# Research Papers — 2025-12 · bundle page 140 of 175

> Full-text markdown of papers #6951–7000 of 8703 published in 2025-12, newest first. AI-readable mirrors of the full 8703-paper 2025-12 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-12/page/140
[← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/139.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-12/page/141.md)

### NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generation

Authors: Daniel Rose, Roxane Axel Jacob, Johannes Kirchmair, Thierry Langer
Published: 2025-12-05
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.05844v3

Transformer-based autoregressive models offer an efficient alternative to diffusion- and flow-matching-based approaches for generating 3D molecules. One challenge remains: standard transformer architectures require a sequential ordering of tokens, which is not inherently defined for the atoms in a molecule. Prior works have addressed this by using canonical atom orderings. However, these approaches are not permutation invariant w.r.t. atoms and bias next-token prediction towards ordering conventions. We overcome this limitation by introducing a novel neighborhood-guided training strategy. Our model, NEAT (Neighborhood-Guided, Efficient, Autoregressive Set Transformer) treats molecular graphs as sets of atoms and learns an order-agnostic distribution over admissible tokens at the graph boundary, thereby ensuring atom-level permutation invariance. NEAT achieves state-of-the-art generation quality on the QM9 and GEOM-Drugs datasets while offering a significant speed advantage over existing baselines.

---

### NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generation

Authors: Daniel Rose, Roxane Axel Jacob, Johannes Kirchmair, Thierry Langer
Published: 2025-12-05
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.05844v1

Autoregressive models are a promising alternative to diffusion-based models for 3D molecular structure generation. However, a key limitation is the assumption of a token order: while text has a natural sequential order, the next token prediction given a molecular graph prefix should be invariant to atom permutations. Previous works sidestepped this mismatch by using canonical orders or focus atoms. We argue that this is unnecessary. We introduce NEAT, a Neighborhood-guided, Efficient, Autoregressive, Set Transformer that treats molecular graphs as sets of atoms and learns the order-agnostic distribution over admissible tokens at the graph boundary with an autoregressive flow model. NEAT approaches state-of-the-art performance in 3D molecular generation with high computational efficiency and atom-level permutation invariance, establishing a practical foundation for scalable molecular design.

---

### Wasserstein Evolution : Evolutionary Optimization as Phase Transition

Authors: Kaichen Ouyang, Mingyang Yu, Zong Ke, Junbo Jacob Lian, Shengwei Fu, Xiaoyang Hao, Shengju Yu, Dayu Hu
Published: 2025-12-05
Categories: cs.NE
arXiv: https://arxiv.org/abs/2512.05837v3

Evolutionary algorithms (EAs) serve as powerful black-box optimizers inspired by biological evolution. However, most existing EAs predominantly focus on heuristic operators such as crossover and mutation, while usually overlooking underlying physical interpretability such as statistical mechanics and thermosdynamics. This theoretical void limits the principled understanding of algorithmic dynamics, hindering the systematic design of evolutionary search beyond ad-hoc heuristics. To bridge this gap, we first point out that evolutionary optimization can be conceptually reframed as a physical phase transition process. Building on this perspective, we establish the theoretical grounds by modeling the optimization dynamics as a Wasserstein gradient flow of free energy. Consequently, a robust and interpretable solver named Wasserstein Evolution (WE) is proposed. WE mathematically frames the trade-off between exploration and exploitation as a competition between potential gradient forces and entropic forces. This formulation guarantees convergence to the Boltzmann distribution, thereby minimizing free energy and maximizing entropy, which promotes highly diverse solutions. Extensive experiments on complex multimodal and physical potential functions demonstrate that WE achieves superior diversity and stability compared to established baselines.

---

### Wasserstein Evolution : Evolutionary Optimization as Phase Transition

Authors: Kaichen Ouyang
Published: 2025-12-05
Categories: cs.NE
arXiv: https://arxiv.org/abs/2512.05837v1

This paper establishes a novel connection between evolutionary computation and statistical physics by formalizing evolutionary optimization as a phase transition process. We introduce Wasserstein Evolution (WE), a principled optimization framework that implements the Wasserstein gradient flow of a free energy functional, mathematically bridging evolutionary dynamics with thermodynamics. WE directly translates the physical competition between potential gradient forces (exploitation) and entropic forces (exploration) into algorithmic dynamics, providing an adaptive, theoretically grounded mechanism for balancing exploration and exploitation. Experiments on challenging benchmark functions demonstrate that WE achieves competitive convergence performance while maintaining dramatically higher population diversity than classical methods (GA, DE, CMA-ES).This superior entropy preservation enables effective navigation of multi-modal landscapes without premature convergence, validating the physical interpretation of optimization as a disorder-to-order transition. Our work provides not only an effective optimization algorithm but also a new paradigm for understanding evolutionary computation through statistical physics.

---

### Testing the Utility of Using Large Language Models to Create Personalized Networks From Therapy Session Transcripts: A Proof of Concept Study

Authors: Clarissa W. Ong, Hiba Arnaout, Kate Sheehan, Estella Fox, Eugen Owtscharow, Iryna Gurevych
Published: 2025-12-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2512.05836v2

Recent advances in psychotherapy have focused on treatment personalization, such as by selecting treatment modules based on individual networks. However, estimating personalized networks typically requires intensive longitudinal data, which is not always feasible to collect. A solution to increase scalability of network-driven treatment personalization is leveraging large language models (LLMs). In this study, we developed an end-to-end pipeline for automatically generating client networks to support case conceptualization and treatment planning. We annotated 8,028 utterances from 77 therapy transcripts (N = 6). In the first stage of the pipeline, we identified clinically relevant processes (binary classification) and their corresponding dimensions (multi-label classification). Then, we introduced a two-step method that grouped the processes into clinically meaningful clusters and generated labels for the clusters. Finally, we generated connections between clusters. Evaluation results generally supported model utility, however, interrater agreement on model performance metrics was inconsistent, ranging from poor to substantial. Qualitative examination of the networks indicated consistency with original study data. Given coherence and interpretability of the generated networks, developing networks from therapy transcripts using LLMs appears to be feasible. Nonetheless, more research is needed to examine whether these networks improve treatment outcomes, including relative to other methods of treatment personalization, such as statistically estimated networks. Potential use cases and limitations of our pipeline are discussed.

---

### Using Large Language Models to Create Personalized Networks From Therapy Sessions

Authors: Clarissa W. Ong, Hiba Arnaout, Kate Sheehan, Estella Fox, Eugen Owtscharow, Iryna Gurevych
Published: 2025-12-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2512.05836v1

Recent advances in psychotherapy have focused on treatment personalization, such as by selecting treatment modules based on personalized networks. However, estimating personalized networks typically requires intensive longitudinal data, which is not always feasible. A solution to facilitate scalability of network-driven treatment personalization is leveraging LLMs. In this study, we present an end-to-end pipeline for automatically generating client networks from 77 therapy transcripts to support case conceptualization and treatment planning. We annotated 3364 psychological processes and their corresponding dimensions in therapy transcripts. Using these data, we applied in-context learning to jointly identify psychological processes and their dimensions. The method achieved high performance even with a few training examples. To organize the processes into networks, we introduced a two-step method that grouped them into clinically meaningful clusters. We then generated explanation-augmented relationships between clusters. Experts found that networks produced by our multi-step approach outperformed those built with direct prompting for clinical utility and interpretability, with up to 90% preferring our approach. In addition, the networks were rated favorably by experts, with scores for clinical relevance, novelty, and usefulness ranging from 72-75%. Our findings provide a proof of concept for using LLMs to create clinically relevant networks from therapy transcripts. Advantages of our approach include bottom-up case conceptualization from client utterances in therapy sessions and identification of latent themes. Networks generated from our pipeline may be used in clinical settings and supervision and training. Future research should examine whether these networks improve treatment outcomes relative to other methods of treatment personalization, including statistically estimated networks.

