# Research Papers — 2025-11 · bundle page 75 of 197

> Full-text markdown of papers #3701–3750 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/75
[← Previous bundle](https://aipod.dev/papers/archive/2025-11/page/74.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-11/page/76.md)

### Model-to-Model Knowledge Transmission (M2KT): A Data-Free Framework for Cross-Model Understanding Transfer

Authors: Pratham Sorte
Published: 2025-11-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.17638v1

Modern artificial intelligence systems depend heavily on large datasets for both training and transferring knowledge between models. Knowledge distillation, transfer learning, and dataset distillation have made such transfers more efficient, yet they remain fundamentally data-driven: a teacher must produce examples, logits, or gradients for a student to learn. In this work, we introduce Model-to-Model Knowledge Transmission (M2KT), a novel paradigm for data-free conceptual transfer between neural networks. M2KT enables models to exchange knowledge packets that encapsulate structured concept embeddings, abstraction graphs, reasoning traces, and provenance metadata. Unlike classical distillation, M2KT operates primarily in concept space rather than example space, and it does not require labeled datasets or teacher-generated outputs during transfer. We formalize the notion of concept manifolds, introduce an inter-model alignment mapping between teacher and student latent spaces, and derive a composite loss that enforces geometric, structural, and reasoning consistency together with explicit safety constraints. We further present algorithmic procedures for teacher-side packet generation and student-side ingestion and verification. Experiments on symbolic reasoning with large language models show that M2KT can achieve approximately 85 to 90 percent of teacher performance while reducing data usage by over 98 percent compared to standard knowledge distillation. This work establishes a theoretical and practical foundation for data-free AI-to-AI knowledge transfer and self-improving model ecosystems.

---

### PocketLLM: Ultimate Compression of Large Language Models via Meta Networks

Authors: Ye Tian, Chengcheng Wang, Jing Han, Yehui Tang, Kai Han
Published: 2025-11-19
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2511.17637v1

As Large Language Models (LLMs) continue to grow in size, storing and transmitting them on edge devices becomes increasingly challenging. Traditional methods like quantization and pruning struggle to achieve extreme compression of LLMs without sacrificing accuracy. In this paper, we introduce PocketLLM, a novel approach to compress LLMs in a latent space via meta-networks. A simple encoder network is proposed to project the weights of LLMs into discrete latent vectors, which are then represented using a compact codebook. A lightweight decoder network is employed to map the codebook's representative vectors back to the original weight space. This method allows for significant compression of the large weights in LLMs, consisting solely of a small decoder, a concise codebook, and an index. Extensive experiments show that PocketLLM achieves superior performance even at significantly high compression ratios, e.g., compressing Llama 2-7B by 10x with a negligible drop in accuracy.

---

### TSRE: Channel-Aware Typical Set Refinement for Out-of-Distribution Detection

Authors: Weijun Gao, Rundong He, Jinyang Dong, Yongshun Gong
Published: 2025-11-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.17636v1

Out-of-Distribution (OOD) detection is a critical capability for ensuring the safe deployment of machine learning models in open-world environments, where unexpected or anomalous inputs can compromise model reliability and performance. Activation-based methods play a fundamental role in OOD detection by mitigating anomalous activations and enhancing the separation between in-distribution (ID) and OOD data. However, existing methods apply activation rectification while often overlooking channel's intrinsic characteristics and distributional skewness, which results in inaccurate typical set estimation. This discrepancy can lead to the improper inclusion of anomalous activations across channels. To address this limitation, we propose a typical set refinement method based on discriminability and activity, which rectifies activations into a channel-aware typical set. Furthermore, we introduce a skewness-based refinement to mitigate distributional bias in typical set estimation. Finally, we leverage the rectified activations to compute the energy score for OOD detection. Experiments on the ImageNet-1K and CIFAR-100 benchmarks demonstrate that our method achieves state-of-the-art performance and generalizes effectively across backbones and score functions.

