Shi, Minfeng Qi, Lefeng Zhang, Di Wang, Yingjie Zhao, Ziying Li, Yalong Xing, Ningran Li
Published: 2025-10-12
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.10633v1

Multimodal text-to-image generation remains constrained by the difficulty of maintaining semantic alignment and professional-level detail across diverse visual domains. We propose a multi-agent reinforcement learning framework that coordinates domain-specialized agents (e.g., focused on architecture, portraiture, and landscape imagery) within two coupled subsystems: a text enhancement module and an image generation module, each augmented with multimodal integration components. Agents are trained using Proximal Policy Optimization (PPO) under a composite reward function that balances semantic similarity, linguistic visual quality, and content diversity. Cross-modal alignment is enforced through contrastive learning, bidirectional attention, and iterative feedback between text and image. Across six experimental settings, our system significantly enriches generated content (word count increased by 1614%) while reducing ROUGE-1 scores by 69.7%. Among fusion methods, Transformer-based strategies achieve the highest composite score (0.521), despite occasional stability issues. Multimodal ensembles yield moderate consistency (ranging from 0.444 to 0.481), reflecting the persistent challenges of cross-modal semantic grounding. These findings underscore the promise of collaborative, specialization-driven architectures for advancing reliable multimodal generative systems.

---

### GraphTARIF: Linear Graph Transformer with Augmented Rank and Improved Focus

Authors: Zhaolin Hu, Kun Li, Hehe Fan, Yi Yang
Published: 2025-10-12
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.10631v2

Linear attention mechanisms have emerged as efficient alternatives to full self-attention in Graph Transformers, offering linear time complexity. However, existing linear attention models often suffer from a significant drop in expressiveness due to low-rank projection structures and overly uniform attention distributions. We theoretically prove that these properties reduce the class separability of node representations, limiting the model's classification ability. To address this, we propose a novel hybrid framework that enhances both the rank and focus of attention. Specifically, we enhance linear attention by attaching a gated local graph network branch to the value matrix, thereby increasing the rank of the resulting attention map. Furthermore, to alleviate the excessive smoothing effect inherent in linear attention, we introduce a learnable log-power function into the attention scores to reduce entropy and sharpen focus. We theoretically show that this function decreases entropy in the attention distribution, enhancing the separability of learned embeddings. Extensive experiments on both homophilic and heterophilic graph benchmarks demonstrate that our method achieves competitive performance while preserving the scalability of linear attention.

---

### FactAppeal: Identifying Epistemic Factual Appeals in News Media

Authors: Guy Mor-Lan, Tamir Sheafer, Shaul R. Shenhav
Published: 2025-10-12
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.10627v2

How is a factual claim made credible? We propose the novel task of Epistemic Appeal Identification, which identifies whether and how factual statements have been anchored by external sources or evidence. To advance research on this task, we present FactAppeal, a manually annotated dataset of 3,226 English-language news sentences. Unlike prior resources that focus solely on claim detection and verification, FactAppeal identifies the nuanced epistemic structures and evidentiary basis underlying these claims and used to support them. FactAppeal contains span-level annotations which identify factual statements and mentions of sources on which they rely. Moreover, the annotations include fine-grained characteristics of factual appeals such as the type of source (e.g. Active Participant, Witness, Expert, Direct Evidence), whether it is mentioned by name, mentions of the source's role and epistemic credentials, attribution to the source via direct or indirect quotation, and other features. We model the task with a range of encoder models and generative decoder models in the 2B-9B parameter range. Our best performing model, based on Gemma 2 9B, achieves a macro-F1 score of 0.73.

---

### ImpMIA: Leveraging Implicit Bias for Membership Inference Attack

Authors: Yuval Golbari, Navve Wasserman, Gal Vardi, Michal Irani
Published: 2025-10-12
Categories: cs.LG, cs.CR, cs.CV
arXiv: https://arxiv.org/abs/2510.10625v3

Determining which data samples were used to train a model, known as Membership Inference Attack (MIA), is a well-studied and important problem with implications on data privacy. SotA methods (which are black-box attacks) rely on training many auxiliary reference models to imitate the behavior of the attacked model. As such, they rely on assumptions which rarely hold in real-world settings: (i) the attacker knows the training hyperparameters; (ii) all available non-training samples come from the same distribution as the training data; and (iii) the fraction of training data in the evaluation set is known. We show that removing these assumptions significantly harms the performance of black-box attacks. We introduce ImpMIA, a Membership Inference Attack that exploits the Implicit Bias of neural networks. Building on the maximum-margin implicit bias theory, ImpMIA uses the Karush-Kuhn-Tucker (KKT) optimality conditions to identify training samples -- those whose gradients most strongly reconstruct the trained model's parameters. Our approach is optimization-based, and requires NO training of reference-models, thus removing the need for any knowledge/assumptions regarding the attacked model's training procedure. While ImpMIA is a white-box attack (a setting which assumes access to model weights), this is becoming increasingly realistic given that many models are publicly available (e.g., via Hugging Face). ImpMIA achieves SotA performance compared to both black and white box attacks in settings where only the model weights are known, and a superset of the training data is available.

---

### SDG-L: A Semiparametric Deep Gaussian Process based Framework for Battery Capacity Prediction

Authors: Hanbing Liu, Yanru Wu, Yang Li, Ercan E. Kuruoglu, Xuan Zhang
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10621v1

Lithium-ion batteries are becoming increasingly omnipresent in energy supply. However, the durability of energy storage using lithium-ion batteries is threatened by their dropping capacity with the growing number of charging/discharging cycles. An accurate capacity prediction is the key to ensure system efficiency and reliability, where the exploitation of battery state information in each cycle has been largely undervalued. In this paper, we propose a semiparametric deep Gaussian process regression framework named SDG-L to give predictions based on the modeling of time series battery state data. By introducing an LSTM feature extractor, the SDG-L is specially designed to better utilize the auxiliary profiling information during charging/discharging process. In experimental studies based on NASA dataset, our proposed method obtains an average test MSE error of 1.2%. We also show that SDG-L achieves better performance compared to existing works and validate the framework using ablation studies.