---

### Vague Knowledge: Information without Transitivity and Partitions

Authors: Kerry Xiao
Published: 2025-12-05
Categories: econ.TH, cs.CL, math.LO, q-fin.GN
arXiv: https://arxiv.org/abs/2512.05833v2

I relax the standard assumptions of transitivity and partition structure in economic models of information to formalize vague knowledge: non-transitive indistinguishability over states. I show that vague knowledge, while failing to partition the state space, remains informative by distinguishing some states from others. Moreover, it can only be faithfully expressed through vague communication with blurred boundaries. My results provide microfoundations for the prevalence of natural language communication and qualitative reasoning in the real world, where knowledge is often vague.

---

### Vague Knowledge: Information without Transitivity and Partitions

Authors: Kerry Xiao
Published: 2025-12-05
Categories: econ.TH, cs.CL, math.LO, q-fin.GN
arXiv: https://arxiv.org/abs/2512.05833v1

I relax the standard assumptions of transitivity and partition structure in economic models of information to formalize vague knowledge: non-transitive indistinguishability over states. I show that vague knowledge, while failing to partition the state space, remains informative by distinguishing some states from others. Moreover, it can only be faithfully expressed through vague communication with blurred boundaries. My results provide microfoundations for the prevalence of natural language communication and qualitative reasoning in the real world, where knowledge is often vague.

---

### Heard or Halted? Gender, Interruptions, and Emotional Tone in U.S. Supreme Court Oral Arguments

Authors: Yifei Tong
Published: 2025-12-05
Categories: cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2512.05832v2

This study examines how interruptions during U.S. Supreme Court oral arguments shape both the semantic content and emotional tone of advocates' speech, with a focus on gendered dynamics in judicial discourse. Using the ConvoKit Supreme Court Corpus (2010-2019), we analyze 12,663 speech chunks from advocate-justice interactions to assess whether interruptions alter the meaning of an advocate's argument and whether interruptions toward female advocates exhibit more negative emotional valence. Semantic shifts are quantified using GloVe-based sentence embeddings, while sentiment is measured through lexicon-based analysis. We find that semantic similarity between pre- and post-interruption speech remains consistently high, suggesting that interruptions do not substantially alter argumentative content. However, interruptions directed at female advocates contain significantly higher levels of negative sentiment. These results deepen empirical understanding of gendered communication in elite institutional settings and demonstrate the value of computational linguistic methods for studying power, discourse, and equity in judicial proceedings.

---

### Heard or Halted? Gender, Interruptions, and Emotional Tone in U.S. Supreme Court Oral Arguments

Authors: Yifei Tong
Published: 2025-12-05
Categories: cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2512.05832v1

This study examines how interruptions during U.S. Supreme Court oral arguments shape both the semantic content and emotional tone of advocates' speech, with a focus on gendered dynamics in judicial discourse. Using the ConvoKit Supreme Court Corpus (2010-2019), we analyze 12,663 speech chunks from advocate-justice interactions to assess whether interruptions alter the meaning of an advocate's argument and whether interruptions toward female advocates exhibit more negative emotional valence. Semantic shifts are quantified using GloVe-based sentence embeddings, while sentiment is measured through lexicon-based analysis. We find that semantic similarity between pre- and post-interruption speech remains consistently high, suggesting that interruptions do not substantially alter argumentative content. However, interruptions directed at female advocates contain significantly higher levels of negative sentiment. These results deepen empirical understanding of gendered communication in elite institutional settings and demonstrate the value of computational linguistic methods for studying power, discourse, and equity in judicial proceedings.

---

### Phase-OTDR Event Detection Using Image-Based Data Transformation and Deep Learning

Authors: Muhammet Cagri Yeke, Samil Sirin, Kivilcim Yuksel, Abdurrahman Gumus
Published: 2025-12-05
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2512.05830v1

This study focuses on event detection in optical fibers, specifically classifying six events using the Phase-OTDR system. A novel approach is introduced to enhance Phase-OTDR data analysis by transforming 1D data into grayscale images through techniques such as Gramian Angular Difference Field, Gramian Angular Summation Field, and Recurrence Plot. These grayscale images are combined into a multi-channel RGB representation, enabling more robust and adaptable analysis using transfer learning models. The proposed methodology achieves high classification accuracies of 98.84% and 98.24% with the EfficientNetB0 and DenseNet121 models, respectively. A 5-fold cross-validation process confirms the reliability of these models, with test accuracy rates of 99.07% and 98.68%. Using a publicly available Phase-OTDR dataset, the study demonstrates an efficient approach to understanding optical fiber events while reducing dataset size and improving analysis efficiency. The results highlight the transformative potential of image-based analysis in interpreting complex fiber optic sensing data, offering significant advancements in the accuracy and reliability of fiber optic monitoring systems. The codes and the corresponding image-based dataset are made publicly available on GitHub to support further research: https://github.com/miralab-ai/Phase-OTDR-event-detection.

---

### Approximation of Box Decomposition Algorithm for Fast Hypervolume-Based Multi-Objective Optimization

Authors: Shuhei Watanabe
Published: 2025-12-05
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.05825v1

Hypervolume (HV)-based Bayesian optimization (BO) is one of the standard approaches for multi-objective decision-making. However, the computational cost of optimizing the acquisition function remains a significant bottleneck, primarily due to the expense of HV improvement calculations. While HV box-decomposition offers an efficient way to cope with the frequent exact improvement calculations, it suffers from super-polynomial memory complexity $O(MN^{\lfloor \frac{M + 1}{2} \rfloor})$ in the worst case as proposed by Lacour et al. (2017). To tackle this problem, Couckuyt et al. (2012) employed an approximation algorithm. However, a rigorous algorithmic description is currently absent from the literature. This paper bridges this gap by providing comprehensive mathematical and algorithmic details of this approximation algorithm.

---

### Multimodal Oncology Agent for IDH1 Mutation Prediction in Low-Grade Glioma

Authors: Hafsa Akebli, Adam Shephard, Vincenzo Della Mea, Nasir Rajpoot
Published: 2025-12-05
Categories: cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2512.05824v1

Low-grade gliomas frequently present IDH1 mutations that define clinically distinct subgroups with specific prognostic and therapeutic implications. This work introduces a Multimodal Oncology Agent (MOA) integrating a histology tool based on the TITAN foundation model for IDH1 mutation prediction in low-grade glioma, combined with reasoning over structured clinical and genomic inputs through PubMed, Google Search, and OncoKB. MOA reports were quantitatively evaluated on 488 patients from the TCGA-LGG cohort against clinical and histology baselines. MOA without the histology tool outperformed the clinical baseline, achieving an F1-score of 0.826 compared to 0.798. When fused with histology features, MOA reached the highest performance with an F1-score of 0.912, exceeding both the histology baseline at 0.894 and the fused histology-clinical baseline at 0.897. These results demonstrate that the proposed agent captures complementary mutation-relevant information enriched through external biomedical sources, enabling accurate IDH1 mutation prediction.

---

### Machine-learning-enabled interpretation of tribological deformation patterns in large-scale MD data

Authors: Hendrik J. Ehrich, Marvin C. May, Stefan J. Eder
Published: 2025-12-05
Categories: cond-mat.mtrl-sci, cs.LG
arXiv: https://arxiv.org/abs/2512.05818v1

Molecular dynamics (MD) simulations have become indispensable for exploring tribological deformation patterns at the atomic scale. However, transforming the resulting high-dimensional data into interpretable deformation pattern maps remains a resource-intensive and largely manual process. In this work, we introduce a data-driven workflow that automates this interpretation step using unsupervised and supervised learning. Grain-orientation-colored computational tomograph pictures obtained from CuNi alloy simulations were first compressed through an autoencoder to a 32-dimensional global feature vector. Despite this strong compression, the reconstructed images retained the essential microstructural motifs: grain boundaries, stacking faults, twins, and partial lattice rotations, while omitting only the finest defects. The learned representations were then combined with simulation metadata (composition, load, time, temperature, and spatial position) to train a CNN-MLP model to predict the dominant deformation pattern. The resulting model achieves a prediction accuracy of approximately 96% on validation data. A refined evaluation strategy, in which an entire spatial region containing distinct grains was excluded from training, provides a more robust measure of generalization. The approach demonstrates that essential tribological deformation signatures can be automatically identified and classified from structural images using Machine Learning. This proof of concept constitutes a first step towards fully automated, data-driven construction of tribological mechanism maps and, ultimately, toward predictive modeling frameworks that may reduce the need for large-scale MD simulation campaigns.