---

### Upstream Probabilistic Meta-Imputation for Multimodal Pediatric Pancreatitis Classification

Authors: Max A. Nelson, Elif Keles, Eminenur Sen Tasci, Merve Yazol, Halil Ertugrul Aktas, Ziliang Hong, Andrea Mia Bejar, Gorkem Durak, Oznur Leman Boyunaga, Ulas Bagci
Published: 2025-11-19
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2511.17635v1

Pediatric pancreatitis is a progressive and debilitating inflammatory condition, including acute pancreatitis and chronic pancreatitis, that presents significant clinical diagnostic challenges. Machine learning-based methods also face diagnostic challenges due to limited sample availability and multimodal imaging complexity. To address these challenges, this paper introduces Upstream Probabilistic Meta-Imputation (UPMI), a light-weight augmentation strategy that operates upstream of a meta-learner in a low-dimensional meta-feature space rather than in image space. Modality-specific logistic regressions (T1W and T2W MRI radiomics) produce probability outputs that are transformed into a 7-dimensional meta-feature vector. Class-conditional Gaussian mixture models (GMMs) are then fit within each cross-validation fold to sample synthetic meta-features that, combined with real meta-features, train a Random Forest (RF) meta-classifier. On 67 pediatric subjects with paired T1W/T2W MRIs, UPMI achieves a mean AUC of 0.908 $\pm$ 0.072, a $\sim$5% relative gain over a real-only baseline (AUC 0.864 $\pm$ 0.061).

---

### Efficient Score Pre-computation for Diffusion Models via Cross-Matrix Krylov Projection

Authors: Kaikwan Lau, Andrew S. Na, Justin W. L. Wan
Published: 2025-11-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.17634v1

This paper presents a novel framework to accelerate score-based diffusion models. It first converts the standard stable diffusion model into the Fokker-Planck formulation which results in solving large linear systems for each image. For training involving many images, it can lead to a high computational cost. The core innovation is a cross-matrix Krylov projection method that exploits mathematical similarities between matrices, using a shared subspace built from ``seed" matrices to rapidly solve for subsequent ``target" matrices. Our experiments show that this technique achieves a 15.8\% to 43.7\% time reduction over standard sparse solvers. Additionally, we compare our method against DDPM baselines in denoising tasks, showing a speedup of up to 115$\times$. Furthermore, under a fixed computational budget, our model is able to produce high-quality images while DDPM fails to generate recognizable content, illustrating our approach is a practical method for efficient generation in resource-limited settings.

---

### BD-Net: Has Depth-Wise Convolution Ever Been Applied in Binary Neural Networks?

Authors: DoYoung Kim, Jin-Seop Lee, Noo-ri Kim, SungJoon Lee, Jee-Hyong Lee
Published: 2025-11-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.17633v1

Recent advances in model compression have highlighted the potential of low-bit precision techniques, with Binary Neural Networks (BNNs) attracting attention for their extreme efficiency. However, extreme quantization in BNNs limits representational capacity and destabilizes training, posing significant challenges for lightweight architectures with depth-wise convolutions. To address this, we propose a 1.58-bit convolution to enhance expressiveness and a pre-BN residual connection to stabilize optimization by improving the Hessian condition number. These innovations enable, to the best of our knowledge, the first successful binarization of depth-wise convolutions in BNNs. Our method achieves 33M OPs on ImageNet with MobileNet V1, establishing a new state-of-the-art in BNNs by outperforming prior methods with comparable OPs. Moreover, it consistently outperforms existing methods across various datasets, including CIFAR-10, CIFAR-100, STL-10, Tiny ImageNet, and Oxford Flowers 102, with accuracy improvements of up to 9.3 percentage points.