---

### DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism

Authors: Chenyu Jiang, Zhenkun Cai, Ye Tian, Zhen Jia, Yida Wang, Chuan Wu
Published: 2025-10-12
Categories: cs.DC, cs.LG
arXiv: https://arxiv.org/abs/2510.10620v1

Context parallelism has emerged as a key technique to support long-context training, a growing trend in generative AI for modern large models. However, existing context parallel methods rely on static parallelization configurations that overlook the dynamic nature of training data, specifically, the variability in sequence lengths and token relationships (i.e., attention patterns) across samples. As a result, these methods often suffer from unnecessary communication overhead and imbalanced computation. In this paper, we present DCP, a dynamic context parallel training framework that introduces fine-grained blockwise partitioning of both data and computation. By enabling flexible mapping of data and computation blocks to devices, DCP can adapt to varying sequence characteristics, effectively reducing communication and improving memory and computation balance. Micro-benchmarks demonstrate that DCP accelerates attention by 1.19x~2.45x under causal masks and 2.15x~3.77x under sparse attention patterns. Additionally, we observe up to 0.94x~1.16x end-to-end training speed-up for causal masks, and 1.00x~1.46x for sparse masks.

---

### A Machine Learning Approach for MIDI to Guitar Tablature Conversion

Authors: Maximos Kaliakatsos-Papakostas, Gregoris Bastas, Dimos Makris, Dorien Herremans, Vassilis Katsouros, Petros Maragos
Published: 2025-10-12
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2510.10619v1

Guitar tablature transcription consists in deducing the string and the fret number on which each note should be played to reproduce the actual musical part. This assignment should lead to playable string-fret combinations throughout the entire track and, in general, preserve parsimonious motion between successive combinations. Throughout the history of guitar playing, specific chord fingerings have been developed across different musical styles that facilitate common idiomatic voicing combinations and motion between them. This paper presents a method for assigning guitar tablature notation to a given MIDI-based musical part (possibly consisting of multiple polyphonic tracks), i.e. no information about guitar-idiomatic expressional characteristics is involved (e.g. bending etc.) The current strategy is based on machine learning and requires a basic assumption about how much fingers can stretch on a fretboard; only standard 6-string guitar tuning is examined. The proposed method also examines the transcription of music pieces that was not meant to be played or could not possibly be played by a guitar (e.g. potentially a symphonic orchestra part), employing a rudimentary method for augmenting musical information and training/testing the system with artificial data. The results present interesting aspects about what the system can achieve when trained on the initial and augmented dataset, showing that the training with augmented data improves the performance even in simple, e.g. monophonic, cases. Results also indicate weaknesses and lead to useful conclusions about possible improvements.

---

### Preserving LLM Capabilities through Calibration Data Curation: From Analysis to Optimization

Authors: Bowei He, Lihao Yin, Huiling Zhen, Shuqi Liu, Han Wu, Xiaokun Zhang, Mingxuan Yuan, Chen Ma
Published: 2025-10-12
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.10618v1

Post-training compression has been a widely employed approach to scale down large language model (LLM) and facilitate efficient inference. In various proposed compression methods, including pruning and quantization, calibration data plays a vital role by informing the weight importance and activation dynamic ranges. However, how calibration data impacts the LLM capability after compression is less explored. Few of the existing works, though recognizing the significance of this study, only investigate the language modeling or commonsense reasoning performance degradation from limited angles, like the data sources or sample amounts. More systematic research is still needed to examine the impacts on different LLM capabilities in terms of compositional properties and domain correspondence of calibration data. In this work, we aim at bridging this gap and further analyze underlying influencing mechanisms from the activation pattern perspective. Especially, we explore the calibration data's impacts on high-level complex reasoning capabilities, like math problem solving and code generation. Delving into the underlying mechanism, we find that the representativeness and diversity in activation space more fundamentally determine the quality of calibration data. Finally, we propose a calibration data curation framework based on such observations and analysis, enhancing the performance of existing post-training compression methods on preserving critical LLM capabilities. Our code is provided in \href{https://github.com/BokwaiHo/COLA.git}{Link}.

---

### Encoder Decoder Generative Adversarial Network Model for Stock Market Prediction

Authors: Bahadur Yadav, Sanjay Kumar Mohanty
Published: 2025-10-12
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2510.10617v1

Forecasting stock prices remains challenging due to the volatile and non-linear nature of financial markets. Despite the promise of deep learning, issues such as mode collapse, unstable training, and difficulty in capturing temporal and feature level correlations have limited the applications of GANs in this domain. We propose a GRU-based Encoder-Decoder GAN (EDGAN) model that strikes a balance between expressive power and simplicity. The model introduces key innovations such as a temporal decoder with residual connections for precise reconstruction, conditioning on static and dynamic covariates for contextual learning, and a windowing mechanism to capture temporal dynamics. Here, the generator uses a dense encoder-decoder framework with residual GRU blocks. Extensive experiments on diverse stock datasets demonstrate that EDGAN achieves superior forecasting accuracy and training stability, even in volatile markets. It consistently outperforms traditional GAN variants in forecasting accuracy and convergence stability under market conditions.

---

### Dynamic Topic Evolution with Temporal Decay and Attention in Large Language Models

Authors: Di Wu, Shuaidong Pan
Published: 2025-10-12
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.10613v2

This paper proposes a modeling framework for dynamic topic evolution based on temporal large language models. The method first uses a large language model to obtain contextual embeddings of text and then introduces a temporal decay function and an attention mechanism. These components allow the model to adjust the importance of semantic units according to time intervals and capture topic variations across different periods. The temporal representations are then mapped into a latent topic space, where a state transition matrix is applied to describe the dynamic evolution of topics. A joint optimization objective constrains both semantic modeling and temporal consistency, ensuring diversity and smoothness in topic generation. The design emphasizes the unified modeling of semantic representation and temporal evolution, which improves topic coherence and diversity while enhancing stability and interpretability over time. Experiments on real-world corpora show that the framework effectively captures the generation, expansion, and decline of topics and outperforms existing models across multiple metrics. Overall, the proposed method provides a systematic solution for understanding dynamic semantic patterns in large-scale text, enriches the research paradigm of topic modeling, and supports complex text analysis tasks in multiple domains.