---

### Utility Boundary of Dataset Distillation: Scaling and Configuration-Coverage Laws

Authors: Zhengquan Luo, Zhiqiang Xu
Published: 2025-12-05
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.05817v3

Dataset distillation (DD) aims to construct compact synthetic datasets that allow models to achieve comparable performance to full-data training while substantially reducing storage and computation. Despite rapid empirical progress, its theoretical foundations remain limited: existing methods (gradient, distribution, trajectory matching) are built on heterogeneous surrogate objectives and optimization assumptions, which makes it difficult to analyze their common principles or provide general guarantees. Moreover, it is still unclear under what conditions distilled data can retain the effectiveness of full datasets when the training configuration, such as optimizer, architecture, or augmentation, changes. To answer these questions, we propose a unified theoretical framework, termed configuration--dynamics--error analysis, which reformulates major DD approaches under a common generalization-error perspective and provides two main results: (i) a scaling law that provides a single-configuration upper bound, characterizing how the error decreases as the distilled sample size increases and explaining the commonly observed performance saturation effect; and (ii) a coverage law showing that the required distilled sample size scales linearly with configuration diversity, with provably matching upper and lower bounds. In addition, our unified analysis reveals that various matching methods are interchangeable surrogates, reducing the same generalization error, clarifying why they can all achieve dataset distillation and providing guidance on how surrogate choices affect sample efficiency and robustness. Experiments across diverse methods and configurations empirically confirm the derived laws, advancing a theoretical foundation for DD and enabling theory-driven design of compact, configuration-robust dataset distillation.

---

### Utility Boundary of Dataset Distillation: Scaling and Configuration-Coverage Laws

Authors: Zhengquan Luo, Zhiqiang Xu
Published: 2025-12-05
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.05817v1

Dataset distillation (DD) aims to construct compact synthetic datasets that allow models to achieve comparable performance to full-data training while substantially reducing storage and computation. Despite rapid empirical progress, its theoretical foundations remain limited: existing methods (gradient, distribution, trajectory matching) are built on heterogeneous surrogate objectives and optimization assumptions, which makes it difficult to analyze their common principles or provide general guarantees. Moreover, it is still unclear under what conditions distilled data can retain the effectiveness of full datasets when the training configuration, such as optimizer, architecture, or augmentation, changes. To answer these questions, we propose a unified theoretical framework, termed configuration--dynamics--error analysis, which reformulates major DD approaches under a common generalization-error perspective and provides two main results: (i) a scaling law that provides a single-configuration upper bound, characterizing how the error decreases as the distilled sample size increases and explaining the commonly observed performance saturation effect; and (ii) a coverage law showing that the required distilled sample size scales linearly with configuration diversity, with provably matching upper and lower bounds. In addition, our unified analysis reveals that various matching methods are interchangeable surrogates, reducing the same generalization error, clarifying why they can all achieve dataset distillation and providing guidance on how surrogate choices affect sample efficiency and robustness. Experiments across diverse methods and configurations empirically confirm the derived laws, advancing a theoretical foundation for DD and enabling theory-driven design of compact, configuration-robust dataset distillation.

---

### UG-FedDA: Uncertainty-Guided Federated Domain Adaptation for Multi-Center Alzheimer's Disease Detection

Authors: Fubao Zhu, Zhanyuan Jia, Zhiguo Wang, Huan Huang, Danyang Sun, Chuang Han, Yanting Li, Jiaofen Nan, Chen Zhao, Weihua Zhou
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05814v1

Alzheimer's disease (AD) is an irreversible neurodegenerative disorder, and early diagnosis is critical for timely intervention. However, most existing classification frameworks face challenges in multicenter studies, as they often neglect inter-site heterogeneity and lack mechanisms to quantify uncertainty, which limits their robustness and clinical applicability. To address these issues, we proposed Uncertainty-Guided Federated Domain Adaptation (UG-FedDA), a novel multicenter AD classification framework that integrates uncertainty quantification (UQ) with federated domain adaptation to handle cross-site structure magnetic resonance imaging (MRI) heterogeneity under privacy constraints. Our approach extracts multi-template region-of-interest (RoI) features using a self-attention transformer, capturing both regional representations and their interactions. UQ is integrated to guide feature alignment, mitigating source-target distribution shifts by down-weighting uncertain samples. Experiments are conducted on three public datasets: the Alzheimer's Disease Neuroimaging Initiative (ADNI), the Australian Imaging, Biomarkers and Lifestyle study (AIBL), and the Open Access Series of Imaging Studies (OASIS). UG-FedDA achieved consistent cross-domain improvements in accuracy, sensitivity, and area under the ROC curve across three classification tasks: AD vs. normal controls (NC), mild cognitive impairment (MCI) vs. AD, and NC vs. MCI. For NC vs. AD, UG-FedDA achieves accuracies of 90.54%, 89.04%, and 77.78% on ADNI, AIBL and OASIS datasets, respectively. For MCI vs. AD, accuracies are 80.20% (ADNI), 71.91% (AIBL), and 79.73% (OASIS). For NC vs. MCI, results are 76.87% (ADNI), 73.91% (AIBL), and 83.73% (OASIS). These results demonstrate that the proposed framework not only adapts efficiently across multiple sites but also preserves strict privacy.

---

### Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation

Authors: Fabian Konstantinidis, Moritz Sackmann, Ulrich Hofmann, Christoph Stiller
Published: 2025-12-05
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2512.05812v5

Scalable multi-agent driving simulation requires behavior models that are both realistic and computationally efficient. We address this by optimizing the behavior model that controls individual traffic participants. To improve efficiency, we adopt an instance-centric scene representation, where each traffic participant and map element is modeled in its own local coordinate frame. This design enables efficient, viewpoint-invariant scene encoding and allows static map tokens to be reused across simulation steps. To model interactions, we employ a query-centric symmetric context encoder with relative positional encodings between local frames. We use Adversarial Inverse Reinforcement Learning to learn the behavior model and propose an adaptive reward transformation that automatically balances robustness and realism during training. Experiments demonstrate that our approach scales efficiently with the number of tokens, significantly reducing training and inference times, while outperforming several agent-centric baselines in terms of positional accuracy and robustness.

---

### Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation

Authors: Fabian Konstantinidis, Moritz Sackmann, Ulrich Hofmann, Christoph Stiller
Published: 2025-12-05
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2512.05812v1

Scalable multi-agent driving simulation requires behavior models that are both realistic and computationally efficient. We address this by optimizing the behavior model that controls individual traffic participants. To improve efficiency, we adopt an instance-centric scene representation, where each traffic participant and map element is modeled in its own local coordinate frame. This design enables efficient, viewpoint-invariant scene encoding and allows static map tokens to be reused across simulation steps. To model interactions, we employ a query-centric symmetric context encoder with relative positional encodings between local frames. We use Adversarial Inverse Reinforcement Learning to learn the behavior model and propose an adaptive reward transformation that automatically balances robustness and realism during training. Experiments demonstrate that our approach scales efficiently with the number of tokens, significantly reducing training and inference times, while outperforming several agent-centric baselines in terms of positional accuracy and robustness.