---

### Smart Manufacturing: MLOps-Enabled Event-Driven Architecture for Enhanced Control in Steel Production

Authors: Bestoun S. Ahmed, Tommaso Azzalin, Andreas Kassler, Andreas Thore, Hans Lindback
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.17632v1

We explore a Digital Twin-Based Approach for Smart Manufacturing to improve Sustainability, Efficiency, and Cost-Effectiveness for a steel production plant. Our system is based on a micro-service edge-compute platform that ingests real-time sensor data from the process into a digital twin over a converged network infrastructure. We implement agile machine learning-based control loops in the digital twin to optimize induction furnace heating, enhance operational quality, and reduce process waste. Key to our approach is a Deep Reinforcement learning-based agent used in our machine learning operation (MLOps) driven system to autonomously correlate the system state with its digital twin to identify correction actions that aim to optimize power settings for the plant. We present the theoretical basis, architectural details, and practical implications of our approach to reduce manufacturing waste and increase production quality. We design the system for flexibility so that our scalable event-driven architecture can be adapted to various industrial applications. With this research, we propose a pivotal step towards the transformation of traditional processes into intelligent systems, aligning with sustainability goals and emphasizing the role of MLOps in shaping the future of data-driven manufacturing.

---

### Enhanced Federated Deep Multi-View Clustering under Uncertainty Scenario

Authors: Bingjun Wei, Xuemei Cao, Jiafen Liu, Haoyang Liang, Xin Yang
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.17631v1

Traditional Federated Multi-View Clustering assumes uniform views across clients, yet practical deployments reveal heterogeneous view completeness with prevalent incomplete, redundant, or corrupted data. While recent approaches model view heterogeneity, they neglect semantic conflicts from dynamic view combinations, failing to address dual uncertainties: view uncertainty (semantic inconsistency from arbitrary view pairings) and aggregation uncertainty (divergent client updates with imbalanced contributions). To address these, we propose a novel Enhanced Federated Deep Multi-View Clustering framework: first align local semantics, hierarchical contrastive fusion within clients resolves view uncertainty by eliminating semantic conflicts; a view adaptive drift module mitigates aggregation uncertainty through global-local prototype contrast that dynamically corrects parameter deviations; and a balanced aggregation mechanism coordinates client updates. Experimental results demonstrate that EFDMVC achieves superior robustness against heterogeneous uncertain views across multiple benchmark datasets, consistently outperforming all state-of-the-art baselines in comprehensive evaluations.

---

### Can we use LLMs to bootstrap reinforcement learning? -- A case study in digital health behavior change

Authors: Nele Albers, Esra Cemre Su de Groot, Loes Keijsers, Manon H. Hillegers, Emiel Krahmer
Published: 2025-11-19
Categories: cs.LG, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2511.17630v1

Personalizing digital applications for health behavior change is a promising route to making them more engaging and effective. This especially holds for approaches that adapt to users and their specific states (e.g., motivation, knowledge, wants) over time. However, developing such approaches requires making many design choices, whose effectiveness is difficult to predict from literature and costly to evaluate in practice. In this work, we explore whether large language models (LLMs) can be used out-of-the-box to generate samples of user interactions that provide useful information for training reinforcement learning models for digital behavior change settings. Using real user data from four large behavior change studies as comparison, we show that LLM-generated samples can be useful in the absence of real data. Comparisons to the samples provided by human raters further show that LLM-generated samples reach the performance of human raters. Additional analyses of different prompting strategies including shorter and longer prompt variants, chain-of-thought prompting, and few-shot prompting show that the relative effectiveness of different strategies depends on both the study and the LLM with also relatively large differences between prompt paraphrases alone. We provide recommendations for how LLM-generated samples can be useful in practice.

---

### Boundary-Aware Adversarial Filtering for Reliable Diagnosis under Extreme Class Imbalance

Authors: Yanxuan Yu, Michael S. Hughes, Julien Lee, Jiacheng Zhou, Andrew F. Laine
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.17629v2

We study classification under extreme class imbalance where recall and calibration are both critical, for example in medical diagnosis scenarios. We propose AF-SMOTE, a mathematically motivated augmentation framework that first synthesizes minority points and then filters them by an adversarial discriminator and a boundary utility model. We prove that, under mild assumptions on the decision boundary smoothness and class-conditional densities, our filtering step monotonically improves a surrogate of F_beta (for beta >= 1) while not inflating Brier score. On MIMIC-IV proxy label prediction and canonical fraud detection benchmarks, AF-SMOTE attains higher recall and average precision than strong oversampling baselines (SMOTE, ADASYN, Borderline-SMOTE, SVM-SMOTE), and yields the best calibration. We further validate these gains across multiple additional datasets beyond MIMIC-IV. Our successful application of AF-SMOTE to a healthcare dataset using a proxy label demonstrates in a disease-agnostic way its practical value in clinical situations, where missing true positive cases in rare diseases can have severe consequences.