---

### UltraScatter: Ray-Based Simulation of Ultrasound Scattering

Authors: Felix Duelmer, Mohammad Farid Azampour, Nassir Navab
Published: 2025-10-12
Categories: physics.med-ph, cs.CV
arXiv: https://arxiv.org/abs/2510.10612v1

Traditional ultrasound simulation methods solve wave equations numerically, achieving high accuracy but at substantial computational cost. Faster alternatives based on convolution with precomputed impulse responses remain relatively slow, often requiring several minutes to generate a full B-mode image. We introduce UltraScatter, a probabilistic ray tracing framework that models ultrasound scattering efficiently and realistically. Tissue is represented as a volumetric field of scattering probability and scattering amplitude, and ray interactions are simulated via free-flight delta tracking. Scattered rays are traced to the transducer, with phase information incorporated through a linear time-of-flight model. Integrated with plane-wave imaging and beamforming, our parallelized ray tracing architecture produces B-mode images within seconds. Validation with phantom data shows realistic speckle and inclusion patterns, positioning UltraScatter as a scalable alternative to wave-based methods.

---

### OmniQuality-R: Advancing Reward Models Through All-Encompassing Quality Assessment

Authors: Yiting Lu, Fengbin Guan, Yixin Gao, Yan Zhong, Xinge Peng, Jiakang Yuan, Yihao Liu, Bo Zhang, Xin Li, Zhibo Chen, Weisi Lin
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10609v1

Current visual evaluation approaches are typically constrained to a single task. To address this, we propose OmniQuality-R, a unified reward modeling framework that transforms multi-task quality reasoning into continuous and interpretable reward signals for policy optimization. Inspired by subjective experiments, where participants are given task-specific instructions outlining distinct assessment principles prior to evaluation, we propose OmniQuality-R, a structured reward modeling framework that transforms multi-dimensional reasoning into continuous and interpretable reward signals. To enable this, we construct a reasoning-enhanced reward modeling dataset by sampling informative plan-reason trajectories via rejection sampling, forming a reliable chain-of-thought (CoT) dataset for supervised fine-tuning (SFT). Building on this, we apply Group Relative Policy Optimization (GRPO) for post-training, using a Gaussian-based reward to support continuous score prediction. To further stabilize the training and improve downstream generalization, we incorporate standard deviation (STD) filtering and entropy gating mechanisms during reinforcement learning. These techniques suppress unstable updates and reduce variance in policy optimization. We evaluate OmniQuality-R on three key IQA tasks: aesthetic quality assessment, technical quality evaluation, and text-image alignment.

---

### ViSurf: Visual Supervised-and-Reinforcement Fine-Tuning for Large Vision-and-Language Models

Authors: Yuqi Liu, Liangyu Chen, Jiazhen Liu, Mingkang Zhu, Zhisheng Zhong, Bei Yu, Jiaya Jia
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10606v4

Post-training Large Vision-and-Language Models (LVLMs) typically involves Supervised Fine-Tuning (SFT) for knowledge injection or Reinforcement Learning with Verifiable Rewards (RLVR) for performance enhancement. However, SFT often leads to sub-optimal performance, while RLVR remains constrained by the model's internal knowledge base. While a sequential SFT $\rightarrow$ RLVR pipeline can be used, it introduces significant computational overhead and suffers from catastrophic forgetting. To address these limitations, we propose ViSurf (\textbf{Vi}sual \textbf{Su}pervised-and-\textbf{R}einforcement \textbf{F}ine-Tuning), a unified, single-stage paradigm that integrates the strengths of both SFT and RLVR. By analyzing their training objectives, we establish a unified framework that injects ground-truth labels directly into RLVR rollouts, facilitating simultaneous external supervision and internal reinforcement. Furthermore, we introduce three novel reward control strategies to ensure training stability and optimization. Extensive experiments demonstrate that ViSurf consistently outperforms standalone SFT, RLVR, and the traditional two-stage pipeline across diverse benchmarks. In-depth analysis corroborates these findings, validating the derivation and design principles of ViSurf.

---

### Budget Allocation for Unknown Value Functions in a Lipschitz Space

Authors: MohammadHossein Bateni, Hossein Esfandiari, Samira HosseinGhorban, Alireza Mirrokni, Radin Shahdaei
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10605v2

Building learning models frequently requires evaluating numerous intermediate models. Examples include models considered during feature selection, model structure search, and parameter tunings. The evaluation of an intermediate model influences subsequent model exploration decisions. Although prior knowledge can provide initial quality estimates, true performance is only revealed after evaluation. In this work, we address the challenge of optimally allocating a bounded budget to explore the space of intermediate models. We formalize this as a general budget allocation problem over unknown-value functions within a Lipschitz space.

---

### FusionGen: Feature Fusion-Based Few-Shot EEG Data Generation

Authors: Yuheng Chen, Dingkun Liu, Xinyao Yang, Xinping Xu, Baicheng Chen, Dongrui Wu
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10604v1

Brain-computer interfaces (BCIs) provide potential for applications ranging from medical rehabilitation to cognitive state assessment by establishing direct communication pathways between the brain and external devices via electroencephalography (EEG). However, EEG-based BCIs are severely constrained by data scarcity and significant inter-subject variability, which hinder the generalization and applicability of EEG decoding models in practical settings. To address these challenges, we propose FusionGen, a novel EEG data generation framework based on disentangled representation learning and feature fusion. By integrating features across trials through a feature matching fusion module and combining them with a lightweight feature extraction and reconstruction pipeline, FusionGen ensures both data diversity and trainability under limited data constraints. Extensive experiments on multiple publicly available EEG datasets demonstrate that FusionGen significantly outperforms existing augmentation techniques, yielding notable improvements in classification accuracy.