---

### Probing the effectiveness of World Models for Spatial Reasoning through Test-time Scaling

Authors: Saurav Jha, M. Jehanzeb Mirza, Wei Lin, Shiqi Yang, Sarath Chandar
Published: 2025-12-05
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2512.05809v1

Vision-Language Models (VLMs) remain limited in spatial reasoning tasks that require multi-view understanding and embodied perspective shifts. Recent approaches such as MindJourney attempt to mitigate this gap through test-time scaling where a world model imagines action-conditioned trajectories and a heuristic verifier selects helpful views from such trajectories. In this work, we systematically examine how such test-time verifiers behave across benchmarks, uncovering both their promise and their pitfalls. Our uncertainty-based analyses show that MindJourney's verifier provides little meaningful calibration, and that random scoring often reduces answer entropy equally well, thus exposing systematic action biases and unreliable reward signals. To mitigate these, we introduce a Verification through Spatial Assertions (ViSA) framework that grounds the test-time reward in verifiable, frame-anchored micro-claims. This principled verifier consistently improves spatial reasoning on the SAT-Real benchmark and corrects trajectory-selection biases through more balanced exploratory behavior. However, on the challenging MMSI-Bench, none of the verifiers, including ours, achieve consistent scaling, suggesting that the current world models form an information bottleneck where imagined views fail to enrich fine-grained reasoning. Together, these findings chart the bad, good, and ugly aspects of test-time verification for world-model-based reasoning. Our code is available at https://github.com/chandar-lab/visa-for-mindjourney.

---

### 3D Path Planning for Robot-assisted Vertebroplasty from Arbitrary Bi-plane X-ray via Differentiable Rendering

Authors: Blanca Inigo, Benjamin D. Killeen, Rebecca Choi, Michelle Song, Ali Uneri, Majid Khan, Christopher Bailey, Axel Krieger, Mathias Unberath
Published: 2025-12-05
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2512.05803v1

Robotic systems are transforming image-guided interventions by enhancing accuracy and minimizing radiation exposure. A significant challenge in robotic assistance lies in surgical path planning, which often relies on the registration of intraoperative 2D images with preoperative 3D CT scans. This requirement can be burdensome and costly, particularly in procedures like vertebroplasty, where preoperative CT scans are not routinely performed. To address this issue, we introduce a differentiable rendering-based framework for 3D transpedicular path planning utilizing bi-planar 2D X-rays. Our method integrates differentiable rendering with a vertebral atlas generated through a Statistical Shape Model (SSM) and employs a learned similarity loss to refine the SSM shape and pose dynamically, independent of fixed imaging geometries. We evaluated our framework in two stages: first, through vertebral reconstruction from orthogonal X-rays for benchmarking, and second, via clinician-in-the-loop path planning using arbitrary-view X-rays. Our results indicate that our method outperformed a normalized cross-correlation baseline in reconstruction metrics (DICE: 0.75 vs. 0.65) and achieved comparable performance to the state-of-the-art model ReVerteR (DICE: 0.77), while maintaining generalization to arbitrary views. Success rates for bipedicular planning reached 82% with synthetic data and 75% with cadaver data, exceeding the 66% and 31% rates of a 2D-to-3D baseline, respectively. In conclusion, our framework facilitates versatile, CT-free 3D path planning for robot-assisted vertebroplasty, effectively accommodating real-world imaging diversity without the need for preoperative CT scans.

---

### Bring Your Dreams to Life: Continual Text-to-Video Customization

Authors: Jiahua Dong, Xudong Wang, Wenqi Liang, Zongyan Han, Meng Cao, Duzhen Zhang, Hanbin Zhao, Zhi Han, Salman Khan, Fahad Shahbaz Khan
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05802v2

Customized text-to-video generation (CTVG) has recently witnessed great progress in generating tailored videos from user-specific text. However, most CTVG methods assume that personalized concepts remain static and do not expand incrementally over time. Additionally, they struggle with forgetting and concept neglect when continuously learning new concepts, including subjects and motions. To resolve the above challenges, we develop a novel Continual Customized Video Diffusion (CCVD) model, which can continuously learn new concepts to generate videos across various text-to-video generation tasks by tackling forgetting and concept neglect. To address catastrophic forgetting, we introduce a concept-specific attribute retention module and a task-aware concept aggregation strategy. They can capture the unique characteristics and identities of old concepts during training, while combining all subject and motion adapters of old concepts based on their relevance during testing. Besides, to tackle concept neglect, we develop a controllable conditional synthesis to enhance regional features and align video contexts with user conditions, by incorporating layer-specific region attention-guided noise estimation. Extensive experimental comparisons demonstrate that our CCVD outperforms existing CTVG baselines on both the DreamVideo and Wan 2.1 backbones. The code is available at https://github.com/JiahuaDong/CCVD.

---

### Bring Your Dreams to Life: Continual Text-to-Video Customization

Authors: Jiahua Dong, Xudong Wang, Wenqi Liang, Zongyan Han, Meng Cao, Duzhen Zhang, Hanbin Zhao, Zhi Han, Salman Khan, Fahad Shahbaz Khan
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05802v1

Customized text-to-video generation (CTVG) has recently witnessed great progress in generating tailored videos from user-specific text. However, most CTVG methods assume that personalized concepts remain static and do not expand incrementally over time. Additionally, they struggle with forgetting and concept neglect when continuously learning new concepts, including subjects and motions. To resolve the above challenges, we develop a novel Continual Customized Video Diffusion (CCVD) model, which can continuously learn new concepts to generate videos across various text-to-video generation tasks by tackling forgetting and concept neglect. To address catastrophic forgetting, we introduce a concept-specific attribute retention module and a task-aware concept aggregation strategy. They can capture the unique characteristics and identities of old concepts during training, while combining all subject and motion adapters of old concepts based on their relevance during testing. Besides, to tackle concept neglect, we develop a controllable conditional synthesis to enhance regional features and align video contexts with user conditions, by incorporating layer-specific region attention-guided noise estimation. Extensive experimental comparisons demonstrate that our CCVD outperforms existing CTVG models. The code is available at https://github.com/JiahuaDong/CCVD.

---

### Mechanistic Interpretability of Antibody Language Models Using SAEs

Authors: Rebonto Haque, Oliver M. Turnbull, Anisha Parsan, Nithin Parsan, John J. Yang, Anna L. Beukenhorst, Charlotte M. Deane
Published: 2025-12-05
Categories: cs.LG, cs.AI, q-bio.QM
arXiv: https://arxiv.org/abs/2512.05794v3

Sparse autoencoders (SAEs) are a mechanistic interpretability technique that have been used to provide insight into learned concepts within large protein language models. Here, we employ TopK and Ordered SAEs to investigate autoregressive antibody language models, and steer their generation. We show that TopK SAEs can reveal biologically meaningful latent features, but high feature-concept correlation does not guarantee causal control over generation. In contrast, Ordered SAEs impose a hierarchical structure that reliably identifies steerable features, but at the expense of more complex and less interpretable activation patterns. These findings advance the mechanistic interpretability of domain-specific protein language models and suggest that, while TopK SAEs suffice for mapping latent features to concepts, Ordered SAEs are preferable when precise generative steering is required.

---

### Mechanistic Interpretability of Antibody Language Models Using SAEs

Authors: Rebonto Haque, Oliver M. Turnbull, Anisha Parsan, Nithin Parsan, John J. Yang, Charlotte M. Deane
Published: 2025-12-05
Categories: cs.LG, cs.AI, q-bio.QM
arXiv: https://arxiv.org/abs/2512.05794v1

Sparse autoencoders (SAEs) are a mechanistic interpretability technique that have been used to provide insight into learned concepts within large protein language models. Here, we employ TopK and Ordered SAEs to investigate an autoregressive antibody language model, p-IgGen, and steer its generation. We show that TopK SAEs can reveal biologically meaningful latent features, but high feature concept correlation does not guarantee causal control over generation. In contrast, Ordered SAEs impose an hierarchical structure that reliably identifies steerable features, but at the expense of more complex and less interpretable activation patterns. These findings advance the mechanistic interpretability of domain-specific protein language models and suggest that, while TopK SAEs are sufficient for mapping latent features to concepts, Ordered SAEs are preferable when precise generative steering is required.