---

### Rectifying Distribution Shift in Cascaded Precipitation Nowcasting

Authors: Fanbo Ju, Haiyuan Shi, Qingjian Ni
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.17628v3

Precipitation nowcasting, which aims to provide high spatio-temporal resolution precipitation forecasts by leveraging current radar observations, is a core task in regional weather forecasting. Recently, the cascaded architecture has emerged as the mainstream paradigm for deep learning-based precipitation nowcasting. This paradigm involves a deterministic model to predict posterior mean, followed by a probabilistic model to generate local stochasticity. However, existing methods commonly overlook the conflation of the systematic distribution shift in deterministic predictions and the local stochasticity. As a result, the distribution shift of the deterministic component contaminates the predictions of the probabilistic component, leading to inaccuracies in precipitation patterns and intensity, particularly over longer lead times. To address this issue, we introduce RectiCast, a two-stage framework that explicitly decouples the rectification of mean-field shift from the generation of local stochasticity via a dual Flow Matching model. In the first stage, a deterministic model generates the posterior mean. In the second stage, we introduce a Rectifier to explicitly learn the distribution shift and produce a rectified mean. Subsequently, a Generator focuses on modeling the local stochasticity conditioned on the rectified mean. Experiments on two radar datasets demonstrate that RectiCast achieves significant performance improvements over existing state-of-the-art methods.

---

### DiffuMamba: High-Throughput Diffusion LMs with Mamba Backbone

Authors: Vaibhav Singh, Oleksiy Ostapenko, Pierre-André Noël, Eugene Belilovsky, Torsten Scholak
Published: 2025-11-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.15927v4

Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive (AR) generation, yet their reliance on Transformer backbones limits inference efficiency due to quadratic attention or KV-cache overhead. We introduce DiffuMamba, a masked diffusion language model built on a bidirectional Mamba backbone that combines the diffusion objective with linear-time sequence modeling, and DiffuMamba-H, a hybrid variant with interleaved attention. Across scales up to 1.3B parameters, our models match Transformer-based diffusion in downstream performance while achieving up to 8.2x and 4.3x higher inference throughput, respectively, on long sequences. We further present a systematic analysis of inference efficiency across modern DLM variants combining asymptotic complexity with empirical measurements. Notably, cache-efficient block diffusion with Mamba mixers emerges as the only strategy that scales linearly with sequence length and achieves the strongest performance across all baselines, suggesting a promising direction for future diffusion-based generation systems.

---

### RB-FT: Rationale-Bootstrapped Fine-Tuning for Video Classification

Authors: Meilong Xu, Di Fu, Jiaxing Zhang, Gong Yu, Jiayu Zheng, Xiaoling Hu, Dongdi Zhao, Feiyang Li, Chao Chen, Yong Cao
Published: 2025-11-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.15923v1

Vision Language Models (VLMs) are becoming increasingly integral to multimedia understanding; however, they often struggle with domain-specific video classification tasks, particularly in cases with limited data. This stems from a critical \textit{rationale gap}, where sparse domain data is insufficient to bridge the semantic distance between complex spatio-temporal content and abstract classification labels. We propose a two-stage self-improvement paradigm to bridge this gap without new annotations. First, we prompt the VLMs to generate detailed textual rationales for each video, compelling them to articulate the domain-specific logic. The VLM is then fine-tuned on these self-generated rationales, utilizing this intermediate supervision to align its representations with the nuances of the target domain. Second, conventional supervised fine-tuning (SFT) is performed on the task labels, achieving markedly higher effectiveness as a result of the model's pre-acquired domain reasoning. Extensive experiments on diverse datasets demonstrate that our method significantly outperforms direct SFT, validating self-generated rationale as an effective, annotation-efficient paradigm for adapting VLMs to domain-specific video analysis.