---

### EA4LLM: A Gradient-Free Approach to Large Language Model Optimization via Evolutionary Algorithms

Authors: WenTao Liu, Siyu Song, Hao Hao, Aimin Zhou
Published: 2025-10-12
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.10603v2

In recent years, large language models (LLMs) have made remarkable progress, with model optimization primarily relying on gradient-based optimizers such as Adam. However, these gradient-based methods impose stringent hardware requirements, demanding high-concurrency, high-memory GPUs. Moreover, they require all neural network operations to be differentiable, thereby excluding many promising non-differentiable architectures from practical use. To address these limitations, we propose EA4LLM, an evolutionary algorithm for optimizing LLMs, and, for the first time, empirically verify full-parameter optimization from the pretraining stage across model sizes ranging from 0.5B to 32B. We conduct extensive experiments and provide key insights into how evolutionary algorithms can effectively optimize neural networks. Our work challenges the prevailing assumption that gradient-based optimization is the only viable approach for training neural networks. It also holds significant potential to reduce the computational cost of training large language models, thereby enabling groups with limited computational resources to participate in deep learning research.

---

### SpikeGrasp: A Benchmark for 6-DoF Grasp Pose Detection from Stereo Spike Streams

Authors: Zhuoheng Gao, Jiyao Zhang, Zhiyong Xie, Hao Dong, Zhaofei Yu, Rongmei Chen, Guozhang Chen, Tiejun Huang
Published: 2025-10-12
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2510.10602v2

Most robotic grasping systems rely on converting sensor data into explicit 3D point clouds, which is a computational step not found in biological intelligence. This paper explores a fundamentally different, neuro-inspired paradigm for 6-DoF grasp detection. We introduce SpikeGrasp, a framework that mimics the biological visuomotor pathway, processing raw, asynchronous events from stereo spike cameras, similarly to retinas, to directly infer grasp poses. Our model fuses these stereo spike streams and uses a recurrent spiking neural network, analogous to high-level visual processing, to iteratively refine grasp hypotheses without ever reconstructing a point cloud. To validate this approach, we built a large-scale synthetic benchmark dataset. Experiments show that SpikeGrasp surpasses traditional point-cloud-based baselines, especially in cluttered and textureless scenes, and demonstrates remarkable data efficiency. By establishing the viability of this end-to-end, neuro-inspired approach, SpikeGrasp paves the way for future systems capable of the fluid and efficient manipulation seen in nature, particularly for dynamic objects.

---

### A Distance Measure for Random Permutation Set: From the Layer-2 Belief Structure Perspective

Authors: Ruolan Cheng, Yong Deng, Serafín Moral, José Ramón Trillo
Published: 2025-10-12
Categories: cs.AI, cs.IT
arXiv: https://arxiv.org/abs/2510.10596v1

Random permutation set (RPS) is a recently proposed framework designed to represent order-structured uncertain information. Measuring the distance between permutation mass functions is a key research topic in RPS theory (RPST). This paper conducts an in-depth analysis of distances between RPSs from two different perspectives: random finite set (RFS) and transferable belief model (TBM). Adopting the layer-2 belief structure interpretation of RPS, we regard RPST as a refinement of TBM, where the order in the ordered focus set represents qualitative propensity. Starting from the permutation, we introduce a new definition of the cumulative Jaccard index to quantify the similarity between two permutations and further propose a distance measure method for RPSs based on the cumulative Jaccard index matrix. The metric and structural properties of the proposed distance measure are investigated, including the positive definiteness analysis of the cumulative Jaccard index matrix, and a correction scheme is provided. The proposed method has a natural top-weightiness property: inconsistencies between higher-ranked elements tend to result in greater distance values. Two parameters are provided to the decision-maker to adjust the weight and truncation depth. Several numerical examples are used to compare the proposed method with the existing method. The experimental results show that the proposed method not only overcomes the shortcomings of the existing method and is compatible with the Jousselme distance, but also has higher sensitivity and flexibility.

---

### A Layered Intuition -- Method Model with Scope Extension for LLM Reasoning

Authors: Hong Su
Published: 2025-10-12
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.10592v1

Existing studies have introduced method-based reasoning and scope extension as approaches to enhance Large Language Model (LLM) performance beyond direct matrix mappings. Building on these foundations, this paper summarizes and integrates these ideas into a unified Intuition-Method Layered Model with Scope Extension, designed to address indirected (unseen) issues more systematically. In this framework, intuition-based thinking provides rapid first-reaction answers, while method-based thinking decouples questions and solutions into transferable reasoning units. Scope extension is then applied to broaden applicability, including vertical (cause analysis), horizontal (parallel and generalized issues), and for the first time, temporal and spatial extensions, which expand reasoning across time and contextual dimensions. These extensions are organized into systematic knowledge trees that interconnect into a knowledge network, thereby increasing adaptability. To quantitatively evaluate this process, we propose the entropy of method extension, which measures the independence and diversity of extensions as an indicator of the system's capacity to solve unseen questions. By logically connecting existing approaches with new extensions and introducing an entropy-based evaluation framework, this work advances toward a more robust and extensible reasoning paradigm for LLMs in real-world problem-solving.

---

### A Simple and Better Baseline for Visual Grounding

Authors: Jingchao Wang, Wenlong Zhang, Dingjiang Huang, Hong Wang, Yefeng Zheng
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10587v1

Visual grounding aims to predict the locations of target objects specified by textual descriptions. For this task with linguistic and visual modalities, there is a latest research line that focuses on only selecting the linguistic-relevant visual regions for object localization to reduce the computational overhead. Albeit achieving impressive performance, it is iteratively performed on different image scales, and at every iteration, linguistic features and visual features need to be stored in a cache, incurring extra overhead. To facilitate the implementation, in this paper, we propose a feature selection-based simple yet effective baseline for visual grounding, called FSVG. Specifically, we directly encapsulate the linguistic and visual modalities into an overall network architecture without complicated iterative procedures, and utilize the language in parallel as guidance to facilitate the interaction between linguistic modal and visual modal for extracting effective visual features. Furthermore, to reduce the computational cost, during the visual feature learning, we introduce a similarity-based feature selection mechanism to only exploit language-related visual features for faster prediction. Extensive experiments conducted on several benchmark datasets comprehensively substantiate that the proposed FSVG achieves a better balance between accuracy and efficiency beyond the current state-of-the-art methods. Code is available at https://github.com/jcwang0602/FSVG.