---

### Fast and Robust Diffusion Posterior Sampling for MR Image Reconstruction Using the Preconditioned Unadjusted Langevin Algorithm

Authors: Moritz Blumenthal, Tina Holliber, Jonathan I. Tamir, Martin Uecker
Published: 2025-12-05
Categories: physics.med-ph, cs.CV, cs.LG, math.PR
arXiv: https://arxiv.org/abs/2512.05791v2

Purpose: The Unadjusted Langevin Algorithm (ULA) in combination with diffusion models can generate high quality MRI reconstructions with uncertainty estimation from highly undersampled k-space data. However, sampling methods such as diffusion posterior sampling (DPS) or likelihood annealing suffer from long reconstruction times and the need for parameter tuning. The purpose of this work is to develop a robust sampling algorithm with fast convergence. Theory and Methods: In the reverse diffusion process used for sampling the posterior, the exact likelihood is multiplied with the diffused prior at all noise scales. To overcome the issue of slow convergence, preconditioning is used. The method is trained on fastMRI data and tested on retrospectively undersampled brain data of a healthy volunteer. Results: For posterior sampling in Cartesian and non-Cartesian accelerated MRI the new approach outperforms annealed sampling and DPS in terms of reconstruction speed and sample quality. Conclusion: The proposed exact likelihood with preconditioning enables rapid and reliable posterior sampling across various MRI reconstruction tasks without the need for parameter tuning.

---

### Learnability Window in Gated Recurrent Neural Networks

Authors: Lorenzo Livi
Published: 2025-12-05
Categories: cs.LG, physics.data-an
arXiv: https://arxiv.org/abs/2512.05790v10

We develop a statistical theory of temporal learnability in recurrent neural networks, quantifying the maximal temporal horizon $\mathcal{H}_N$ over which gradient-based learning can recover lag-dependent structure at finite sample size $N$. The theory is built on the effective learning rate envelope $f(\ell)$, a function that captures how gating mechanisms and adaptive optimizers jointly shape the coupling between state-space dynamics and parameter updates during Backpropagation Through Time. Under heavy-tailed ($α$-stable) fluctuations, where empirical averages concentrate at rate $N^{-1/κ_α}$ with $κ_α= α/(α-1)$, the interplay between envelope decay and statistical concentration yields explicit scaling laws for the growth of $\mathcal{H}_N$: logarithmic, polynomial, and exponential temporal learning regimes emerge according to the decay law of $f(\ell)$. These results identify envelope decay as the key determinant of temporal learnability. Slower attenuation of $f(\ell)$ enlarges $\mathcal{H}_N$, while heavy-tailed fluctuations compress it by weakening statistical concentration. Moreover, envelope geometry outweighs dataset size: slowing the envelope's decay enlarges $\mathcal{H}_N$ more than adding data, so more complex architectures that realize slower-decaying envelopes can be more data-efficient than simpler ones. Experiments across multiple gated architectures and optimizers corroborate these structural predictions.

---

### Learnability Window in Gated Recurrent Neural Networks

Authors: Lorenzo Livi
Published: 2025-12-05
Categories: cs.LG, physics.data-an
arXiv: https://arxiv.org/abs/2512.05790v1

We develop a theoretical framework that explains how gating mechanisms determine the learnability window $\mathcal{H}_N$ of recurrent neural networks, defined as the largest temporal horizon over which gradient information remains statistically recoverable. While classical analyses emphasize numerical stability of Jacobian products, we show that stability alone is insufficient: learnability is governed instead by the \emph{effective learning rates} $μ_{t,\ell}$, per-lag and per-neuron quantities obtained from first-order expansions of gate-induced Jacobian products in Backpropagation Through Time. These effective learning rates act as multiplicative filters that control both the magnitude and anisotropy of gradient transport. Under heavy-tailed ($α$-stable) gradient noise, we prove that the minimal sample size required to detect a dependency at lag~$\ell$ satisfies $N(\ell)\propto f(\ell)^{-α}$, where $f(\ell)=\|μ_{t,\ell}\|_1$ is the effective learning rate envelope. This leads to an explicit formula for $\mathcal{H}_N$ and closed-form scaling laws for logarithmic, polynomial, and exponential decay of $f(\ell)$. The theory predicts that broader or more heterogeneous gate spectra produce slower decay of $f(\ell)$ and hence larger learnability windows, whereas heavier-tailed noise compresses $\mathcal{H}_N$ by slowing statistical concentration. By linking gate-induced time-scale structure, gradient noise, and sample complexity, the framework identifies the effective learning rates as the fundamental quantities that govern when -- and for how long -- gated recurrent networks can learn long-range temporal dependencies.

---

### Curvature-Regularized Variational Autoencoder for 3D Scene Reconstruction from Sparse Depth

Authors: Maryam Yousefi, Soodeh Bakhshandeh
Published: 2025-12-05
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2512.05783v1

When depth sensors provide only 5% of needed measurements, reconstructing complete 3D scenes becomes difficult. Autonomous vehicles and robots cannot tolerate the geometric errors that sparse reconstruction introduces. We propose curvature regularization through a discrete Laplacian operator, achieving 18.1% better reconstruction accuracy than standard variational autoencoders. Our contribution challenges an implicit assumption in geometric deep learning: that combining multiple geometric constraints improves performance. A single well-designed regularization term not only matches but exceeds the effectiveness of complex multi-term formulations. The discrete Laplacian offers stable gradients and noise suppression with just 15% training overhead and zero inference cost. Code and models are available at https://github.com/Maryousefi/GeoVAE-3D.

---

### Active Video Perception: Iterative Evidence Seeking for Agentic Long Video Understanding

Authors: Ziyang Wang, Honglu Zhou, Shijie Wang, Junnan Li, Caiming Xiong, Silvio Savarese, Mohit Bansal, Michael S. Ryoo, Juan Carlos Niebles
Published: 2025-12-05
Categories: cs.CV, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2512.05774v2

Long video understanding (LVU) is challenging because answering real-world queries often depends on sparse, temporally dispersed cues buried in hours of mostly redundant and irrelevant content. While agentic pipelines improve video reasoning capabilities, prevailing frameworks rely on a query-agnostic captioner to perceive video information, which wastes computation on irrelevant content and blurs fine-grained temporal and spatial information. Motivated by active perception theory, we argue that LVU agents should actively decide what, when, and where to observe, and continuously assess whether the current observation is sufficient to answer the query. We present Active Video Perception (AVP), an evidence-seeking framework that treats the video as an interactive environment and acquires compact, queryrelevant evidence directly from pixels. Concretely, AVP runs an iterative plan-observe-reflect process with MLLM agents. In each round, a planner proposes targeted video interactions, an observer executes them to extract time-stamped evidence, and a reflector evaluates the sufficiency of the evidence for the query, either halting with an answer or triggering further observation. Across five LVU benchmarks, AVP achieves highest overall accuracy with significant improvements. Notably, AVP outperforms the best agentic method by 5.7% in average overall accuracy while only requires 18.4% inference time and 12.4% input tokens.

---

### Active Video Perception: Iterative Evidence Seeking for Agentic Long Video Understanding

Authors: Ziyang Wang, Honglu Zhou, Shijie Wang, Junnan Li, Caiming Xiong, Silvio Savarese, Mohit Bansal, Michael S. Ryoo, Juan Carlos Niebles
Published: 2025-12-05
Categories: cs.CV, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2512.05774v1

Long video understanding (LVU) is challenging because answering real-world queries often depends on sparse, temporally dispersed cues buried in hours of mostly redundant and irrelevant content. While agentic pipelines improve video reasoning capabilities, prevailing frameworks rely on a query-agnostic captioner to perceive video information, which wastes computation on irrelevant content and blurs fine-grained temporal and spatial information. Motivated by active perception theory, we argue that LVU agents should actively decide what, when, and where to observe, and continuously assess whether the current observation is sufficient to answer the query. We present Active Video Perception (AVP), an evidence-seeking framework that treats the video as an interactive environment and acquires compact, queryrelevant evidence directly from pixels. Concretely, AVP runs an iterative plan-observe-reflect process with MLLM agents. In each round, a planner proposes targeted video interactions, an observer executes them to extract time-stamped evidence, and a reflector evaluates the sufficiency of the evidence for the query, either halting with an answer or triggering further observation. Across five LVU benchmarks, AVP achieves highest performance with significant improvements. Notably, AVP outperforms the best agentic method by 5.7% in average accuracy while only requires 18.4% inference time and 12.4% input tokens.