---

### Thinking, Faithful and Stable: Mitigating Hallucinations in LLMs

Authors: Chelsea Zou, Yiheng Yao, Basant Khalil
Published: 2025-11-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.15921v1

This project develops a self correcting framework for large language models (LLMs) that detects and mitigates hallucinations during multi-step reasoning. Rather than relying solely on final answer correctness, our approach leverages fine grained uncertainty signals: 1) self-assessed confidence alignment, and 2) token-level entropy spikes to detect unreliable and unfaithful reasoning in real time. We design a composite reward function that penalizes unjustified high confidence and entropy spikes, while encouraging stable and accurate reasoning trajectories. These signals guide a reinforcement learning (RL) policy that makes the model more introspective and shapes the model's generation behavior through confidence-aware reward feedback, improving not just outcome correctness but the coherence and faithfulness of their intermediate reasoning steps. Experiments show that our method improves both final answer accuracy and reasoning calibration, with ablations validating the individual contribution of each signal.

---

### AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization

Authors: Genghan Zhang, Shaowei Zhu, Anjiang Wei, Zhenyu Song, Allen Nie, Zhen Jia, Nandita Vijaykumar, Yida Wang, Kunle Olukotun
Published: 2025-11-19
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2511.15915v2

We present AccelOpt, a self-improving large language model (LLM) agentic system that autonomously optimizes kernels for emerging AI acclerators, eliminating the need for expert-provided hardware-specific optimization knowledge. AccelOpt explores the kernel optimization space through iterative generation, informed by an optimization memory that curates experiences and insights from previously encountered slow-fast kernel pairs. We build NKIBench, a new benchmark suite of AWS Trainium accelerator kernels with varying complexity extracted from real-world LLM workloads to evaluate the effectiveness of AccelOpt. Our evaluation confirms that AccelOpt's capability improves over time, boosting the average percentage of peak throughput from $49\%$ to $61\%$ on Trainium 1 and from $45\%$ to $59\%$ on Trainium 2 for NKIBench kernels. Moreover, AccelOpt is highly cost-effective: using open-source models, it matches the kernel improvements of Claude Sonnet 4 while being $26\times$ cheaper. The code is open-sourced at https://github.com/zhang677/AccelOpt.

---

### Unified all-atom molecule generation with neural fields

Authors: Matthieu Kirchmeyer, Pedro O. Pinheiro, Emma Willett, Karolis Martinkus, Joseph Kleinhenz, Emily K. Makowski, Andrew M. Watkins, Vladimir Gligorijevic, Richard Bonneau, Saeed Saremi
Published: 2025-11-19
Categories: cs.LG, q-bio.BM
arXiv: https://arxiv.org/abs/2511.15906v1

Generative models for structure-based drug design are often limited to a specific modality, restricting their broader applicability. To address this challenge, we introduce FuncBind, a framework based on computer vision to generate target-conditioned, all-atom molecules across atomic systems. FuncBind uses neural fields to represent molecules as continuous atomic densities and employs score-based generative models with modern architectures adapted from the computer vision literature. This modality-agnostic representation allows a single unified model to be trained on diverse atomic systems, from small to large molecules, and handle variable atom/residue counts, including non-canonical amino acids. FuncBind achieves competitive in silico performance in generating small molecules, macrocyclic peptides, and antibody complementarity-determining region loops, conditioned on target structures. FuncBind also generated in vitro novel antibody binders via de novo redesign of the complementarity-determining region H3 loop of two chosen co-crystal structures. As a final contribution, we introduce a new dataset and benchmark for structure-conditioned macrocyclic peptide generation. The code is available at https://github.com/prescient-design/funcbind.

