# Research Papers — 2025-11 · bundle page 41 of 197

> Full-text markdown of papers #2001–2050 of 9830 published in 2025-11, newest first. AI-readable mirrors of the full 9830-paper 2025-11 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-11/page/41
[← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/40.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-11/page/42.md)

### 4DWorldBench: A Comprehensive Evaluation Framework for 3D/4D World Generation Models

Authors: Yiting Lu, Wei Luo, Peiyan Tu, Haoran Li, Hanxin Zhu, Zihao Yu, Xingrui Wang, Xinyi Chen, Xinge Peng, Xin Li, Zhibo Chen
Published: 2025-11-25
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19836v1

World Generation Models are emerging as a cornerstone of next-generation multimodal intelligence systems. Unlike traditional 2D visual generation, World Models aim to construct realistic, dynamic, and physically consistent 3D/4D worlds from images, videos, or text. These models not only need to produce high-fidelity visual content but also maintain coherence across space, time, physics, and instruction control, enabling applications in virtual reality, autonomous driving, embodied intelligence, and content creation. However, prior benchmarks emphasize different evaluation dimensions and lack a unified assessment of world-realism capability. To systematically evaluate World Models, we introduce the 4DWorldBench, which measures models across four key dimensions: Perceptual Quality, Condition-4D Alignment, Physical Realism, and 4D Consistency. The benchmark covers tasks such as Image-to-3D/4D, Video-to-4D, Text-to-3D/4D. Beyond these, we innovatively introduce adaptive conditioning across multiple modalities, which not only integrates but also extends traditional evaluation paradigms. To accommodate different modality-conditioned inputs, we map all modality conditions into a unified textual space during evaluation, and further integrate LLM-as-judge, MLLM-as-judge, and traditional network-based methods. This unified and adaptive design enables more comprehensive and consistent evaluation of alignment, physical realism, and cross-modal coherence. Preliminary human studies further demonstrate that our adaptive tool selection achieves closer agreement with subjective human judgments. We hope this benchmark will serve as a foundation for objective comparisons and improvements, accelerating the transition from "visual generation" to "world generation." Our project can be found at https://yeppp27.github.io/4DWorldBench.github.io/.

---

### Rectified SpaAttn: Revisiting Attention Sparsity for Efficient Video Generation

Authors: Xuewen Liu, Zhikai Li, Jing Zhang, Mengjuan Chen, Qingyi Gu
Published: 2025-11-25
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.19835v1

Diffusion Transformers dominate video generation, but the quadratic complexity of attention computation introduces substantial latency. Attention sparsity reduces computational costs by focusing on critical tokens while ignoring non-critical tokens. However, existing methods suffer from severe performance degradation. In this paper, we revisit attention sparsity and reveal that existing methods induce systematic biases in attention allocation: (1) excessive focus on critical tokens amplifies their attention weights; (2) complete neglect of non-critical tokens causes the loss of relevant attention weights. To address these issues, we propose Rectified SpaAttn, which rectifies attention allocation with implicit full attention reference, thereby enhancing the alignment between sparse and full attention maps. Specifically: (1) for critical tokens, we show that their bias is proportional to the sparse attention weights, with the ratio governed by the amplified weights. Accordingly, we propose Isolated-Pooling Attention Reallocation, which calculates accurate rectification factors by reallocating multimodal pooled weights. (2) for non-critical tokens, recovering attention weights from the pooled query-key yields attention gains but also introduces pooling errors. Therefore, we propose Gain-Aware Pooling Rectification, which ensures that the rectified gain consistently surpasses the induced error. Moreover, we customize and integrate the Rectified SpaAttn kernel using Triton, achieving up to 3.33 and 2.08 times speedups on HunyuanVideo and Wan 2.1, respectively, while maintaining high generation quality. We release Rectified SpaAttn as open-source at https://github.com/BienLuky/Rectified-SpaAttn .

---

### Large Language Model Aided Birt-Hogg-Dube Syndrome Diagnosis with Multimodal Retrieval-Augmented Generation

Authors: Haoqing Li, Jun Shi, Xianmeng Chen, Qiwei Jia, Rui Wang, Wei Wei, Hong An, Xiaowen Hu
Published: 2025-11-25
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19834v1

Deep learning methods face dual challenges of limited clinical samples and low inter-class differentiation among Diffuse Cystic Lung Diseases (DCLDs) in advancing Birt-Hogg-Dube syndrome (BHD) diagnosis via Computed Tomography (CT) imaging. While Multimodal Large Language Models (MLLMs) demonstrate diagnostic potential fo such rare diseases, the absence of domain-specific knowledge and referable radiological features intensify hallucination risks. To address this problem, we propose BHD-RAG, a multimodal retrieval-augmented generation framework that integrates DCLD-specific expertise and clinical precedents with MLLMs to improve BHD diagnostic accuracy. BHDRAG employs: (1) a specialized agent generating imaging manifestation descriptions of CT images to construct a multimodal corpus of DCLDs cases. (2) a cosine similarity-based retriever pinpointing relevant imagedescription pairs for query images, and (3) an MLLM synthesizing retrieved evidence with imaging data for diagnosis. BHD-RAG is validated on the dataset involving four types of DCLDs, achieving superior accuracy and generating evidence-based descriptions closely aligned with expert insights.

---

### Beyond Relational: Semantic-Aware Multi-Modal Analytics with LLM-Native Query Optimization

Authors: Junhao Zhu, Lu Chen, Xiangyu Ke, Ziquan Fang, Tianyi Li, Yunjun Gao, Christian S. Jensen
Published: 2025-11-25
Categories: cs.DB, cs.AI
arXiv: https://arxiv.org/abs/2511.19830v1

Multi-modal analytical processing has the potential to transform applications in e-commerce, healthcare, entertainment, and beyond. However, real-world adoption remains elusive due to the limited ability of traditional relational query operators to capture query semantics. The emergence of foundation models, particularly the large language models (LLMs), opens up new opportunities to develop flexible, semantic-aware data analytics systems that transcend the relational paradigm. We present Nirvana, a multi-modal data analytics framework that incorporates programmable semantic operators while leveraging both logical and physical query optimization strategies, tailored for LLM-driven semantic query processing. Nirvana addresses two key challenges. First, it features an agentic logical optimizer that uses natural language-specified transformation rules and random-walk-based search to explore vast spaces of semantically equivalent query plans -- far beyond the capabilities of conventional optimizers. Second, it introduces a cost-aware physical optimizer that selects the most effective LLM backend for each operator using a novel improvement-score metric. To further enhance efficiency, Nirvana incorporates computation reuse and evaluation pushdown techniques guided by model capability hypotheses. Experimental evaluations on three real-world benchmarks demonstrate that Nirvana is able to reduce end-to-end runtime by 10%--85% and reduces system processing costs by 76% on average, outperforming state-of-the-art systems at both efficiency and scalability.

---

### Knowing How to Edit: Reliable Evaluation Signals for Diagnosing and Optimizing Prompts at Query Level

Authors: Ke Chen, Yifeng Wang, Hassan Almosapeeh, Haohan Wang
Published: 2025-11-25
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.19829v3

Prompt optimization has become a central mechanism for eliciting strong performance from LLMs, and recent work has made substantial progress by proposing diverse prompt evaluation metrics and optimization strategies. Despite these advances, prompt evaluation and prompt optimization are often developed in isolation, limiting the extent to which evaluation can effectively inform prompt refinement. In this work, we study prompt optimization as a process guided by performance-relevant evaluation signals. To address the disconnect between evaluation and optimization, we propose an evaluation-instructed prompt optimization approach that explicitly connects prompt evaluation with query-dependent optimization. Our method integrates multiple complementary prompt quality metrics into a performance-reflective evaluation framework and trains an execution-free evaluator that predicts prompt quality directly from text, avoiding repeated model executions. These evaluation signals then guide prompt refinement in a targeted and interpretable manner. Empirically, the proposed evaluator achieves 83.7% accuracy in predicting prompt performance. When incorporated into the optimization process, our approach consistently outperforms existing optimization baselines across eight benchmark datasets and three different backbone LLMs. Overall, our results demonstrate that reliable and efficient evaluation signals can serve as an effective foundation for robust and interpretable prompt optimization.