---

### AGI Requires a Coordination Layer on Top of Pattern Repositories

Authors: Edward Y. Chang
Published: 2025-12-05
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.05765v2

In this paper we argue that influential critiques dismissing Large Language Models (LLMs) as a dead end for AGI misidentify the bottleneck: they confuse the ocean with the net. Pattern repositories are the necessary System-1 substrate; the missing component is a System-2 coordination layer that recruits relevant patterns, verifies their use, preserves state, and governs convergence. We separate two uses of control that are often conflated. Semantic anchoring, formalized by UCCT (Unified Contextual Control Theory), binds labels and task intent to learned pattern regions through a phase transition governed by effective support (rho_d), representational mismatch (d_r), and an adaptive anchoring budget (gamma log k). Trace-answer verification, implemented by Recursive Causal Audit (RCA), tests whether a final causal judgment is warranted by its own reasoning trace under pressure. We translate these ideas into MACI, a multi-agent coordination stack that integrates diversity and control via baiting (PID-modulated debate), filtering (Socratic and causal audit), and persistence (transactional memory). Empirical validation on causal judgment and the sycophancy-paranoia trade-off demonstrates that static prompting fails where adaptive control succeeds. By reframing common objections as testable coordination failures, we argue that the path to AGI runs through LLMs, not around them. Capability is not coordination.

---

### The Missing Layer of AGI: From Pattern Alchemy to Coordination Physics

Authors: Edward Y. Chang
Published: 2025-12-05
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.05765v1

Influential critiques argue that Large Language Models (LLMs) are a dead end for AGI: "mere pattern matchers" structurally incapable of reasoning or planning. We argue this conclusion misidentifies the bottleneck: it confuses the ocean with the net. Pattern repositories are the necessary System-1 substrate; the missing component is a System-2 coordination layer that selects, constrains, and binds these patterns. We formalize this layer via UCCT, a theory of semantic anchoring that models reasoning as a phase transition governed by effective support (rho_d), representational mismatch (d_r), and an adaptive anchoring budget (gamma log k). Under this lens, ungrounded generation is simply an unbaited retrieval of the substrate's maximum likelihood prior, while "reasoning" emerges when anchors shift the posterior toward goal-directed constraints. We translate UCCT into architecture with MACI, a coordination stack that implements baiting (behavior-modulated debate), filtering (Socratic judging), and persistence (transactional memory). By reframing common objections as testable coordination failures, we argue that the path to AGI runs through LLMs, not around them.

---

### Towards agent-based-model informed neural networks

Authors: Nino Antulov-Fantulin
Published: 2025-12-05
Categories: cs.LG, cs.SI, nlin.AO, physics.soc-ph
arXiv: https://arxiv.org/abs/2512.05764v2

In this article, we present a framework for designing neural networks that remain consistent with the underlying principles of agent-based models. We begin by highlighting the limitations of standard neural differential equations in modeling complex systems, where physical invariants (like energy) are often absent but other constraints (like mass conservation, information locality, bounded rationality) must be enforced. To address this, we introduce Agent-Based-Model informed Neural Networks (ABM-NNs), which leverage restricted graph neural networks and hierarchical decomposition to learn interpretable, structure-preserving dynamics. We validate the framework across three case studies of increasing complexity: (i) a generalized Generalized Lotka--Volterra system, where we recover ground-truth parameters from short trajectories in presence of interventions; (ii) a graph-based SIR contagion model, where our method outperforms state-of-the-art graph learning baselines (GCN, GraphSAGE, Graph Transformer) in out-of-sample forecasting and noise robustness; and (iii) a real-world macroeconomic model of the ten largest economies, where we learn coupled GDP dynamics from empirical data and demonstrate counterfactual analysis for policy interventions

---

### Towards agent-based-model informed neural networks

Authors: Nino Antulov-Fantulin
Published: 2025-12-05
Categories: cs.LG, cs.SI, nlin.AO, physics.soc-ph
arXiv: https://arxiv.org/abs/2512.05764v1

In this article, we present a framework for designing neural networks that remain consistent with the underlying principles of agent-based models. We begin by highlighting the limitations of standard neural differential equations in modeling complex systems, where physical invariants (like energy) are often absent but other constraints (like mass conservation, network locality, bounded rationality) must be enforced. To address this, we introduce Agent-Based-Model informed Neural Networks(ABM-NNs), which leverage restricted graph neural networks and hierarchical decomposition to learn interpretable, structure-preserving dynamics. We validate the framework across three case studies of increasing complexity: (i) a generalized Generalized Lotka--Volterra system, where we recover ground-truth parameters from short trajectories in presence of interventions; (ii) a graph-based SIR contagion model, where our method outperforms state-of-the-art graph learning baselines (GCN, GraphSAGE, Graph Transformer) in out-of-sample forecasting and noise robustness; and (iii) a real-world macroeconomic model of the ten largest economies, where we learn coupled GDP dynamics from empirical data and demonstrate gradient-based counterfactual analysis for policy interventions.

---

### FNOPT: Resolution-Agnostic, Self-Supervised Cloth Simulation using Meta-Optimization with Fourier Neural Operators

Authors: Ruochen Chen, Thuy Tran, Shaifali Parashar
Published: 2025-12-05
Categories: cs.CV, cs.GR
arXiv: https://arxiv.org/abs/2512.05762v1

We present FNOpt, a self-supervised cloth simulation framework that formulates time integration as an optimization problem and trains a resolution-agnostic neural optimizer parameterized by a Fourier neural operator (FNO). Prior neural simulators often rely on extensive ground truth data or sacrifice fine-scale detail, and generalize poorly across resolutions and motion patterns. In contrast, FNOpt learns to simulate physically plausible cloth dynamics and achieves stable and accurate rollouts across diverse mesh resolutions and motion patterns without retraining. Trained only on a coarse grid with physics-based losses, FNOpt generalizes to finer resolutions, capturing fine-scale wrinkles and preserving rollout stability. Extensive evaluations on a benchmark cloth simulation dataset demonstrate that FNOpt outperforms prior learning-based approaches in out-of-distribution settings in both accuracy and robustness. These results position FNO-based meta-optimization as a compelling alternative to previous neural simulators for cloth, thus reducing the need for curated data and improving cross-resolution reliability.

---

### Evolutionary System 2 Reasoning: An Empirical Proof

Authors: Zeyuan Ma, Wenqi Huang, Guo-Huan Song, Hongshu Guo, Sijie Ma, Zhiguang Cao, Yue-Jiao Gong
Published: 2025-12-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2512.05760v1

Machine intelligence marks the ultimate dream of making machines' intelligence comparable to human beings. While recent progress in Large Language Models (LLMs) show substantial specific skills for a wide array of downstream tasks, they more or less fall shorts in general intelligence. Following correlation between intelligence and system 2 reasoning (slow thinking), in this paper, we aim to answering a worthwhile research question: could machine intelligence such as LLMs be evolved to acquire reasoning ability (not specific skill) just like our human beings? To this end, we propose evolutionary reasoning optimization (ERO) framework which performs survival of the fittest over a population of LLMs to search for individual with strong reasoning ability. Given a reasoning task, ERO first initializes multiple LLMs as a population, after which an evolutionary strategy evolves the population to maximize quantified reasoning score of the best individual. Based on experiments on representative testsuites, we claim two surprising empirical discoveries: i) the latest LLMs such as GPT-5 still show limited system 2 reasoning ability; ii) with simple evolution-loop of ERO, a relatively weak model (Qwen-7B) could be enhanced to emerge powerful reasoning ability. Our project can be accessed at https://github.com/MetaEvo/ERO for reproduction needs.