---

### Bayesian Semiparametric Causal Inference: Targeted Doubly Robust Estimation of Treatment Effects

Authors: Gözde Sert, Abhishek Chakrabortty, Anirban Bhattacharya
Published: 2025-11-19
Categories: stat.ME, econ.EM, math.ST, stat.ML
arXiv: https://arxiv.org/abs/2511.15904v1

We propose a semiparametric Bayesian methodology for estimating the average treatment effect (ATE) within the potential outcomes framework using observational data with high-dimensional nuisance parameters. Our method introduces a Bayesian debiasing procedure that corrects for bias arising from nuisance estimation and employs a targeted modeling strategy based on summary statistics rather than the full data. These summary statistics are identified in a debiased manner, enabling the estimation of nuisance bias via weighted observables and facilitating hierarchical learning of the ATE. By combining debiasing with sample splitting, our approach separates nuisance estimation from inference on the target parameter, reducing sensitivity to nuisance model specification. We establish that, under mild conditions, the marginal posterior for the ATE satisfies a Bernstein-von Mises theorem when both nuisance models are correctly specified and remains consistent and robust when only one is correct, achieving Bayesian double robustness. This ensures asymptotic efficiency and frequentist validity. Extensive simulations confirm the theoretical results, demonstrating accurate point estimation and credible intervals with nominal coverage, even in high-dimensional settings. The proposed framework can also be extended to other causal estimands, and its key principles offer a general foundation for advancing Bayesian semiparametric inference more broadly.

---

### EEG Emotion Recognition Through Deep Learning

Authors: Roman Dolgopolyi, Antonis Chatzipanagiotou
Published: 2025-11-19
Categories: eess.SP, cs.LG
arXiv: https://arxiv.org/abs/2511.15902v1

An advanced emotion classification model was developed using a CNN-Transformer architecture for emotion recognition from EEG brain wave signals, effectively distinguishing among three emotional states, positive, neutral and negative. The model achieved a testing accuracy of 91%, outperforming traditional models such as SVM, DNN, and Logistic Regression. Training was conducted on a custom dataset created by merging data from SEED, SEED-FRA, and SEED-GER repositories, comprising 1,455 samples with EEG recordings labeled according to emotional states. The combined dataset represents one of the largest and most culturally diverse collections available. Additionally, the model allows for the reduction of the requirements of the EEG apparatus, by leveraging only 5 electrodes of the 62. This reduction demonstrates the feasibility of deploying a more affordable consumer-grade EEG headset, thereby enabling accessible, at-home use, while also requiring less computational power. This advancement sets the groundwork for future exploration into mood changes induced by media content consumption, an area that remains underresearched. Integration into medical, wellness, and home-health platforms could enable continuous, passive emotional monitoring, particularly beneficial in clinical or caregiving settings where traditional behavioral cues, such as facial expressions or vocal tone, are diminished, restricted, or difficult to interpret, thus potentially transforming mental health diagnostics and interventions...

---

### Global Resolution: Optimal Multi-Draft Speculative Sampling via Convex Minimization

Authors: Rahul Krishna Thomas, Arka Pal
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.15898v1

Speculative sampling reduces the latency of autoregressive decoding for target model LLMs without sacrificing inference quality, by using a cheap draft model to suggest a candidate token and a verification criterion to accept or resample this token. To improve acceptance and decoding efficiency, recent work has explored the multi-draft extension, where at each step $n$ draft tokens are generated, and the verification criterion is a distribution conditioned on these. When this criterion maximizes the probability of accepting some draft token, it is called the optimal transport (OT). However, finding the OT is difficult, as it is the solution of a linear program (OTLP) in over $V^n$ variables, with $V$ being the vocabulary size. Two recent theoretical works have reframed the OTLP in terms of importance sampling or subset selection. In this work, we prove that these formulations are equivalent to an exponentially large relaxed OTLP, so it remains infeasible to solve. Then, we reverse engineer subset selection to formulate the OTLP as a max-flow problem. With a novel application of polymatroid theory, we reduce the exponentially large OTLP to a convex optimization problem in at most $V$ variables. This allows us to devise an algorithm for optimal $n$-draft speculative sampling when the $n$ tokens are chosen i.i.d. from a single draft model, which can be tuned to arbitrary accuracy. Finally, we measure acceptance rates and algorithm runtimes for various $n$ and top-$k$ draft sampling settings. Our findings give the first multi-draft algorithm with 90% acceptance and under 100 ms of overhead per generated token with negligible deviation from the target model distribution.