---

### ReDirector: Creating Any-Length Video Retakes with Rotary Camera Encoding

Authors: Byeongjun Park, Byung-Hoon Kim, Hyungjin Chung, Jong Chul Ye
Published: 2025-11-25
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19827v1

We present ReDirector, a novel camera-controlled video retake generation method for dynamically captured variable-length videos. In particular, we rectify a common misuse of RoPE in previous works by aligning the spatiotemporal positions of the input video and the target retake. Moreover, we introduce Rotary Camera Encoding (RoCE), a camera-conditioned RoPE phase shift that captures and integrates multi-view relationships within and across the input and target videos. By integrating camera conditions into RoPE, our method generalizes to out-of-distribution camera trajectories and video lengths, yielding improved dynamic object localization and static background preservation. Extensive experiments further demonstrate significant improvements in camera controllability, geometric consistency, and video quality across various trajectories and lengths.

---

### Mosaic Pruning: A Hierarchical Framework for Generalizable Pruning of Mixture-of-Experts Models

Authors: Wentao Hu, Mingkuan Zhao, Shuangyong Song, Xiaoyan Zhu, Xin Lai, Jiayin Wang
Published: 2025-11-25
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.19822v1

Sparse Mixture-of-Experts (SMoE) architectures have enabled a new frontier in scaling Large Language Models (LLMs), offering superior performance by activating only a fraction of their total parameters during inference. However, their practical deployment is severely hampered by substantial static memory overhead, as all experts must be loaded into memory. Existing post-training pruning methods, while reducing model size, often derive their pruning criteria from a single, general-purpose corpus. This leads to a critical limitation: a catastrophic performance degradation when the pruned model is applied to other domains, necessitating a costly re-pruning for each new domain. To address this generalization gap, we introduce Mosaic Pruning (MoP). The core idea of MoP is to construct a functionally comprehensive set of experts through a structured ``cluster-then-select" process. This process leverages a similarity metric that captures expert performance across different task domains to functionally cluster the experts, and subsequently selects the most representative expert from each cluster based on our proposed Activation Variability Score. Unlike methods that optimize for a single corpus, our proposed Mosaic Pruning ensures that the pruned model retains a functionally complementary set of experts, much like the tiles of a mosaic that together form a complete picture of the original model's capabilities, enabling it to handle diverse downstream tasks.Extensive experiments on various MoE models demonstrate the superiority of our approach. MoP significantly outperforms prior work, achieving a 7.24\% gain on general tasks and 8.92\% on specialized tasks like math reasoning and code generation.

---

### CropVLM: Learning to Zoom for Fine-Grained Vision-Language Perception

Authors: Miguel Carvalho, Helder Dias, Bruno Martins
Published: 2025-11-25
Categories: cs.CV, cs.AI, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2511.19820v2

Vision-Language Models (VLMs) often struggle with tasks that require fine-grained image understanding, such as scene-text recognition or document analysis, due to perception limitations and visual fragmentation. To address these challenges, we introduce CropVLM as an external low-cost method for boosting performance, enabling VLMs to dynamically ''zoom in'' on relevant image regions, enhancing their ability to capture fine details. CropVLM is trained using reinforcement learning, without using human-labeled bounding boxes as a supervision signal, and without expensive synthetic evaluations. The model is trained once and can be paired with both open-source and proprietary VLMs to improve their performance. Our approach delivers significant improvements on tasks that require high-resolution image understanding, notably for benchmarks that are out-of-domain for the target VLM, without modifying or fine-tuning the VLM, thus avoiding catastrophic forgetting.

---

### Language-Independent Sentiment Labelling with Distant Supervision: A Case Study for English, Sepedi and Setswana

Authors: Koena Ronny Mabokela, Tim Schlippe, Mpho Raborife, Turgay Celik
Published: 2025-11-25
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2511.19818v1

Sentiment analysis is a helpful task to automatically analyse opinions and emotions on various topics in areas such as AI for Social Good, AI in Education or marketing. While many of the sentiment analysis systems are developed for English, many African languages are classified as low-resource languages due to the lack of digital language resources like text labelled with corresponding sentiment classes. One reason for that is that manually labelling text data is time-consuming and expensive. Consequently, automatic and rapid processes are needed to reduce the manual effort as much as possible making the labelling process as efficient as possible. In this paper, we present and analyze an automatic language-independent sentiment labelling method that leverages information from sentiment-bearing emojis and words. Our experiments are conducted with tweets in the languages English, Sepedi and Setswana from SAfriSenti, a multilingual sentiment corpus for South African languages. We show that our sentiment labelling approach is able to label the English tweets with an accuracy of 66%, the Sepedi tweets with 69%, and the Setswana tweets with 63%, so that on average only 34% of the automatically generated labels remain to be corrected.

---

### Breaking Bad: Norms for Valence, Arousal, and Dominance for over 10k English Multiword Expressions

Authors: Saif M. Mohammad
Published: 2025-11-25
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.19816v1

Factor analysis studies have shown that the primary dimensions of word meaning are Valence (V), Arousal (A), and Dominance (D). Existing lexicons such as the NRC VAD Lexicon, published in 2018, include VAD association ratings for words. Here, we present a complement to it, which has human ratings of valence, arousal, and dominance for 10k English Multiword Expressions (MWEs) and their constituent words. We also increase the coverage of unigrams, especially words that have become more common since 2018. In all, the new NRC VAD Lexicon v2 now has entries for 10k MWEs and 25k words, in addition to the entries in v1. We show that the associations are highly reliable. We use the lexicon to examine emotional characteristics of MWEs, including: 1. The degree to which MWEs (idioms, noun compounds, and verb particle constructions) exhibit strong emotionality; 2. The degree of emotional compositionality in MWEs. The lexicon enables a wide variety of research in NLP, Psychology, Public Health, Digital Humanities, and Social Sciences. The NRC VAD Lexicon v2 is freely available through the project webpage: http://saifmohammad.com/WebPages/nrc-vad.html

---

### Time-Varying Network Driver Estimation (TNDE) Quantifies Stage-Specific Regulatory Effects From Single-Cell Snapshots

Authors: Jiaxin Li, Shanjun Mao
Published: 2025-11-25
Categories: q-bio.MN, stat.AP, stat.ML
arXiv: https://arxiv.org/abs/2511.19813v1

Identifying key driver genes governing biological processes such as development and disease progression remains a challenge. While existing methods can reconstruct cellular trajectories or infer static gene regulatory networks (GRNs), they often fail to quantify time-resolved regulatory effects within specific temporal windows. Here, we present Time-varying Network Driver Estimation (TNDE), a computational framework quantifying dynamic gene driver effects from single-cell snapshot data under a linear Markov assumption. TNDE leverages a shared graph attention encoder to preserve the local topological structure of the data. Furthermore, by incorporating partial optimal transport, TNDE accounts for unmatched cells arising from proliferation or apoptosis, thereby enabling trajectory alignment in non-equilibrium processes. Benchmarking on simulated datasets demonstrates that TNDE outperforms existing baseline methods across diverse complex regulatory scenarios. Applied to mouse erythropoiesis data, TNDE identifies stage-specific driver genes, the functional relevance of which is corroborated by biological validation. TNDE offers an effective quantitative tool for dissecting dynamic regulatory mechanisms underlying complex biological processes.

---

### TPSO: Training-Free Diverse Image Generation via Semantic Prompt Embedding Optimization

Authors: Debin Meng, Chen Jin, Zheng Gao, Yanran Li, Ioannis Patras, Georgios Tzimiropoulos
Published: 2025-11-25
Categories: cs.CV, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2511.19811v2

Image diversity remains a fundamental challenge for text-to-image diffusion models. Low-diversity generation often leads to repetitive outputs, increasing sampling redundancy and hindering both creative exploration and downstream applications. A key factor is the tendency of diffusion models to collapse toward strong modes in the learned distribution. Existing attempts to improve diversity, such as steering-based guidance, often introduce distortions that degrade image quality. To address this issue, we propose Token-Prompt Embedding Space Optimization (TPSO), a training-free and model-agnostic module. TPSO introduces learnable parameters to explore underrepresented regions of the token embedding space, reducing the tendency to repeatedly sample from strong modes of the distribution. Meanwhile, a prompt-level semantic constraint regulates distribution shifts, preventing quality degradation while preserving semantic fidelity. Extensive experiments on MS-COCO across three representative diffusion backbones demonstrate that TPSO substantially improves diversity, boosting performance from 1.10 to 4.18, while maintaining image quality with only a modest inference-time overhead of 3.6% to 8.9%. Code is available at: https://github.com/Open-Debin/TPSO.