---

### Compositional Symmetry as Compression: Lie Pseudogroup Structure in Algorithmic Agents

Authors: Giulio Ruffini
Published: 2025-10-12
Categories: cs.LG, cs.AI, cs.IT, q-bio.NC
arXiv: https://arxiv.org/abs/2510.10586v1

In the algorithmic (Kolmogorov) view, agents are programs that track and compress sensory streams using generative programs. We propose a framework where the relevant structural prior is simplicity (Solomonoff) understood as \emph{compositional symmetry}: natural streams are well described by (local) actions of finite-parameter Lie pseudogroups on geometrically and topologically complex low-dimensional configuration manifolds (latent spaces). Modeling the agent as a generic neural dynamical system coupled to such streams, we show that accurate world-tracking imposes (i) \emph{structural constraints} -- equivariance of the agent's constitutive equations and readouts -- and (ii) \emph{dynamical constraints}: under static inputs, symmetry induces conserved quantities (Noether-style labels) in the agent dynamics and confines trajectories to reduced invariant manifolds; under slow drift, these manifolds move but remain low-dimensional. This yields a hierarchy of reduced manifolds aligned with the compositional factorization of the pseudogroup, providing a geometric account of the ``blessing of compositionality'' in deep models. We connect these ideas to the Spencer formalism for Lie pseudogroups and formulate a symmetry-based, self-contained version of predictive coding in which higher layers receive only \emph{coarse-grained residual transformations} (prediction-error coordinates) along symmetry directions unresolved at lower layers.

---

### Equipping Vision Foundation Model with Mixture of Experts for Out-of-Distribution Detection

Authors: Shizhen Zhao, Jiahui Liu, Xin Wen, Haoru Tan, Xiaojuan Qi
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10584v1

Pre-trained vision foundation models have transformed many computer vision tasks. Despite their strong ability to learn discriminative and generalizable features crucial for out-of-distribution (OOD) detection, their impact on this task remains underexplored. Motivated by this gap, we systematically investigate representative vision foundation models for OOD detection. Our findings reveal that a pre-trained DINOv2 model, even without fine-tuning on in-domain (ID) data, naturally provides a highly discriminative feature space for OOD detection, achieving performance comparable to existing state-of-the-art methods without requiring complex designs. Beyond this, we explore how fine-tuning foundation models on in-domain (ID) data can enhance OOD detection. However, we observe that the performance of vision foundation models remains unsatisfactory in scenarios with a large semantic space. This is due to the increased complexity of decision boundaries as the number of categories grows, which complicates the optimization process. To mitigate this, we propose the Mixture of Feature Experts (MoFE) module, which partitions features into subspaces, effectively capturing complex data distributions and refining decision boundaries. Further, we introduce a Dynamic-$β$ Mixup strategy, which samples interpolation weights from a dynamic beta distribution. This adapts to varying levels of learning difficulty across categories, improving feature learning for more challenging categories. Extensive experiments demonstrate the effectiveness of our approach, significantly outperforming baseline methods.

---

### Injecting Frame-Event Complementary Fusion into Diffusion for Optical Flow in Challenging Scenes

Authors: Haonan Wang, Hanyu Zhou, Haoyue Liu, Luxin Yan
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10577v1

Optical flow estimation has achieved promising results in conventional scenes but faces challenges in high-speed and low-light scenes, which suffer from motion blur and insufficient illumination. These conditions lead to weakened texture and amplified noise and deteriorate the appearance saturation and boundary completeness of frame cameras, which are necessary for motion feature matching. In degraded scenes, the frame camera provides dense appearance saturation but sparse boundary completeness due to its long imaging time and low dynamic range. In contrast, the event camera offers sparse appearance saturation, while its short imaging time and high dynamic range gives rise to dense boundary completeness. Traditionally, existing methods utilize feature fusion or domain adaptation to introduce event to improve boundary completeness. However, the appearance features are still deteriorated, which severely affects the mostly adopted discriminative models that learn the mapping from visual features to motion fields and generative models that generate motion fields based on given visual features. So we introduce diffusion models that learn the mapping from noising flow to clear flow, which is not affected by the deteriorated visual features. Therefore, we propose a novel optical flow estimation framework Diff-ABFlow based on diffusion models with frame-event appearance-boundary fusion.

---

### UniFlow: A Unified Pixel Flow Tokenizer for Visual Understanding and Generation

Authors: Zhengrong Yue, Haiyu Zhang, Xiangyu Zeng, Boyu Chen, Chenting Wang, Shaobin Zhuang, Lu Dong, Yi Wang, Limin Wang, Yali Wang
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10575v2

Tokenizer is a crucial component for both visual understanding and generation. To advance toward the ultimate goal of universal modeling, recent research has focused on developing a unified tokenizer. However, existing tokenizers face a significant performance trade-off between understanding and generation, stemming from the inherent conflict between high-level semantic abstraction and low-level pixel reconstruction. To tackle this challenge, we propose a generic and unified tokenizer, namely UniFlow, by flexibly adapting any visual encoder with a concise reconstruction decoder. Specifically, we introduce layer-wise adaptive self-distillation applied to the well-pretrained visual encoders, which enables UniFlow to simultaneously inherit the strong semantic features for visual understanding and flexibly adapt to model fine-grained details for visual generation. Moreover, we propose a lightweight patch-wise pixel flow decoder, which efficiently achieves high-fidelity pixel reconstruction by modeling a conditional flow from the noisy state back to the patch-wise pixel domain. By leveraging the semantic features as visual conditions for the decoder, we effectively alleviate the training conflicts between understanding and generation. Furthermore, the patch-wise learning strategy simplifies the data distribution, thereby improving training efficiency. Extensive experiments across 13 challenging benchmarks spanning 7 widely studied visual understanding and generation tasks demonstrate that UniFlow achieves a win-win outcome. For instance, our 7B UniFlow-XL not only surpasses the 14B TokenFlow-XL by 6.05% on average understanding benchmarks, but also achieves a competitive results in both visual reconstruction and generation, surpassing UniTok by 0.15 in rFID and 0.09 in gFID (without guidance), respectively.