---

### Cross-Balancing for Data-Informed Design and Efficient Analysis of Observational Studies

Authors: Ying Jin, José Zubizarreta
Published: 2025-11-19
Categories: stat.ME, math.ST, stat.AP, stat.ML
arXiv: https://arxiv.org/abs/2511.15896v1

Causal inference starts with a simple idea: compare groups that differ by treatment, not much else. Traditionally, similar groups are constructed using only observed covariates; however, it remains a long-standing challenge to incorporate available outcome data into the study design while preserving valid inference. In this paper, we study the general problem of covariate adjustment, effect estimation, and statistical inference when balancing features are constructed or selected with the aid of outcome information from the data. We propose cross-balancing, a method that uses sample splitting to separate the error in feature construction from the error in weight estimation. Our framework addresses two cases: one where the features are learned functions and one where they are selected from a potentially high-dimensional dictionary. In both cases, we establish mild and general conditions under which cross-balancing produces consistent, asymptotically normal, and efficient estimators. In the learned-function case, cross-balancing achieves finite-sample bias reduction relative to plug-in-type estimators, and is multiply robust when the learned features converge at slow rates. In the variable-selection case, cross-balancing only requires a product condition on how well the selected variables approximate true functions. We illustrate cross-balancing in extensive simulations and an observational study, showing that careful use of outcome information can substantially improve both estimation and inference while maintaining interpretability.

---

### Decomposing Theory of Mind: How Emotional Processing Mediates ToM Abilities in LLMs

Authors: Ivan Chulo, Ananya Joshi
Published: 2025-11-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.15895v1

Recent work shows activation steering substantially improves language models' Theory of Mind (ToM) (Bortoletto et al. 2024), yet the mechanisms of what changes occur internally that leads to different outputs remains unclear. We propose decomposing ToM in LLMs by comparing steered versus baseline LLMs' activations using linear probes trained on 45 cognitive actions. We applied Contrastive Activation Addition (CAA) steering to Gemma-3-4B and evaluated it on 1,000 BigToM forward belief scenarios (Gandhi et al. 2023), we find improved performance on belief attribution tasks (32.5\% to 46.7\% accuracy) is mediated by activations processing emotional content : emotion perception (+2.23), emotion valuing (+2.20), while suppressing analytical processes: questioning (-0.78), convergent thinking (-1.59). This suggests that successful ToM abilities in LLMs are mediated by emotional understanding, not analytical reasoning.

---

### Mind the Motions: Benchmarking Theory-of-Mind in Everyday Body Language

Authors: Seungbeen Lee, Jinhong Jeong, Donghyun Kim, Yejin Son, Youngjae Yu
Published: 2025-11-19
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.15887v2

Our ability to interpret others' mental states through nonverbal cues (NVCs) is fundamental to our survival and social cohesion. While existing Theory of Mind (ToM) benchmarks have primarily focused on false-belief tasks and reasoning with asymmetric information, they overlook other mental states beyond belief and the rich tapestry of human nonverbal communication. We present Motion2Mind, a framework for evaluating the ToM capabilities of machines in interpreting NVCs. Leveraging an expert-curated body-language reference as a proxy knowledge base, we build Motion2Mind, a carefully curated video dataset with fine-grained nonverbal cue annotations paired with manually verified psychological interpretations. It encompasses 222 types of nonverbal cues and 397 mind states. Our evaluation reveals that current AI systems struggle significantly with NVC interpretation, exhibiting not only a substantial performance gap in Detection, as well as patterns of over-interpretation in Explanation compared to human annotators.

---

### What Really Counts? Examining Step and Token Level Attribution in Multilingual CoT Reasoning

Authors: Jeremias Ferrao, Ezgi Basar, Khondoker Ittehadul Islam, Mahrokh Hassani
Published: 2025-11-19
Categories: cs.CL
arXiv: https://arxiv.org/abs/2511.15886v1

This study investigates the attribution patterns underlying Chain-of-Thought (CoT) reasoning in multilingual LLMs. While prior works demonstrate the role of CoT prompting in improving task performance, there are concerns regarding the faithfulness and interpretability of the generated reasoning chains. To assess these properties across languages, we applied two complementary attribution methods--ContextCite for step-level attribution and Inseq for token-level attribution--to the Qwen2.5 1.5B-Instruct model using the MGSM benchmark. Our experimental results highlight key findings such as: (1) attribution scores excessively emphasize the final reasoning step, particularly in incorrect generations; (2) structured CoT prompting significantly improves accuracy primarily for high-resource Latin-script languages; and (3) controlled perturbations via negation and distractor sentences reduce model accuracy and attribution coherence. These findings highlight the limitations of CoT prompting, particularly in terms of multilingual robustness and interpretive transparency.