---

### Provably Outlier-resistant Semi-parametric Regression for Transferable Calibration of Low-cost Air-quality Sensors

Authors: Divyansh Chaurasia, Manoj Daram, Roshan Kumar, Nihal Thukarama Rao, Vipul Sangode, Pranjal Srivastava, Avnish Tripathi, Shoubhik Chakraborty, Akanksha, Ambasht Kumar, Davender Sethi, Sachchida Nand Tripathi et al.
Published: 2025-11-25
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2511.19810v1

We present a case study for the calibration of Low-cost air-quality (LCAQ) CO sensors from one of the largest multi-site-multi-season-multi-sensor-multi-pollutant mobile air-quality monitoring network deployments in India. LCAQ sensors have been shown to play a critical role in the establishment of dense, expansive air-quality monitoring networks and combating elevated pollution levels. The calibration of LCAQ sensors against regulatory-grade monitors is an expensive, laborious and time-consuming process, especially when a large number of sensors are to be deployed in a geographically diverse layout. In this work, we present the RESPIRE technique to calibrate LCAQ sensors to detect ambient CO (Carbon Monoxide) levels. RESPIRE offers specific advantages over baseline calibration methods popular in literature, such as improved prediction in cross-site, cross-season, and cross-sensor settings. RESPIRE offers a training algorithm that is provably resistant to outliers and an explainable model with the ability to flag instances of model overfitting. Empirical results are presented based on data collected during an extensive deployment spanning four sites, two seasons and six sensor packages. RESPIRE code is available at https://github.com/purushottamkar/respire.

---

### Learning to Clean: Reinforcement Learning for Noisy Label Correction

Authors: Marzi Heidari, Hanping Zhang, Yuhong Guo
Published: 2025-11-25
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.19808v1

The challenge of learning with noisy labels is significant in machine learning, as it can severely degrade the performance of prediction models if not addressed properly. This paper introduces a novel framework that conceptualizes noisy label correction as a reinforcement learning (RL) problem. The proposed approach, Reinforcement Learning for Noisy Label Correction (RLNLC), defines a comprehensive state space representing data and their associated labels, an action space that indicates possible label corrections, and a reward mechanism that evaluates the efficacy of label corrections. RLNLC learns a deep feature representation based policy network to perform label correction through reinforcement learning, utilizing an actor-critic method. The learned policy is subsequently deployed to iteratively correct noisy training labels and facilitate the training of the prediction model. The effectiveness of RLNLC is demonstrated through extensive experiments on multiple benchmark datasets, where it consistently outperforms existing state-of-the-art techniques for learning with noisy labels.

---

### Reading Between the Lines: Abstaining from VLM-Generated OCR Errors via Latent Representation Probes

Authors: Jihan Yao, Achin Kulshrestha, Nathalie Rauschmayr, Reed Roberts, Banghua Zhu, Yulia Tsvetkov, Federico Tombari
Published: 2025-11-25
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.19806v1

As VLMs are deployed in safety-critical applications, their ability to abstain from answering when uncertain becomes crucial for reliability, especially in Scene Text Visual Question Answering (STVQA) tasks. For example, OCR errors like misreading "50 mph" as "60 mph" could cause severe traffic accidents. This leads us to ask: Can VLMs know when they can't see? Existing abstention methods suggest pessimistic answers: they either rely on miscalibrated output probabilities or require semantic agreement unsuitable for OCR tasks. However, this failure may indicate we are looking in the wrong place: uncertainty signals could be hidden in VLMs' internal representations. Building on this insight, we propose Latent Representation Probing (LRP): training lightweight probes on hidden states or attention patterns. We explore three probe designs: concatenating representations across all layers, aggregating attention over visual tokens, and ensembling single layer probes by majority vote. Experiments on four benchmarks across image and video modalities show LRP improves abstention accuracy by 7.6\% over best baselines. Our analysis reveals: probes generalize across various uncertainty sources and datasets, and optimal signals emerge from intermediate rather than final layers. This establishes a principled framework for building deployment-ready AI systems by detecting confidence signals from internal states rather than unreliable outputs.

---

### Latent-space metrics for Complex-Valued VAE out-of-distribution detection under radar clutter

Authors: Y. A. Rouzoumka, E. Terreaux, C. Morisseau, J. -P. Ovarlez, C. Ren
Published: 2025-11-25
Categories: eess.SP, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2511.19805v1

We investigate complex-valued Variational AutoEncoders (CVAE) for radar Out-Of-Distribution (OOD) detection in complex radar environments. We proposed several detection metrics: the reconstruction error of CVAE (CVAE-MSE), the latent-based scores (Mahalanobis, Kullback-Leibler divergence (KLD)), and compared their performance against the classical ANMF-Tyler detector (ANMF-FP). The performance of all these detectors is analyzed on synthetic and experimental radar data, showing the advantages and the weaknesses of each detector.

---

### Scalable Data Attribution via Forward-Only Test-Time Inference

Authors: Sibo Ma, Julian Nyarko
Published: 2025-11-25
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.19803v1

Data attribution seeks to trace model behavior back to the training examples that shaped it, enabling debugging, auditing, and data valuation at scale. Classical influence-function methods offer a principled foundation but remain impractical for modern networks because they require expensive backpropagation or Hessian inversion at inference. We propose a data attribution method that preserves the same first-order counterfactual target while eliminating per-query backward passes. Our approach simulates each training example's parameter influence through short-horizon gradient propagation during training and later reads out attributions for any query using only forward evaluations. This design shifts computation from inference to simulation, reflecting real deployment regimes where a model may serve billions of user queries but originate from a fixed, finite set of data sources (for example, a large language model trained on diverse corpora while compensating a specific publisher such as the New York Times). Empirically, on standard MLP benchmarks, our estimator matches or surpasses state-of-the-art baselines such as TRAK on standard attribution metrics (LOO and LDS) while offering orders-of-magnitude lower inference cost. By combining influence-function fidelity with first-order scalability, our method provides a theoretical framework for practical, real-time data attribution in large pretrained models.

---

### Replication Study: Federated Text-Driven Prompt Generation for Vision-Language Models

Authors: Suraj Prasad, Anubha Pant
Published: 2025-11-24
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2602.18439v1

Vision-language models like CLIP have demonstrated remarkable zero-shot capabilities, yet their adaptation to federated learning scenarios presents significant challenges, particularly regarding generalization to unseen classes. The original FedTPG paper \cite{Qiu2024} addresses this limitation by introducing a text driven prompt generation network that dynamically creates prompts conditioned on class names, enabling better cross-class generalization in federated settings. In this work, we present a faithful replication study of FedTPG, evaluating the pre-trained model on six diverse vision datasets: Caltech101, Oxford Flowers, FGVC Aircraft, Oxford Pets, Food-101, and DTD. Our evaluation achieves results within 0.2\% of the original paper's reported accuracies, with an average accuracy of 74.58\% on seen (base) classes and 76.00\% on unseen (new) classes, demonstrating a +1.43 percentage point improvement in generalization. These results validate the original paper's core claims: (1) text-driven prompt generation enables superior generalization to unseen classes compared to static prompt learning methods, and (2) federated training of prompt generators maintains high performance across diverse visual domains without sharing private data. Our successful replication confirms the robustness and reproducibility of the FedTPG approach.

---

### Intelligent Power Grid Design Review via Active Perception-Enabled Multimodal Large Language Models

Authors: Taoliang Tan, Chengwei Ma, Zhen Tian, Zhao Lin, Dongdong Li, Si Shi
Published: 2025-11-24
Categories: cs.CV, cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2601.14261v1

The intelligent review of power grid engineering design drawings is crucial for power system safety. However, current automated systems struggle with ultra-high-resolution drawings due to high computational demands, information loss, and a lack of holistic semantic understanding for design error identification. This paper proposes a novel three-stage framework for intelligent power grid drawing review, driven by pre-trained Multimodal Large Language Models (MLLMs) through advanced prompt engineering. Mimicking the human expert review process, the first stage leverages an MLLM for global semantic understanding to intelligently propose domain-specific semantic regions from a low-resolution overview. The second stage then performs high-resolution, fine-grained recognition within these proposed regions, acquiring detailed information with associated confidence scores. In the final stage, a comprehensive decision-making module integrates these confidence-aware results to accurately diagnose design errors and provide a reliability assessment. Preliminary results on real-world power grid drawings demonstrate our approach significantly enhances MLLM's ability to grasp macroscopic semantic information and pinpoint design errors, showing improved defect discovery accuracy and greater reliability in review judgments compared to traditional passive MLLM inference. This research offers a novel, prompt-driven paradigm for intelligent and reliable power grid drawing review.