---

### Deep semi-supervised approach based on consistency regularization and similarity learning for weeds classification

Authors: Farouq Benchallal, Adel Hafiane, Nicolas Ragot, Raphael Canals
Published: 2025-10-12
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.10573v1

Weed species classification represents an important step for the development of automated targeting systems that allow the adoption of precision agriculture practices. To reduce costs and yield losses caused by their presence. The identification of weeds is a challenging problem due to their shared similarities with crop plants and the variability related to the differences in terms of their types. Along with the variations in relation to changes in field conditions. Moreover, to fully benefit from deep learning-based methods, large fully annotated datasets are needed. This requires time intensive and laborious process for data labeling, which represents a limitation in agricultural applications. Hence, for the aim of improving the utilization of the unlabeled data, regarding conditions of scarcity in terms of the labeled data available during the learning phase and provide robust and high classification performance. We propose a deep semi-supervised approach, that combines consistency regularization with similarity learning. Through our developed deep auto-encoder architecture, experiments realized on the DeepWeeds dataset and inference in noisy conditions demonstrated the effectiveness and robustness of our method in comparison to state-of-the-art fully supervised deep learning models. Furthermore, we carried out ablation studies for an extended analysis of our proposed joint learning strategy.

---

### Understanding Self-supervised Contrastive Learning through Supervised Objectives

Authors: Byeongchan Lee
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10572v1

Self-supervised representation learning has achieved impressive empirical success, yet its theoretical understanding remains limited. In this work, we provide a theoretical perspective by formulating self-supervised representation learning as an approximation to supervised representation learning objectives. Based on this formulation, we derive a loss function closely related to popular contrastive losses such as InfoNCE, offering insight into their underlying principles. Our derivation naturally introduces the concepts of prototype representation bias and a balanced contrastive loss, which help explain and improve the behavior of self-supervised learning algorithms. We further show how components of our theoretical framework correspond to established practices in contrastive learning. Finally, we empirically validate the effect of balancing positive and negative pair interactions. All theoretical proofs are provided in the appendix, and our code is included in the supplementary material.

---

### Multitask Learning with Learned Task Relationships

Authors: Zirui Wan, Stefan Vlaski
Published: 2025-10-12
Categories: cs.LG, cs.DC, cs.MA
arXiv: https://arxiv.org/abs/2510.10570v1

Classical consensus-based strategies for federated and decentralized learning are statistically suboptimal in the presence of heterogeneous local data or task distributions. As a result, in recent years, there has been growing interest in multitask or personalized strategies, which allow individual agents to benefit from one another in pursuing locally optimal models without enforcing consensus. Existing strategies require either precise prior knowledge of the underlying task relationships or are fully non-parametric and instead rely on meta-learning or proximal constructions. In this work, we introduce an algorithmic framework that strikes a balance between these extremes. By modeling task relationships through a Gaussian Markov Random Field with an unknown precision matrix, we develop a strategy that jointly learns both the task relationships and the local models, allowing agents to self-organize in a way consistent with their individual data distributions. Our theoretical analysis quantifies the quality of the learned relationship, and our numerical experiments demonstrate its practical effectiveness.

---

### BitMar: Low-Bit Multimodal Fusion with Episodic Memory for Edge Devices

Authors: Euhid Aman, Esteban Carlin, Hsing-Kuo Pao, Giovanni Beltrame, Ghaluh Indah Permata Sari, Yie-Tarng Chen
Published: 2025-10-12
Categories: cs.CL, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2510.10560v1

Cross-attention transformers and other multimodal vision-language models excel at grounding and generation; however, their extensive, full-precision backbones make it challenging to deploy them on edge devices. Memory-augmented architectures enhance the utilization of past context; however, most works rarely pair them with aggressive edge-oriented quantization. We introduce BitMar, a quantized multimodal transformer that proposes an external human-like episodic memory for effective image-text generation on hardware with limited resources. BitMar utilizes 1.58-bit encoders, one for text (BitNet-style) and one for vision (DiNOv2-based), to create compact embeddings that are combined and used to query a fixed-size key-value episodic memory. During vector retrieval, the BitNet decoder applies per-layer conditioning, which increases the contextual relevance of generated content. The decoder also employs attention sinks with a sliding-window mechanism to process long or streaming inputs under tight memory budgets. The combination of per-layer conditioning and sliding-window attention achieves a strong quality-speed trade-off, delivering competitive captioning and multimodal understanding at low latency with a small model footprint. These characteristics make BitMar well-suited for edge deployment.

---

### Multi-scale Frequency-Aware Adversarial Network for Parkinson's Disease Assessment Using Wearable Sensors

Authors: Weiming Zhao, Xulong Wang, Jun Qi, Yun Yang, Po Yang
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10558v1

Severity assessment of Parkinson's disease (PD) using wearable sensors offers an effective, objective basis for clinical management. However, general-purpose time series models often lack pathological specificity in feature extraction, making it difficult to capture subtle signals highly correlated with PD.Furthermore, the temporal sparsity of PD symptoms causes key diagnostic features to be easily "diluted" by traditional aggregation methods, further complicating assessment. To address these issues, we propose the Multi-scale Frequency-Aware Adversarial Multi-Instance Network (MFAM). This model enhances feature specificity through a frequency decomposition module guided by medical prior knowledge. Furthermore, by introducing an attention-based multi-instance learning (MIL) framework, the model can adaptively focus on the most diagnostically valuable sparse segments.We comprehensively validated MFAM on both the public PADS dataset for PD versus differential diagnosis (DD) binary classification and a private dataset for four-class severity assessment. Experimental results demonstrate that MFAM outperforms general-purpose time series models in handling complex clinical time series with specificity, providing a promising solution for automated assessment of PD severity.