---

### Label-Efficient Point Cloud Segmentation with Active Learning

Authors: Johannes Meyer, Jasper Hoffmann, Felix Schulz, Dominik Merkle, Daniel Buescher, Alexander Reiterer, Joschka Boedecker, Wolfram Burgard
Published: 2025-12-05
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2512.05759v1

Semantic segmentation of 3D point cloud data often comes with high annotation costs. Active learning automates the process of selecting which data to annotate, reducing the total amount of annotation needed to achieve satisfactory performance. Recent approaches to active learning for 3D point clouds are often based on sophisticated heuristics for both, splitting point clouds into annotatable regions and selecting the most beneficial for further neural network training. In this work, we propose a novel and easy-to-implement strategy to separate the point cloud into annotatable regions. In our approach, we utilize a 2D grid to subdivide the point cloud into columns. To identify the next data to be annotated, we employ a network ensemble to estimate the uncertainty in the network output. We evaluate our method on the S3DIS dataset, the Toronto-3D dataset, and a large-scale urban 3D point cloud of the city of Freiburg, which we labeled in parts manually. The extensive evaluation shows that our method yields performance on par with, or even better than, complex state-of-the-art methods on all datasets. Furthermore, we provide results suggesting that in the context of point clouds the annotated area can be a more meaningful measure for active learning algorithms than the number of annotated points.

---

### USV: Unified Sparsification for Accelerating Video Diffusion Models

Authors: Xinjian Wu, Hongmei Wang, Yuan Zhou, Qinglin Lu
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05754v1

The scalability of high-fidelity video diffusion models (VDMs) is constrained by two key sources of redundancy: the quadratic complexity of global spatio-temporal attention and the computational overhead of long iterative denoising trajectories. Existing accelerators -- such as sparse attention and step-distilled samplers -- typically target a single dimension in isolation and quickly encounter diminishing returns, as the remaining bottlenecks become dominant. In this work, we introduce USV (Unified Sparsification for Video diffusion models), an end-to-end trainable framework that overcomes this limitation by jointly orchestrating sparsification across both the model's internal computation and its sampling process. USV learns a dynamic, data- and timestep-dependent sparsification policy that prunes redundant attention connections, adaptively merges semantically similar tokens, and reduces denoising steps, treating them not as independent tricks but as coordinated actions within a single optimization objective. This multi-dimensional co-design enables strong mutual reinforcement among previously disjoint acceleration strategies. Extensive experiments on large-scale video generation benchmarks demonstrate that USV achieves up to 83.3% speedup in the denoising process and 22.7% end-to-end acceleration, while maintaining high visual fidelity. Our results highlight unified, dynamic sparsification as a practical path toward efficient, high-quality video generation.

---

### A Fast Anti-Jamming Cognitive Radar Deployment Algorithm Based on Reinforcement Learning

Authors: Wencheng Cai, Xuchao Gao, Congying Han, Mingqiang Li, Tiande Guo
Published: 2025-12-05
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.05753v2

The fast deployment of cognitive radar to counter jamming remains a critical challenge in modern warfare, where more efficient deployment leads to quicker detection of targets. Existing methods are primarily based on evolutionary algorithms, which are time-consuming and prone to falling into local optima. We tackle these drawbacks via the efficient inference of neural networks and propose a brand new framework: Fast Anti-Jamming Radar Deployment Algorithm (FARDA). We first model the radar deployment problem as an end-to-end task and design deep reinforcement learning algorithms to solve it, where we develop integrated neural modules to perceive heatmap information and a brand new reward format. Empirical results demonstrate that our method achieves coverage comparable to evolutionary algorithms while deploying radars approximately 7,000 times faster. Further ablation experiments confirm the necessity of each component of FARDA.

---

### A Fast Anti-Jamming Cognitive Radar Deployment Algorithm Based on Reinforcement Learning

Authors: Wencheng Cai, Xuchao Gao, Congying Han, Mingqiang Li, Tiande Guo
Published: 2025-12-05
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.05753v1

The fast deployment of cognitive radar to counter jamming remains a critical challenge in modern warfare, where more efficient deployment leads to quicker detection of targets. Existing methods are primarily based on evolutionary algorithms, which are time-consuming and prone to falling into local optima. We tackle these drawbacks via the efficient inference of neural networks and propose a brand new framework: Fast Anti-Jamming Radar Deployment Algorithm (FARDA). We first model the radar deployment problem as an end-to-end task and design deep reinforcement learning algorithms to solve it, where we develop integrated neural modules to perceive heatmap information and a brand new reward format. Empirical results demonstrate that our method achieves coverage comparable to evolutionary algorithms while deploying radars approximately 7,000 times faster. Further ablation experiments confirm the necessity of each component of FARDA.

---

### Exoplanet formation inference using conditional invertible neural networks

Authors: Remo Burn, Victor F. Ksoll, Hubert Klahr, Thomas Henning
Published: 2025-12-05
Categories: astro-ph.EP, cs.NE, physics.data-an
arXiv: https://arxiv.org/abs/2512.05751v1

The interpretation of the origin of observed exoplanets is usually done only qualitatively due to uncertainties of key parameters in planet formation models. To allow a quantitative methodology which traces back in time to the planet birth locations, we train recently developed conditional invertible neural networks (cINN) on synthetic data from a global planet formation model which tracks growth from dust grains to evolved final giant planets. In addition to deterministic single planet formation runs, we also include gravitationally interacting planets in multiplanetary systems, which include some measure of chaos. For the latter case, we treat them as individual planets or choose the two or three planets most likely to be discovered by telescopes. We find that training on multiplanetary data, each planet treated as individual point, is promising. The single-planet data only covers a small range of planets and does not extrapolate well to planet properties not included in the training data. Extension to planetary systems will require more training data due to the higher dimensionality of the problem.

---

### Capturing Classic Authorial Style in Long-Form Story Generation with GRPO Fine-Tuning

Authors: Jinlong Liu, Mohammed Bahja, Venelin Kovatchev, Mark Lee
Published: 2025-12-05
Categories: cs.CL
arXiv: https://arxiv.org/abs/2512.05747v3

Evaluating and optimising authorial style in long-form story generation remains challenging because style is often assessed with ad hoc prompting and is frequently conflated with overall writing quality. We propose a two-stage pipeline. First, we train a dedicated style-similarity judge by fine-tuning a sentence-transformer with authorship-verification supervision, and calibrate its similarity outputs into a bounded $(0,1)$ reward. Second, we use this judge as the primary reward in Group Relative Policy Optimization (GRPO) to fine-tune an 8B story generator for style-conditioned writing, avoiding the accept/reject supervision required by Direct Preference Optimization (DPO). Across four target authors (Mark Twain, Jane Austen, Charles Dickens, Thomas Hardy), the GRPO-trained 8B model achieves higher style scores than open-weight baselines, with an average style score of 0.893 across authors. These results suggest that AV-calibrated reward modelling provides a practical mechanism for controllable style transfer in long-form generation under a moderate model size and training budget.

---

### Capturing Classic Authorial Style in Long-Form Story Generation with GRPO Fine-Tuning

Authors: Jinlong Liu, Mohammed Bahja, Venelin Kovatchev, Mark Lee
Published: 2025-12-05
Categories: cs.CL
arXiv: https://arxiv.org/abs/2512.05747v1

Recent advances in large language models (LLMs) show impressive performance in open-ended story generation, but fine-grained stylistic control remains limited. Existing methods often rely on shallow cues (e.g., names or topics) to simulate authorial style, without robust evaluation. In this work, we present a training framework for style-conditioned story generation using Group Relative Policy Optimization (GRPO) and a custom multi-reward setup. The style reward is derived from a fine-tuned sentence transformer using authorship verification (AV) signals, combined with content and completeness scores to stabilize long-form narrative generation. We conduct experiments using fiction by Mark Twain, a prominent 19th-century American author, with The Adventures of Huckleberry Finn serving as the reference style exemplar. Our 8B model outperforms larger baselines such as GPT-4o and Claude Sonnet 4 in AV-style metrics, achieving a style score of 0.628 and competitive content quality. Results demonstrate the feasibility of agentic stylistic generation with moderate model size and task-specific training. While the output is clearly style-aligned, narrative completeness remains a challenge, indicating future work is needed to better model global coherence and story resolution.