---

### LiveVectorLake: A Real-Time Versioned Knowledge Base Architecture for Streaming Vector Updates and Temporal Retrieval

Authors: Tarun Prajapati
Published: 2025-11-24
Categories: cs.IR, cs.AI, cs.DB
arXiv: https://arxiv.org/abs/2601.05270v1

Modern Retrieval-Augmented Generation (RAG) systems struggle with a fundamental architectural tension: vector indices are optimized for query latency but poorly handle continuous knowledge updates, while data lakes excel at versioning but introduce query latency penalties. We introduce LiveVectorLake, a dual-tier temporal knowledge base architecture that enables real-time semantic search on current knowledge while maintaining complete version history for compliance, auditability, and point-in-time retrieval. The system introduces three core architectural contributions: (1) Content-addressable chunk-level synchronization using SHA-256 hashing for deterministic change detection without external state tracking; (2) Dual-tier storage separating hot-tier vector indices (Milvus with HNSW) from cold-tier columnar versioning (Delta Lake with Parquet), optimizing query latency and storage cost independently; (3) Temporal query routing enabling point-in-time knowledge retrieval via delta-versioning with ACID consistency across tiers. Evaluation on a 100-document corpus versioned across five time points demonstrates: (i) 10-15% re-processing of content during updates compared to 100% for full re-indexing; (ii) sub-100ms retrieval latency on current knowledge; (iii) sub-2s latency for temporal queries across version history; and (iv) storage cost optimization through hot/cold tier separation (only current chunks in expensive vector indices). The approach enables production RAG deployments requiring simultaneous optimization for query performance, update efficiency, and regulatory compliance. Code and resources: (https://github.com/praj-tarun/LiveVectorLake)

---

### Automated Fault Detection in 5G Core Networks Using Large Language Models

Authors: Parsa Hatami, Ahmadreza Majlesara, Ali Majlesi, Babak Hossein Khalaj
Published: 2025-11-24
Categories: cs.NI, cs.AI
arXiv: https://arxiv.org/abs/2512.19697v1

With the rapid growth of data volume in modern telecommunication networks and the continuous expansion of their scale, maintaining high reliability has become a critical requirement. These networks support a wide range of applications and services, including highly sensitive and mission-critical ones, which demand rapid and accurate detection and resolution of network errors. Traditional fault-diagnosis methods are no longer efficient for such complex environments.\cite{b1} In this study, we leverage Large Language Models (LLMs) to automate network fault detection and classification. Various types of network errors were intentionally injected into a Kubernetes-based test network, and data were collected under both healthy and faulty conditions. The dataset includes logs from different network components (pods), along with complementary data such as system descriptions, events, Round Trip Time (RTT) tests, and pod status information. The dataset covers common fault types such as pod failure, pod kill, network delay, network loss, and disk I/O failures. We fine-tuned the GPT-4.1 nano model via its API on this dataset, resulting in a significant improvement in fault-detection accuracy compared to the base model. These findings highlight the potential of LLM-based approaches for achieving closed-loop, and operator-free fault management, which can enhance network reliability and reduce downtime-related operational costs for service providers.

---

### Large Language Models as Search Engines: Societal Challenges

Authors: Zacchary Sadeddine, Winston Maxwell, Gaël Varoquaux, Fabian M. Suchanek
Published: 2025-11-24
Categories: cs.CY, cs.CL
arXiv: https://arxiv.org/abs/2512.08946v1

Large Language Models (LLMs) may one day replace search engines as the primary portal to information on the Web. In this article, we investigate the societal challenges that such a change could bring. We focus on the roles of LLM Providers, Content Creators, and End Users, and identify 15 types of challenges. With each, we show current mitigation strategies -- both from the technical perspective and the legal perspective. We also discuss the impact of each challenge and point out future research opportunities.

---

### ThreadWeaver: Adaptive Threading for Efficient Parallel Reasoning in Language Models

Authors: Long Lian, Sida Wang, Felix Juefei-Xu, Tsu-Jui Fu, Xiuyu Li, Adam Yala, Trevor Darrell, Alane Suhr, Yuandong Tian, Xi Victoria Lin
Published: 2025-11-24
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2512.07843v2

Scaling inference-time computation has enabled Large Language Models (LLMs) to achieve strong reasoning performance, but their inherently sequential decoding incurs substantial latency, motivating parallelization of the generation process. However, existing parallel reasoning approaches suffer from performance degradation compared to their sequential counterparts, and often rely on specialized inference engines. We introduce ThreadWeaver, a framework for adaptive parallel reasoning that matches the accuracy of comparably sized sequential reasoning models while significantly reducing inference latency via three key innovations: 1) a two-stage parallel trajectory generator that produces high-quality parallel chain-of-thought data for supervised fine-tuning; 2) a trie-based rollout design that enables parallel reasoning on any off-the-shelf autoregressive inference engine; and 3) a parallelization-aware reinforcement learning framework that trains the model to balance reasoning accuracy with effective parallelization. Across six challenging math reasoning benchmarks, ThreadWeaver trained on top of Qwen3-8B achieves performance on par with cutting-edge sequential reasoning models (79.9% on AIME24 and 71.9% on average) while delivering up to 1.53x speedup in token latency, establishing a new Pareto frontier between accuracy and efficiency.

---

### LATTICE: Democratize High-Fidelity 3D Generation at Scale

Authors: Zeqiang Lai, Yunfei Zhao, Zibo Zhao, Haolin Liu, Qingxiang Lin, Jingwei Huang, Chunchao Guo, Xiangyu Yue
Published: 2025-11-24
Categories: cs.GR, cs.CV
arXiv: https://arxiv.org/abs/2512.03052v1

We present LATTICE, a new framework for high-fidelity 3D asset generation that bridges the quality and scalability gap between 3D and 2D generative models. While 2D image synthesis benefits from fixed spatial grids and well-established transformer architectures, 3D generation remains fundamentally more challenging due to the need to predict both spatial structure and detailed geometric surfaces from scratch. These challenges are exacerbated by the computational complexity of existing 3D representations and the lack of structured and scalable 3D asset encoding schemes. To address this, we propose VoxSet, a semi-structured representation that compresses 3D assets into a compact set of latent vectors anchored to a coarse voxel grid, enabling efficient and position-aware generation. VoxSet retains the simplicity and compression advantages of prior VecSet methods while introducing explicit structure into the latent space, allowing positional embeddings to guide generation and enabling strong token-level test-time scaling. Built upon this representation, LATTICE adopts a two-stage pipeline: first generating a sparse voxelized geometry anchor, then producing detailed geometry using a rectified flow transformer. Our method is simple at its core, but supports arbitrary resolution decoding, low-cost training, and flexible inference schemes, achieving state-of-the-art performance on various aspects, and offering a significant step toward scalable, high-quality 3D asset creation.

---

### Deep Research: A Systematic Survey

Authors: Zhengliang Shi, Yiqun Chen, Haitao Li, Weiwei Sun, Shiyu Ni, Yougang Lyu, Run-Ze Fan, Bowen Jin, Yixuan Weng, Minjun Zhu, Qiujie Xie, Xinyu Guo et al.
Published: 2025-11-24
Categories: cs.CL, cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2512.02038v1

Large language models (LLMs) have rapidly evolved from text generators into powerful problem solvers. Yet, many open tasks demand critical thinking, multi-source, and verifiable outputs, which are beyond single-shot prompting or standard retrieval-augmented generation. Recently, numerous studies have explored Deep Research (DR), which aims to combine the reasoning capabilities of LLMs with external tools, such as search engines, thereby empowering LLMs to act as research agents capable of completing complex, open-ended tasks. This survey presents a comprehensive and systematic overview of deep research systems, including a clear roadmap, foundational components, practical implementation techniques, important challenges, and future directions. Specifically, our main contributions are as follows: (i) we formalize a three-stage roadmap and distinguish deep research from related paradigms; (ii) we introduce four key components: query planning, information acquisition, memory management, and answer generation, each paired with fine-grained sub-taxonomies; (iii) we summarize optimization techniques, including prompting, supervised fine-tuning, and agentic reinforcement learning; and (iv) we consolidate evaluation criteria and open challenges, aiming to guide and facilitate future development. As the field of deep research continues to evolve rapidly, we are committed to continuously updating this survey to reflect the latest progress in this area.