---

### MRS-YOLO Railroad Transmission Line Foreign Object Detection Based on Improved YOLO11 and Channel Pruning

Authors: Siyuan Liu, Junting Lin
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10553v1

Aiming at the problems of missed detection, false detection and low detection efficiency in transmission line foreign object detection under railway environment, we proposed an improved algorithm MRS-YOLO based on YOLO11. Firstly, a multi-scale Adaptive Kernel Depth Feature Fusion (MAKDF) module is proposed and fused with the C3k2 module to form C3k2_MAKDF, which enhances the model's feature extraction capability for foreign objects of different sizes and shapes. Secondly, a novel Re-calibration Feature Fusion Pyramid Network (RCFPN) is designed as a neck structure to enhance the model's ability to integrate and utilize multi-level features effectively. Then, Spatial and Channel Reconstruction Detect Head (SC_Detect) based on spatial and channel preprocessing is designed to enhance the model's overall detection performance. Finally, the channel pruning technique is used to reduce the redundancy of the improved model, drastically reduce Parameters and Giga Floating Point Operations Per Second (GFLOPs), and improve the detection efficiency. The experimental results show that the mAP50 and mAP50:95 of the MRS-YOLO algorithm proposed in this paper are improved to 94.8% and 86.4%, respectively, which are 0.7 and 2.3 percentage points higher compared to the baseline, while Parameters and GFLOPs are reduced by 44.2% and 17.5%, respectively. It is demonstrated that the improved algorithm can be better applied to the task of foreign object detection in railroad transmission lines.

---

### ELAIPBench: A Benchmark for Expert-Level Artificial Intelligence Paper Understanding

Authors: Xinbang Dai, Huikang Hu, Yongrui Chen, Jiaqi Li, Rihui Jin, Yuyang Zhang, Xiaoguang Li, Lifeng Shang, Guilin Qi
Published: 2025-10-12
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.10549v2

While large language models (LLMs) excel at many domain-specific tasks, their ability to deeply comprehend and reason about full-length academic papers remains underexplored. Existing benchmarks often fall short of capturing such depth, either due to surface-level question design or unreliable evaluation metrics. To address this gap, we introduce ELAIPBench, a benchmark curated by domain experts to evaluate LLMs' comprehension of artificial intelligence (AI) research papers. Developed through an incentive-driven, adversarial annotation process, ELAIPBench features 403 multiple-choice questions from 137 papers. It spans three difficulty levels and emphasizes non-trivial reasoning rather than shallow retrieval. Our experiments show that the best-performing LLM achieves an accuracy of only 39.95%, far below human performance. Moreover, we observe that frontier LLMs equipped with a thinking mode or a retrieval-augmented generation (RAG) system fail to improve final results-even harming accuracy due to overthinking or noisy retrieval. These findings underscore the significant gap between current LLM capabilities and genuine comprehension of academic papers.

---

### GLOFNet -- A Multimodal Dataset for GLOF Monitoring and Prediction

Authors: Zuha Fatima, Muhammad Anser Sohaib, Muhammad Talha, Sidra Sultana, Ayesha Kanwal, Nazia Perwaiz
Published: 2025-10-12
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.10546v1

Glacial Lake Outburst Floods (GLOFs) are rare but destructive hazards in high mountain regions, yet predictive research is hindered by fragmented and unimodal data. Most prior efforts emphasize post-event mapping, whereas forecasting requires harmonized datasets that combine visual indicators with physical precursors. We present GLOFNet, a multimodal dataset for GLOF monitoring and prediction, focused on the Shisper Glacier in the Karakoram. It integrates three complementary sources: Sentinel-2 multispectral imagery for spatial monitoring, NASA ITS_LIVE velocity products for glacier kinematics, and MODIS Land Surface Temperature records spanning over two decades. Preprocessing included cloud masking, quality filtering, normalization, temporal interpolation, augmentation, and cyclical encoding, followed by harmonization across modalities. Exploratory analysis reveals seasonal glacier velocity cycles, long-term warming of ~0.8 K per decade, and spatial heterogeneity in cryospheric conditions. The resulting dataset, GLOFNet, is publicly available to support future research in glacial hazard prediction. By addressing challenges such as class imbalance, cloud contamination, and coarse resolution, GLOFNet provides a structured foundation for benchmarking multimodal deep learning approaches to rare hazard prediction.

---

### PAC-Bayesian Reinforcement Learning Trains Generalizable Policies

Authors: Abdelkrim Zitouni, Mehdi Hennequin, Juba Agoun, Ryan Horache, Nadia Kabachi, Omar Rivasplata
Published: 2025-10-12
Categories: cs.LG, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2510.10544v3

We derive a novel PAC-Bayesian generalization bound for reinforcement learning that explicitly accounts for Markov dependencies in the data, through the chain's mixing time. This contributes to overcoming challenges in obtaining generalization guarantees for reinforcement learning, where the sequential nature of data breaks the independence assumptions underlying classical bounds. The new bound provides non-vacuous certificates for modern off-policy algorithms such as Soft Actor-Critic. We demonstrate the practical utility of the bound through PB-SAC, a novel algorithm that optimizes the bound during training to guide exploration. Experiments across several continuous control tasks show that the proposed approach provides meaningful confidence certificates while maintaining competitive performance.

---

### Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods?

Authors: Zihan Chen, Yiming Zhang, Hengguang Zhou, Zenghui Ding, Yining Sun, Cho-Jui Hsieh
Published: 2025-10-12
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.10541v2

Current benchmarks are inadequate for evaluating progress in reinforcement learning (RL) for large language models (LLMs).Despite recent benchmark gains reported for RL, we find that training on these benchmarks' training sets achieves nearly the same performance as training directly on the test sets, suggesting that the benchmarks cannot reliably separate further progress.To study this phenomenon, we introduce a diagnostic suite and the Oracle Performance Gap (OPG) metric that quantifies the performance difference between training on the train split versus the test split of a benchmark. We further analyze this phenomenon with stress tests and find that, despite strong benchmark scores, existing RL methods struggle to generalize across distribution shifts, varying levels of difficulty, and counterfactual scenarios: shortcomings that current benchmarks fail to reveal.We conclude that current benchmarks are insufficient for evaluating generalization and propose three core principles for designing more faithful benchmarks: sufficient difficulty, balanced evaluation, and distributional robustness.