---

### HQ-DM: Single Hadamard Transformation-Based Quantization-Aware Training for Low-Bit Diffusion Models

Authors: Shizhuo Mao, Hongtao Zou, Qihu Xie, Song Chen, Yi Kang
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05746v3

Diffusion models have demonstrated significant applications in the field of image generation. However, their high computational and memory costs pose challenges for deployment. Model quantization has emerged as a promising solution to reduce storage overhead and accelerate inference. Nevertheless, existing quantization methods for diffusion models struggle to mitigate outliers in activation matrices during inference, leading to substantial performance degradation under low-bit quantization scenarios. To address this, we propose HQ-DM, a novel Quantization-Aware Training framework that applies Single Hadamard Transformation to activation matrices. This approach effectively reduces activation outliers while preserving model performance under quantization. Compared to traditional Double Hadamard Transformation, our proposed scheme offers distinct advantages by seamlessly supporting INT convolution operations while preventing the amplification of weight outliers. For conditional generation on the ImageNet 256x256 dataset using the LDM-4 model, our W4A4 and W4A3 quantization schemes improve the Inception Score by 12.8% and 467.73%, respectively, over the existing state-of-the-art method.

---

### HQ-DM: Single Hadamard Transformation-Based Quantization-Aware Training for Low-Bit Diffusion Models

Authors: Shizhuo Mao, Hongtao Zou, Qihu Xie, Song Chen, Yi Kang
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05746v1

Diffusion models have demonstrated significant applications in the field of image generation. However, their high computational and memory costs pose challenges for deployment. Model quantization has emerged as a promising solution to reduce storage overhead and accelerate inference. Nevertheless, existing quantization methods for diffusion models struggle to mitigate outliers in activation matrices during inference, leading to substantial performance degradation under low-bit quantization scenarios. To address this, we propose HQ-DM, a novel Quantization-Aware Training framework that applies Single Hadamard Transformation to activation matrices. This approach effectively reduces activation outliers while preserving model performance under quantization. Compared to traditional Double Hadamard Transformation, our proposed scheme offers distinct advantages by seamlessly supporting INT convolution operations while preventing the amplification of weight outliers. For conditional generation on the ImageNet 256x256 dataset using the LDM-4 model, our W4A4 and W4A3 quantization schemes improve the Inception Score by 12.8% and 467.73%, respectively, over the existing state-of-the-art method.

---

### Distilling Expert Surgical Knowledge: How to train local surgical VLMs for anatomy explanation in Complete Mesocolic Excision

Authors: Lennart Maack, Julia-Kristin Graß, Lisa-Marie Toscha, Nathaniel Melling, Alexander Schlaefer
Published: 2025-12-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.05740v1

Recently, Vision Large Language Models (VLMs) have demonstrated high potential in computer-aided diagnosis and decision-support. However, current VLMs show deficits in domain specific surgical scene understanding, such as identifying and explaining anatomical landmarks during Complete Mesocolic Excision. Additionally, there is a need for locally deployable models to avoid patient data leakage to large VLMs, hosted outside the clinic. We propose a privacy-preserving framework to distill knowledge from large, general-purpose LLMs into an efficient, local VLM. We generate an expert-supervised dataset by prompting a teacher LLM without sensitive images, using only textual context and binary segmentation masks for spatial information. This dataset is used for Supervised Fine-Tuning (SFT) and subsequent Direct Preference Optimization (DPO) of the locally deployable VLM. Our evaluation confirms that finetuning VLMs with our generated datasets increases surgical domain knowledge compared to its base VLM by a large margin. Overall, this work validates a data-efficient and privacy-conforming way to train a surgical domain optimized, locally deployable VLM for surgical scene understanding.

---

### KANFormer for Predicting Fill Probabilities via Survival Analysis in Limit Order Books

Authors: Jinfeng Zhong, Emmanuel Bacry, Agathe Guilloux, Jean-François Muzy
Published: 2025-12-05
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.05734v1

This paper introduces KANFormer, a novel deep-learning-based model for predicting the time-to-fill of limit orders by leveraging both market- and agent-level information. KANFormer combines a Dilated Causal Convolutional network with a Transformer encoder, enhanced by Kolmogorov-Arnold Networks (KANs), which improve nonlinear approximation. Unlike existing models that rely solely on a series of snapshots of the limit order book, KANFormer integrates the actions of agents related to LOB dynamics and the position of the order in the queue to more effectively capture patterns related to execution likelihood. We evaluate the model using CAC 40 index futures data with labeled orders. The results show that KANFormer outperforms existing works in both calibration (Right-Censored Log-Likelihood, Integrated Brier Score) and discrimination (C-index, time-dependent AUC). We further analyze feature importance over time using SHAP (SHapley Additive exPlanations). Our results highlight the benefits of combining rich market signals with expressive neural architectures to achieve accurate and interpretabl predictions of fill probabilities.

---

### Efficient Text Classification with Conformal In-Context Learning

Authors: Ippokratis Pantelidis, Korbinian Randl, Aron Henriksson
Published: 2025-12-05
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2512.05732v1

Large Language Models (LLMs) demonstrate strong in-context learning abilities, yet their effectiveness in text classification depends heavily on prompt design and incurs substantial computational cost. Conformal In-Context Learning (CICLe) has been proposed as a resource-efficient framework that integrates a lightweight base classifier with Conformal Prediction to guide LLM prompting by adaptively reducing the set of candidate classes. However, its broader applicability and efficiency benefits beyond a single domain have not yet been systematically explored. In this paper, we present a comprehensive evaluation of CICLe across diverse NLP classification benchmarks. The results show that CICLe consistently improves over its base classifier and outperforms few-shot prompting baselines when the sample size is sufficient for training the base classifier, and performs comparably in low-data regimes. In terms of efficiency, CICLe reduces the number of shots and prompt length by up to 34.45% and 25.16%, respectively, and enables the use of smaller models with competitive performance. CICLe is furthermore particularly advantageous for text classification tasks with high class imbalance. These findings highlight CICLe as a practical and scalable approach for efficient text classification, combining the robustness of traditional classifiers with the adaptability of LLMs, and achieving substantial gains in data and computational efficiency.

---

### Teaching Language Models Mechanistic Explainability Through MechSMILES

Authors: Théo A. Neukomm, Zlatko Jončev, Philippe Schwaller
Published: 2025-12-05
Categories: cs.LG, physics.chem-ph
arXiv: https://arxiv.org/abs/2512.05722v2

Chemical reaction mechanisms are the foundation of how chemists evaluate reactivity and feasibility, yet current Computer-Assisted Synthesis Planning (CASP) systems operate without this mechanistic reasoning. We introduce a computational framework that teaches language models to predict reaction mechanisms through arrow-pushing formalism, a century-old notation that tracks electron flow while enforcing conservation of mass and charge. This mechanistic understanding enables three capabilities that are difficult or impossible with current methods: post-hoc validation of CASP proposals by reconstructing physically plausible electron pathways, holistic atom-to-atom mapping that tracks all atoms including hydrogens, and extraction of catalyst-aware reaction templates that distinguish recycled catalysts from spectator species. Central to our approach is MechSMILES, a compact textual format encoding molecular structure and electron flow through three arrow types, designed within a Python-based environment that enforces conservation laws and eliminates the possibility of atom hallucination. We trained and benchmarked models on four mechanism prediction tasks of increasing complexity using the main mechanistic datasets in the literature. On our most challenging task, predicting complete mechanisms given only reactants, conditions, and the desired product, our models achieve 93.2\% and 73.3\% pathway retrieval on the FlowER and mech-USPTO-31k datasets respectively, with top-3 retrieval reaching 97.6\% and 86.5\%. Furthermore, the framework rapidly learns new reaction classes, with strong mechanistic predictions for ozonolysis and Suzuki cross-coupling emerging from as few as 40 training examples each. By grounding predictions in physically meaningful electron movements, this work provides an architecture-agnostic, open-source foundation for more explainable and chemically valid CASP.

---