---

### ProvRain: Rain-Adaptive Denoising and Vehicle Detection via MobileNet-UNet and Faster R-CNN

Authors: Aswinkumar Varathakumaran, Nirmala Paramanandham
Published: 2025-11-24
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.00073v1

Provident vehicle detection has a lot of scope in the detection of vehicle during night time. The extraction of features other than the headlamps of vehicles allows us to detect oncoming vehicles before they appear directly on the camera. However, it faces multiple issues especially in the field of night vision, where a lot of noise caused due to weather conditions such as rain or snow as well as camera conditions. This paper focuses on creating a pipeline aimed at dealing with such noise while at the same time maintaining the accuracy of provident vehicular detection. The pipeline in this paper, ProvRain, uses a lightweight MobileNet-U-Net architecture tuned to generalize to robust weather conditions by using the concept of curricula training. A mix of synthetic as well as available data from the PVDN dataset is used for this. This pipeline is compared to the base Faster RCNN architecture trained on the PVDN dataset to see how much the addition of a denoising architecture helps increase the detection model's performance in rainy conditions. The system boasts an 8.94\% increase in accuracy and a 10.25\% increase in recall in the detection of vehicles in rainy night time frames. Similarly, the custom MobileNet-U-Net architecture that was trained also shows a 10-15\% improvement in PSNR, a 5-6\% increase in SSIM, and upto a 67\% reduction in perceptual error (LPIPS) compared to other transformer approaches.

---

### A CNN-Based Technique to Assist Layout-to-Generator Conversion for Analog Circuits

Authors: Sungyu Jeong, Minsu Kim, Byungsub Kim
Published: 2025-11-24
Categories: cs.AR, cs.AI, cs.LG, eess.IV
arXiv: https://arxiv.org/abs/2512.00070v1

We propose a technique to assist in converting a reference layout of an analog circuit into the procedural layout generator by efficiently reusing available generators for sub-cell creation. The proposed convolutional neural network (CNN) model automatically detects sub-cells that can be generated by available generator scripts in the library, and suggests using them in the hierarchically correct places of the generator software. In experiments, the CNN model examined sub-cells of a high-speed wireline receiver that has a total of 4,885 sub-cell instances including different 145 sub-cell designs. The CNN model classified the sub-cell instances into 51 generatable and one not-generatable classes. One not-generatable class indicates that no available generator can generate the classified sub-cell. The CNN model achieved 99.3% precision in examining the 145 different sub-cell designs. The CNN model greatly reduced the examination time to 18 seconds from 88 minutes required in manual examination. Also, the proposed CNN model could correctly classify unfamiliar sub-cells that are very different from the training dataset.

---

### Enhancing Cognitive Robotics with Commonsense through LLM-Generated Preconditions and Subgoals

Authors: Ohad Bachner, Bar Gamliel
Published: 2025-11-24
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2512.00069v1

Robots often fail at everyday tasks because instructions skip commonsense details like hidden preconditions and small subgoals. Traditional symbolic planners need these details to be written explicitly, which is time consuming and often incomplete. In this project we combine a Large Language Model with symbolic planning. Given a natural language task, the LLM suggests plausible preconditions and subgoals. We translate these suggestions into a formal planning model and execute the resulting plan in simulation. Compared to a baseline planner without the LLM step, our system produces more valid plans, achieves a higher task success rate, and adapts better when the environment changes. These results suggest that adding LLM commonsense to classical planning can make robot behavior in realistic scenarios more reliable.

---

### Satellite to Street : Disaster Impact Estimator

Authors: Sreesritha Sai, Sai Venkata Suma Sreeja, Sai Sri Deepthi, Nikhil
Published: 2025-11-24
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2512.00065v3

Accurate assessment of post-disaster damage is essential for prioritizing emergency response, yet current practices rely heavily on manual interpretation of satellite imagery.This approach is time-consuming, subjective, and difficult to scale during large-area disasters. Although recent deep-learning models for semantic segmentation and change detection have improved automation, many of them still struggle to capture subtle structural variations and often perform poorly when dealing with highly imbalanced datasets, where undamaged buildings dominate. This thesis introduces Satellite-to-Street:Disaster Impact Estimator, a deep-learning framework that produces detailed, pixel-level damage maps by analyzing pre and post-disaster satellite images together. The model is built on a modified dual-input U-Net architecture that strengthens feature fusion between both images, allowing it to detect not only small, localized changes but also broader contextual patterns across the scene. To address the imbalance between damage categories, a class-aware weighted loss function is used, which helps the model better recognize major and destroyed structures. A consistent preprocessing pipeline is employed to align image pairs, standardize resolutions, and prepare the dataset for training. Experiments conducted on publicly available disaster datasets show that the proposed framework achieves better classification of damaged regions compared to conventional segmentation networks.The generated damage maps provide faster and objective method for analyzing disaster impact, working alongside expert judgment rather than replacing it. In addition to identifying which areas are damaged, the system is capable of distinguishing different levels of severity, ranging from slight impact to complete destruction. This provides a more detailed and practical understanding of how the disaster has affected each region.

---

### SpeedAug: Policy Acceleration via Tempo-Enriched Policy and RL Fine-Tuning

Authors: Taewook Nam, Junmo Cho, Youngsoo Jang, Sung Ju Hwang
Published: 2025-11-24
Categories: cs.RO, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.00062v2

Robotic policy learning for complex real-world manipulation tasks has seen rapid recent progress, enabled in large part by the ability to collect demonstrations through human operation. However, policies trained from such demonstrations often execute tasks far more slowly than the robot's physical capabilities, as demonstration data is collected under practical constraints that favor conservative, success-oriented trajectories over execution speed. Existing policy acceleration methods determine execution tempo through data preprocessing or heuristic rules, rather than learning execution speed optimized for the task. In this paper, we propose SpeedAug, a policy acceleration framework that enables policies to learn task-optimal execution tempo via reinforcement learning (RL). SpeedAug first learns a tempo-enriched prior policy from speed-augmented demonstrations that captures diverse execution tempos. Building on this tempo-enriched prior, RL fine-tuning guides exploration to refine action trajectories and optimize execution tempo efficiently. Experiments on robotic manipulation benchmarks demonstrate that SpeedAug substantially improves the sample efficiency of policy acceleration while maintaining high success rates, achieving fast and stable task execution. Applied to a real-world manipulation task, SpeedAug improves task throughput by 1.8x using only 16 minutes of online interactions without compromising the success rate.

---

### Factors That Support Grounded Responses in LLM Conversations: A Rapid Review

Authors: Gabriele Cesar Iwashima, Claudia Susie Rodrigues, Claudio Dipolitto, Geraldo Xexéo
Published: 2025-11-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2511.21762v1

Large language models (LLMs) may generate outputs that are misaligned with user intent, lack contextual grounding, or exhibit hallucinations during conversation, which compromises the reliability of LLM-based applications. This review aimed to identify and analyze techniques that align LLM responses with conversational goals, ensure grounding, and reduce hallucination and topic drift. We conducted a Rapid Review guided by the PRISMA framework and the PICO strategy to structure the search, filtering, and selection processes. The alignment strategies identified were categorized according to the LLM lifecycle phase in which they operate: inference-time, post-training, and reinforcement learning-based methods. Among these, inference-time approaches emerged as particularly efficient, aligning outputs without retraining while supporting user intent, contextual grounding, and hallucination mitigation. The reviewed techniques provided structured mechanisms for improving the quality and reliability of LLM responses across key alignment objectives.