---

### Detecting Hallucinations in Authentic LLM-Human Interactions

Authors: Yujie Ren, Niklas Gruhlke, Anne Lauscher
Published: 2025-10-12
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.10539v2

As large language models (LLMs) are increasingly applied in sensitive domains such as medicine and law, hallucination detection has become a critical task. Although numerous benchmarks have been proposed to advance research in this area, most of them are artificially constructed--either through deliberate hallucination induction or simulated interactions--rather than derived from genuine LLM-human dialogues. Consequently, these benchmarks fail to fully capture the characteristics of hallucinations that occur in real-world usage. To address this limitation, we introduce AuthenHallu, the first hallucination detection benchmark built entirely from authentic LLM-human interactions. For AuthenHallu, we select and annotate samples from genuine LLM-human dialogues, thereby providing a faithful reflection of how LLMs hallucinate in everyday user interactions. Statistical analysis shows that hallucinations occur in 31.4% of the query-response pairs in our benchmark, and this proportion increases dramatically to 60.0% in challenging domains such as Math & Number Problems. Furthermore, we explore the potential of using vanilla LLMs themselves as hallucination detectors and find that, despite some promise, their current performance remains insufficient in real-world scenarios. The data and code are publicly available at https://github.com/TAI-HAMBURG/AuthenHallu.

---

### MCE: Towards a General Framework for Handling Missing Modalities under Imbalanced Missing Rates

Authors: Binyu Zhao, Wei Zhang, Zhaonian Zou
Published: 2025-10-12
Categories: cs.CV, cs.LG, cs.MM
arXiv: https://arxiv.org/abs/2510.10534v2

Multi-modal learning has made significant advances across diverse pattern recognition applications. However, handling missing modalities, especially under imbalanced missing rates, remains a major challenge. This imbalance triggers a vicious cycle: modalities with higher missing rates receive fewer updates, leading to inconsistent learning progress and representational degradation that further diminishes their contribution. Existing methods typically focus on global dataset-level balancing, often overlooking critical sample-level variations in modality utility and the underlying issue of degraded feature quality. We propose Modality Capability Enhancement (MCE) to tackle these limitations. MCE includes two synergistic components: i) Learning Capability Enhancement (LCE), which introduces multi-level factors to dynamically balance modality-specific learning progress, and ii) Representation Capability Enhancement (RCE), which improves feature semantics and robustness through subset prediction and cross-modal completion tasks. Comprehensive evaluations on four multi-modal benchmarks show that MCE consistently outperforms state-of-the-art methods under various missing configurations. The final published version is now available at https://doi.org/10.1016/j.patcog.2025.112591. Our code is available at https://github.com/byzhaoAI/MCE.

---

### Layout-Independent License Plate Recognition via Integrated Vision and Language Models

Authors: Elham Shabaninia, Fatemeh Asadi-zeydabadi, Hossein Nezamabadi-pour
Published: 2025-10-12
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.10533v1

This work presents a pattern-aware framework for automatic license plate recognition (ALPR), designed to operate reliably across diverse plate layouts and challenging real-world conditions. The proposed system consists of a modern, high-precision detection network followed by a recognition stage that integrates a transformer-based vision model with an iterative language modelling mechanism. This unified recognition stage performs character identification and post-OCR refinement in a seamless process, learning the structural patterns and formatting rules specific to license plates without relying on explicit heuristic corrections or manual layout classification. Through this design, the system jointly optimizes visual and linguistic cues, enables iterative refinement to improve OCR accuracy under noise, distortion, and unconventional fonts, and achieves layout-independent recognition across multiple international datasets (IR-LPR, UFPR-ALPR, AOLP). Experimental results demonstrate superior accuracy and robustness compared to recent segmentation-free approaches, highlighting how embedding pattern analysis within the recognition stage bridges computer vision and language modelling for enhanced adaptability in intelligent transportation and surveillance applications.

---

### Reinforced Domain Selection for Continuous Domain Adaptation

Authors: Hanbing Liu, Huaze Tang, Yanru Wu, Yang Li, Xiao-Ping Zhang
Published: 2025-10-12
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.10530v1

Continuous Domain Adaptation (CDA) effectively bridges significant domain shifts by progressively adapting from the source domain through intermediate domains to the target domain. However, selecting intermediate domains without explicit metadata remains a substantial challenge that has not been extensively explored in existing studies. To tackle this issue, we propose a novel framework that combines reinforcement learning with feature disentanglement to conduct domain path selection in an unsupervised CDA setting. Our approach introduces an innovative unsupervised reward mechanism that leverages the distances between latent domain embeddings to facilitate the identification of optimal transfer paths. Furthermore, by disentangling features, our method facilitates the calculation of unsupervised rewards using domain-specific features and promotes domain adaptation by aligning domain-invariant features. This integrated strategy is designed to simultaneously optimize transfer paths and target task performance, enhancing the effectiveness of domain adaptation processes. Extensive empirical evaluations on datasets such as Rotated MNIST and ADNI demonstrate substantial improvements in prediction accuracy and domain selection efficiency, establishing our method's superiority over traditional CDA approaches.

---

### Merlin's Whisper: Enabling Efficient Reasoning in Large Language Models via Black-box Persuasive Prompting

Authors: Heming Xia, Cunxiao Du, Rui Li, Chak Tou Leong, Yongqi Li, Wenjie Li
Published: 2025-10-12
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.10528v3

Large reasoning models (LRMs) have demonstrated remarkable proficiency in tackling complex tasks through step-by-step thinking. However, this lengthy reasoning process incurs substantial computational and latency overheads, hindering the practical deployment of LRMs. This work presents a new approach to mitigating overthinking in LRMs via black-box persuasive prompting. By treating LRMs as black-box communicators, we investigate how to persuade them to generate concise responses without compromising accuracy. We introduce Whisper, an iterative refinement framework that generates high-quality persuasive prompts from diverse perspectives. Experiments across multiple benchmarks demonstrate that Whisper consistently reduces token usage while preserving performance. Notably, Whisper achieves a 3x reduction in average response length on simple GSM8K questions for the Qwen3 model series and delivers an average ~40% token reduction across all benchmarks. For closed-source APIs, Whisper reduces token usage on MATH-500 by 46% for Claude-3.7 and 50% for Gemini-2.5. Further analysis reveals the broad applicability of Whisper across data domains, model scales, and families, underscoring the potential of black-box persuasive prompting as a practical strategy for enhancing LRM efficiency.

---