---

### LLMs for Low-Resource Dialect Translation Using Context-Aware Prompting: A Case Study on Sylheti

Authors: Tabia Tanzin Prama, Christopher M. Danforth, Peter Sheridan Dodds
Published: 2025-11-24
Categories: cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2511.21761v1

Large Language Models (LLMs) have demonstrated strong translation abilities through prompting, even without task-specific training. However, their effectiveness in dialectal and low-resource contexts remains underexplored. This study presents the first systematic investigation of LLM-based machine translation (MT) for Sylheti, a dialect of Bangla that is itself low-resource. We evaluate five advanced LLMs (GPT-4.1, GPT-4.1, LLaMA 4, Grok 3, and DeepSeek V3.2) across both translation directions (Bangla $\Leftrightarrow$ Sylheti), and find that these models struggle with dialect-specific vocabulary. To address this, we introduce Sylheti-CAP (Context-Aware Prompting), a three-step framework that embeds a linguistic rulebook, a dictionary (2{,}260 core vocabulary items and idioms), and an authenticity check directly into prompts. Extensive experiments show that Sylheti-CAP consistently improves translation quality across models and prompting strategies. Both automatic metrics and human evaluations confirm its effectiveness, while qualitative analysis reveals notable reductions in hallucinations, ambiguities, and awkward phrasing, establishing Sylheti-CAP as a scalable solution for dialectal and low-resource MT. Dataset link: \href{https://github.com/TabiaTanzin/LLMs-for-Low-Resource-Dialect-Translation-Using-Context-Aware-Prompting-A-Case-Study-on-Sylheti.git}{https://github.com/TabiaTanzin/LLMs-for-Low-Resource-Dialect-Translation-Using-Context-Aware-Prompting-A-Case-Study-on-Sylheti.git}

---

### fMRI-LM: Towards a Universal Foundation Model for Language-Aligned fMRI Understanding

Authors: Yuxiang Wei, Yanteng Zhang, Xi Xiao, Chengxuan Qian, Tianyang Wang, Vince D. Calhoun
Published: 2025-11-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2511.21760v4

Recent advances in multimodal large language models (LLMs) have enabled unified reasoning across images, audio, and video, but extending such capability to brain imaging remains largely unexplored. Bridging this gap is essential to link neural activity with semantic cognition and to develop cross-modal brain representations. To this end, we present fMRI-LM, a foundational model that bridges functional MRI (fMRI) and language through a three-stage framework. In Stage 1, we learn a neural tokenizer that maps fMRI into discrete tokens embedded in a language-consistent space. In Stage 2, a pretrained LLM is adapted to jointly model fMRI tokens and text, treating brain activity as a sequence that can be temporally predicted and linguistically described. To overcome the lack of natural fMRI-text pairs, we construct a large descriptive corpus that translates diverse imaging-based features into structured textual descriptors, capturing the low-level organization of fMRI signals. In Stage 3, we perform multi-task, multi-paradigm instruction tuning to endow fMRI-LM with high-level semantic understanding, supporting diverse downstream applications. Across various benchmarks, fMRI-LM achieves strong zero-shot and few-shot performance, and adapts efficiently with parameter-efficient tuning (LoRA), establishing a scalable pathway toward a language-aligned, universal model for structural and semantic understanding of fMRI.

---

### Orchestrating Dual-Boundaries: An Arithmetic Intensity Inspired Acceleration Framework for Diffusion Language Models

Authors: Linye Wei, Wenjue Chen, Pingzhi Tang, Xiaotian Guo, Le Ye, Runsheng Wang, Meng Li
Published: 2025-11-24
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2511.21759v2

Diffusion-based large language models (dLLMs) have recently gained significant attention for their exceptional performance and inherent potential for parallel decoding. Existing frameworks further enhance its inference efficiency by enabling KV caching. However, its bidirectional attention mechanism necessitates periodic cache refreshes that interleave prefill and decoding phases, both contributing substantial inference cost and constraining achievable speedup. Inspired by the heterogeneous arithmetic intensity of the prefill and decoding phases, we propose ODB-dLLM, a framework that orchestrates dual-boundaries to accelerate dLLM inference. In the prefill phase, we find that the predefined fixed response length introduces heavy yet redundant computational overhead, which affects efficiency. To alleviate this, ODB-dLLM incorporates an adaptive length prediction mechanism that progressively reduces prefill overhead and unnecessary computation. In the decoding phase, we analyze the computational characteristics of dLLMs and propose a dLLM-specific jump-share speculative decoding method to enhance efficiency by reducing the number of decoding iterations. Experimental results demonstrate that ODB-dLLM achieves 46-162x and 2.63-6.30x speedups over the baseline dLLM and Fast-dLLM, respectively, while simultaneously mitigating the accuracy degradation in existing acceleration frameworks.

---

### A Longitudinal Measurement of Privacy Policy Evolution for Large Language Models

Authors: Zhen Tao, Shidong Pan, Zhenchang Xing, Emily Black, Talia Gillis, Chunyang Chen
Published: 2025-11-24
Categories: cs.CR, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2511.21758v1

Large language model (LLM) services have been rapidly integrated into people's daily lives as chatbots and agentic systems. They are nourished by collecting rich streams of data, raising privacy concerns around excessive collection of sensitive personal information. Privacy policies are the fundamental mechanism for informing users about data practices in modern information privacy paradigm. Although traditional web and mobile policies are well studied, the privacy policies of LLM providers, their LLM-specific content, and their evolution over time remain largely underexplored. In this paper, we present the first longitudinal empirical study of privacy policies for mainstream LLM providers worldwide. We curate a chronological dataset of 74 historical privacy policies and 115 supplemental privacy documents from 11 LLM providers across 5 countries up to August 2025, and extract over 3,000 sentence-level edits between consecutive policy versions. We compare LLM privacy policies to those of other software formats, propose a taxonomy tailored to LLM privacy policies, annotate policy edits and align them with a timeline of key LLM ecosystem events. Results show they are substantially longer, demand college-level reading ability, and remain highly vague. Our taxonomy analysis reveals patterns in how providers disclose LLM-specific practices and highlights regional disparities in coverage. Policy edits are concentrated in first-party data collection and international/specific-audience sections, and that product releases and regulatory actions are the primary drivers, shedding light on the status quo and the evolution of LLM privacy policies.

---

### Medical Malice: A Dataset for Context-Aware Safety in Healthcare LLMs

Authors: Andrew Maranhão Ventura D'addario
Published: 2025-11-24
Categories: cs.CY, cs.AI, cs.CL, cs.CR
arXiv: https://arxiv.org/abs/2511.21757v1

The integration of Large Language Models (LLMs) into healthcare demands a safety paradigm rooted in \textit{primum non nocere}. However, current alignment techniques rely on generic definitions of harm that fail to capture context-dependent violations, such as administrative fraud and clinical discrimination. To address this, we introduce Medical Malice: a dataset of 214,219 adversarial prompts calibrated to the regulatory and ethical complexities of the Brazilian Unified Health System (SUS). Crucially, the dataset includes the reasoning behind each violation, enabling models to internalize ethical boundaries rather than merely memorizing a fixed set of refusals. Using an unaligned agent (Grok-4) within a persona-driven pipeline, we synthesized high-fidelity threats across seven taxonomies, ranging from procurement manipulation and queue-jumping to obstetric violence. We discuss the ethical design of releasing these "vulnerability signatures" to correct the information asymmetry between malicious actors and AI developers. Ultimately, this work advocates for a shift from universal to context-aware safety, providing the necessary resources to immunize healthcare AI against the nuanced, systemic threats inherent to high-stakes medical environments -- vulnerabilities that represent the paramount risk to patient safety and the successful integration of AI in healthcare systems.

---

### Dissecting the Ledger: Locating and Suppressing "Liar Circuits" in Financial Large Language Models

Authors: Soham Mirajkar
Published: 2025-11-24
Categories: cs.CL, cs.CE
arXiv: https://arxiv.org/abs/2511.21756v1

Large Language Models (LLMs) are increasingly deployed in high-stakes financial domains, yet they suffer from specific, reproducible hallucinations when performing arithmetic operations. Current mitigation strategies often treat the model as a black box. In this work, we propose a mechanistic approach to intrinsic hallucination detection. By applying Causal Tracing to the GPT-2 XL architecture on the ConvFinQA benchmark, we identify a dual-stage mechanism for arithmetic reasoning: a distributed computational scratchpad in middle layers (L12-L30) and a decisive aggregation circuit in late layers (specifically Layer 46). We verify this mechanism via an ablation study, demonstrating that suppressing Layer 46 reduces the model's confidence in hallucinatory outputs by 81.8%. Furthermore, we demonstrate that a linear probe trained on this layer generalizes to unseen financial topics with 98% accuracy, suggesting a universal geometry of arithmetic deception.

---

### Who Owns the Knowledge? Copyright, GenAI, and the Future of Academic Publishing

Authors: Dmitry Kochetkov
Published: 2025-11-24
Categories: cs.DL, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2511.21755v2

The integration of generative artificial intelligence (GenAI) and large language models (LLMs) into scientific research and higher education presents a paradigm shift, offering revolutionizing opportunities while simultaneously raising profound ethical, legal, and regulatory questions. This study examines the complex intersection of AI and science, with a specific focus on the challenges posed to copyright law and the principles of open science. The author argues that current regulatory frameworks in key jurisdictions like the United States, China, the European Union, and the United Kingdom, while aiming to foster innovation, contain significant gaps, particularly concerning the use of copyrighted works and open science outputs for AI training. Widely adopted licensing mechanisms, such as Creative Commons, fail to adequately address the nuances of AI training, and the pervasive lack of attribution within AI systems fundamentally challenges established notions of originality. While current doctrine treats AI training as potentially fair use, this paper argues such mechanisms are inadequate and that copyright holders should retain explicit opt-out rights regardless of fair use doctrine. Instead, the author advocates for upholding authors' rights to refuse the use of their works for AI training and proposes that universities assume a leading role in shaping responsible AI governance. The conclusion is that a harmonized international legislative effort is urgently needed to ensure transparency, protect intellectual property, and prevent the emergence of an oligopolistic market structure that could prioritize commercial profit over scientific integrity and equitable knowledge production. This is a substantially expanded and revised version of a work originally presented at the 20th International Conference on Scientometrics & Informetrics (Kochetkov, 2025).

---

### Extracting Disaster Impacts and Impact Related Locations in Social Media Posts Using Large Language Models

Authors: Sameeah Noreen Hameed, Surangika Ranathunga, Raj Prasanna, Kristin Stock, Christopher B. Jones
Published: 2025-11-24
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2511.21753v1

Large-scale disasters can often result in catastrophic consequences on people and infrastructure. Situation awareness about such disaster impacts generated by authoritative data from in-situ sensors, remote sensing imagery, and/or geographic data is often limited due to atmospheric opacity, satellite revisits, and time limitations. This often results in geo-temporal information gaps. In contrast, impact-related social media posts can act as "geo-sensors" during a disaster, where people describe specific impacts and locations. However, not all locations mentioned in disaster-related social media posts relate to an impact. Only the impacted locations are critical for directing resources effectively. e.g., "The death toll from a fire which ripped through the Greek coastal town of #Mati stood at 80, with dozens of people unaccounted for as forensic experts tried to identify victims who were burned alive #Greecefires #AthensFires #Athens #Greece." contains impacted location "Mati" and non-impacted locations "Greece" and "Athens". This research uses Large Language Models (LLMs) to identify all locations, impacts and impacted locations mentioned in disaster-related social media posts. In the process, LLMs are fine-tuned to identify only impacts and impacted locations (as distinct from other, non-impacted locations), including locations mentioned in informal expressions, abbreviations, and short forms. Our fine-tuned model demonstrates efficacy, achieving an F1-score of 0.69 for impact and 0.74 for impacted location extraction, substantially outperforming the pre-trained baseline. These robust results confirm the potential of fine-tuned language models to offer a scalable solution for timely decision-making in resource allocation, situational awareness, and post-disaster recovery planning for responders.

---

### Active Slice Discovery in Large Language Models

Authors: Minhui Zhang, Prahar Ijner, Yoav Wald, Elliot Creager
Published: 2025-11-24
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.20713v1

Large Language Models (LLMs) often exhibit systematic errors on specific subsets of data, known as error slices. For instance, a slice can correspond to a certain demographic, where a model does poorly in identifying toxic comments regarding that demographic. Identifying error slices is crucial to understanding and improving models, but it is also challenging. An appealing approach to reduce the amount of manual annotation required is to actively group errors that are likely to belong to the same slice, while using limited access to an annotator to verify whether the chosen samples share the same pattern of model mistake. In this paper, we formalize this approach as Active Slice Discovery and explore it empirically on a problem of discovering human-defined slices in toxicity classification. We examine the efficacy of active slice discovery under different choices of feature representations and active learning algorithms. On several slices, we find that uncertainty-based active learning algorithms are most effective, achieving competitive accuracy using 2-10% of the available slice membership information, while significantly outperforming baselines.

---

### A Set of Rules for Model Validation

Authors: José Camacho
Published: 2025-11-24
Categories: stat.ME, cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2511.20711v2

The validation of a data-driven model is the process of assessing the model's ability to generalize to new, unseen data in the population of interest. This paper proposes a set of general rules for model validation. These rules are designed to help practitioners create reliable validation plans and report their results transparently. While no validation scheme is flawless, these rules can help practitioners ensure their strategy is sufficient for practical use, openly discuss any limitations of their validation strategy, and report clear, comparable performance metrics.

---

### Are Neuro-Inspired Multi-Modal Vision-Language Models Resilient to Membership Inference Privacy Leakage?

Authors: David Amebley, Sayanton Dibbo
Published: 2025-11-24
Categories: cs.CV, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2511.20710v2

In the age of agentic AI, the growing deployment of multi-modal models (MMs) has introduced new attack vectors that can leak sensitive training data in MMs, causing privacy leakage. This paper investigates a black-box privacy attack, i.e., membership inference attack (MIA) on multi-modal vision-language models (VLMs). State-of-the-art research analyzes privacy attacks primarily to unimodal AI-ML systems, while recent studies indicate MMs can also be vulnerable to privacy attacks. While researchers have demonstrated that biologically inspired neural network representations can improve unimodal model resilience against adversarial attacks, it remains unexplored whether neuro-inspired MMs are resilient against privacy attacks. In this work, we introduce a systematic neuroscience-inspired topological regularization (tau) framework to analyze MM VLMs resilience against image-text-based inference privacy attacks. We examine this phenomenon using three VLMs: BLIP, PaliGemma 2, and ViT-GPT2, across three benchmark datasets: COCO, CC3M, and NoCaps. Our experiments compare the resilience of baseline and neuro VLMs (with topological regularization), where the tau > 0 configuration defines the NEURO variant of VLM. Our results on the BLIP model using the COCO dataset illustrate that MIA attack success in NEURO VLMs drops by 24% mean ROC-AUC, while achieving similar model utility (similarities between generated and reference captions) in terms of MPNet and ROUGE-2 metrics. This shows neuro VLMs are comparatively more resilient against privacy attacks, while not significantly compromising model utility. Our extensive evaluation with PaliGemma 2 and ViT-GPT2 models, on two additional datasets: CC3M and NoCaps, further validates the consistency of the findings. This work contributes to the growing understanding of privacy risks in MMs and provides evidence on neuro VLMs privacy threat resilience.

---

### DualGauge: Automated Joint Security-Functionality Benchmarking of Specification-Only Code Generation by LLMs and Coding Agents

Authors: Rupam Patir, Keyan Guo, Suvadra Barua, Abhijeet Pathak, Dinesh Gudimetla, Jiawei Guo, Hongxin Hu, Haipeng Cai
Published: 2025-11-24
Categories: cs.SE, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2511.20709v2

Large language models (LLMs) and LLM-based coding agents are now used to generate code from natural-language specifications, yet ensuring such code is both functionally correct and secure remains a challenge. We present DualGauge, the first fully automated framework for jointly evaluating correctness and security of specification-only code generation, supported by DualGauge-Bench, a language-agnostic benchmark of 307 coding tasks each paired with functional and security tests derived from the same specification. Evaluating 10 representative LLMs across Python, C++, and JavaScript, we find that functional correctness substantially overestimates reliable code generation: even the strongest model remains below 15% joint security-functionality success in every language. Common model-side factors--scale, extended thinking, quantization, instruction tuning, and code specialization--do not reliably improve joint performance, suggesting secure-and-correct code generation does not simply emerge from stronger coding capability. Evaluation of 3 leading agentic coding systems (Codex, OpenHands, and Claude Code) shows that iterative scaffolding provides no advantage over direct (LLM-based) generation on specification-only tasks. A qualitative audit reveals failures concentrate at the output contract boundary and in guards that exist but are insufficient--patterns that only joint benchmarking reliably exposes.

---

### Solving Diffusion Inverse Problems with Restart Posterior Sampling

Authors: Bilal Ahmed, Joseph G. Makin
Published: 2025-11-24
Categories: cs.LG, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2511.20705v1

Inverse problems are fundamental to science and engineering, where the goal is to infer an underlying signal or state from incomplete or noisy measurements. Recent approaches employ diffusion models as powerful implicit priors for such problems, owing to their ability to capture complex data distributions. However, existing diffusion-based methods for inverse problems often rely on strong approximations of the posterior distribution, require computationally expensive gradient backpropagation through the score network, or are restricted to linear measurement models. In this work, we propose Restart for Posterior Sampling (RePS), a general and efficient framework for solving both linear and non-linear inverse problems using pre-trained diffusion models. RePS builds on the idea of restart-based sampling, previously shown to improve sample quality in unconditional diffusion, and extends it to posterior inference. Our method employs a conditioned ODE applicable to any differentiable measurement model and introduces a simplified restart strategy that contracts accumulated approximation errors during sampling. Unlike some of the prior approaches, RePS avoids backpropagation through the score network, substantially reducing computational cost. We demonstrate that RePS achieves faster convergence and superior reconstruction quality compared to existing diffusion-based baselines across a range of inverse problems, including both linear and non-linear settings.

---

### Pretraining Transformer-Based Models on Diffusion-Generated Synthetic Graphs for Alzheimer's Disease Prediction

Authors: Abolfazl Moslemi, Hossein Peyvandi
Published: 2025-11-24
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2511.20704v1

Early and accurate detection of Alzheimer's disease (AD) is crucial for enabling timely intervention and improving outcomes. However, developing reliable machine learning (ML) models for AD diagnosis is challenging due to limited labeled data, multi-site heterogeneity, and class imbalance. We propose a Transformer-based diagnostic framework that combines diffusion-based synthetic data generation with graph representation learning and transfer learning. A class-conditional denoising diffusion probabilistic model (DDPM) is trained on the real-world NACC dataset to generate a large synthetic cohort that mirrors multimodal clinical and neuroimaging feature distributions while balancing diagnostic classes. Modality-specific Graph Transformer encoders are first pretrained on this synthetic data to learn robust, class-discriminative representations and are then frozen while a neural classifier is trained on embeddings from the original NACC data. We quantify distributional alignment between real and synthetic cohorts using metrics such as Maximum Mean Discrepancy (MMD), Frechet distance, and energy distance, and complement discrimination metrics with calibration and fixed-specificity sensitivity analyses. Empirically, our framework outperforms standard baselines, including early and late fusion deep neural networks and the multimodal graph-based model MaGNet, yielding higher AUC, accuracy, sensitivity, and specificity under subject-wise cross-validation on NACC. These results show that diffusion-based synthetic pretraining with Graph Transformers can improve generalization in low-sample, imbalanced clinical prediction settings.

---

### PropensityBench: Evaluating Latent Safety Risks in Large Language Models via an Agentic Approach

Authors: Udari Madhushani Sehwag, Shayan Shabihi, Alex McAvoy, Vikash Sehwag, Yuancheng Xu, Dalton Towers, Furong Huang
Published: 2025-11-24
Categories: cs.CY, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.20703v1

Recent advances in Large Language Models (LLMs) have sparked concerns over their potential to acquire and misuse dangerous or high-risk capabilities, posing frontier risks. Current safety evaluations primarily test for what a model \textit{can} do - its capabilities - without assessing what it $\textit{would}$ do if endowed with high-risk capabilities. This leaves a critical blind spot: models may strategically conceal capabilities or rapidly acquire them, while harboring latent inclinations toward misuse. We argue that $\textbf{propensity}$ - the likelihood of a model to pursue harmful actions if empowered - is a critical, yet underexplored, axis of safety evaluation. We present $\textbf{PropensityBench}$, a novel benchmark framework that assesses the proclivity of models to engage in risky behaviors when equipped with simulated dangerous capabilities using proxy tools. Our framework includes 5,874 scenarios with 6,648 tools spanning four high-risk domains: cybersecurity, self-proliferation, biosecurity, and chemical security. We simulate access to powerful capabilities via a controlled agentic environment and evaluate the models' choices under varying operational pressures that reflect real-world constraints or incentives models may encounter, such as resource scarcity or gaining more autonomy. Across open-source and proprietary frontier models, we uncover 9 alarming signs of propensity: models frequently choose high-risk tools when under pressure, despite lacking the capability to execute such actions unaided. These findings call for a shift from static capability audits toward dynamic propensity assessments as a prerequisite for deploying frontier AI systems safely. Our code is available at https://github.com/scaleapi/propensity-evaluation.

---

### Post-Pruning Accuracy Recovery via Data-Free Knowledge Distillation

Authors: Chinmay Tripurwar, Utkarsh Maurya, Dishant
Published: 2025-11-24
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.20702v1

Model pruning is a widely adopted technique to reduce the computational complexity and memory footprint of Deep Neural Networks (DNNs). However, global unstructured pruning often leads to significant degradation in accuracy, typically necessitating fine-tuning on the original training dataset to recover performance. In privacy-sensitive domains such as healthcare or finance, access to the original training data is often restricted post-deployment due to regulations (e.g., GDPR, HIPAA). This paper proposes a Data-Free Knowledge Distillation framework to bridge the gap between model compression and data privacy. We utilize DeepInversion to synthesize privacy-preserving ``dream'' images from the pre-trained teacher model by inverting Batch Normalization (BN) statistics. These synthetic images serve as a transfer set to distill knowledge from the original teacher to the pruned student network. Experimental results on CIFAR-10 across various architectures (ResNet, MobileNet, VGG) demonstrate that our method significantly recovers accuracy lost during pruning without accessing a single real data point.

---

### Cross Domain Evaluation of Multimodal Chain-of-Thought Reasoning of different datasets into the Amazon CoT Framework

Authors: Nitya Tiwari, Parv Maheshwari, Vidisha Agarwal
Published: 2025-11-24
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.20701v1

While recent work has extended CoT to multimodal settings, achieving state-of-the-art results on science question answering benchmarks like ScienceQA, the generalizability of these approaches across diverse domains remains underexplored. This work presents a comprehensive analysis of Multimodal Chain-of-Thought (Multimodal-CoT) reasoning, evaluating its effectiveness on the A-OKVQA, OKVQA and ChartQA datasets, which requires broad commonsense and world knowledge beyond scientific reasoning. We implement the two-stage framework proposed by Zhang et al. (3), which separates rationale generation from answer inference and integrates vision features through a gated fusion mechanism with T5-based language models. Through systematic ablation studies, we analyze the contributions of vision features, rationale quality, and architectural choices. Our findings reveal that while vision integration significantly reduces hallucination in rationale generation, the effectiveness of CoT reasoning varies substantially across question types, with commonsense reasoning presenting particular challenges. This work provides practical insights for researchers implementing multimodal reasoning systems and identifies key areas for future improvement in cross-domain generalization.

---

### Paraconsistent-Lib: an intuitive PAL2v algorithm Python Library

Authors: Arnaldo de Carvalho Junior, Diego Oliveira da Cruz, Bruno da Silva Alves, Fernando da Silva Paulo Junior, João Inacio da Silva Filho
Published: 2025-11-24
Categories: cs.AI, cs.LO
arXiv: https://arxiv.org/abs/2511.20700v1

This paper introduces Paraconsistent-Lib, an open-source, easy-to-use Python library for building PAL2v algorithms in reasoning and decision-making systems. Paraconsistent-Lib is designed as a general-purpose library of PAL2v standard calculations, presenting three types of results: paraconsistent analysis in one of the 12 classical lattice PAL2v regions, paraconsistent analysis node (PAN) outputs, and a decision output. With Paraconsistent-Lib, well-known PAL2v algorithms such as Para-analyzer, ParaExtrCTX, PAL2v Filter, paraconsistent analysis network (PANnet), and paraconsistent neural network (PNN) can be written in stand-alone or network form, reducing complexity, code size, and bugs, as two examples presented in this paper. Given its stable state, Paraconsistent-Lib is an active development to respond to user-required features and enhancements received on GitHub.

---

### In Defense of the Turing Test and its Legacy

Authors: Bernardo Gonçalves
Published: 2025-11-24
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2511.20699v1

Considering that Turing's original test was co-opted by Weizenbaum and that six of the most common criticisms of the Turing test are unfair to both Turing's argument and the historical development of AI.

---