---

### AnyBox: Efficient Zero-Shot 9DoF Pose Estimation of Boxes for Robotic Manipulation

Authors: Yintao Ma, Sajjad Pakdamansavoji, Charles Eret, Rui Heng Yang, Xuan Zhao, Yingxue Zhang, Tongtong Cao, Amir Rasouli
Published: 2025-11-19
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.15884v2

Recovering the 9D pose of objects, both their 6D pose and 3D dimensions, under clutter and occlusion is a core requirement for warehouse automation, logistics, and manufacturing. Model-based methods are accurate but assume an instance-specific CAD model for every object, which is costly to maintain as inventories change. Model-free and category-level methods relax this assumption, yet they remain vulnerable to the symmetry, weak texture, and heavy occlusion that characterize stacked storage boxes, and they ignore the strong structural priors such scenes provide. We present \textbf{AnyBox}, an efficient zero-shot framework that exploits the geometric regularity of boxes to jointly recover pose and dimensions from a single RGB-D observation. Starting from a canonical category template, AnyBox alternates between pose and scale estimation, using the discrepancy between the reprojected template and the observed mask to drive a binary search over box dimensions. Two lightweight components make this practical: a depth-consistency filter that rejects the implausible hypotheses induced by box symmetry, and an early-stopping rule that replaces the remaining search with a single closed-form update. On public benchmarks and an in-house warehouse dataset, AnyBox improves detection AP by up to 36 points, more than doubling the previous best, and approaches instance-level pipelines that have access to ground-truth CAD models. These gains transfer downstream, raising success by 28\% on a cluttered robotic box-shelving task.

---

### Automatic Uncertainty-Aware Synthetic Data Bootstrapping for Historical Map Segmentation

Authors: Lukas Arzoumanidis, Julius Knechtel, Jan-Henrik Haunert, Youness Dehbi
Published: 2025-11-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.15875v2

The automated analysis of historical documents, particularly maps, has drastically benefited from advances in deep learning and its success across various computer vision applications. However, most deep learning-based methods heavily rely on large amounts of annotated training data, which are typically unavailable for historical maps, especially for those belonging to specific, homogeneous cartographic domains, also known as corpora. Creating high-quality training data suitable for machine learning often takes a significant amount of time and involves extensive manual effort. While synthetic training data can alleviate the scarcity of real-world samples, it often lacks the affinity (realism) and diversity (variation) necessary for effective learning. By transferring the cartographic style of a historical map corpus onto modern vector data, we bootstrap an effectively unlimited number of synthetic historical maps suitable for tasks such as land-cover interpretation of a homogeneous historical map corpus. We propose an automatic deep generative approach and an alternative manual stochastic degradation technique to emulate the visual uncertainty and noise, also known as aleatoric uncertainty, commonly observed in historical map scans. To quantitatively evaluate the effectiveness and applicability of our approach, the bootstrapped training datasets were employed for domain-adaptive semantic segmentation on a homogeneous map corpus using a Self-Constructing Graph Convolutional Network, enabling a comprehensive assessment of the impact of our data bootstrapping methods.

---

### WALDO: Where Unseen Model-based 6D Pose Estimation Meets Occlusion

Authors: Sajjad Pakdamansavoji, Yintao Ma, Amir Rasouli, Tongtong Cao
Published: 2025-11-19
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2511.15874v1

Accurate 6D object pose estimation is vital for robotics, augmented reality, and scene understanding. For seen objects, high accuracy is often attainable via per-object fine-tuning but generalizing to unseen objects remains a challenge. To address this problem, past arts assume access to CAD models at test time and typically follow a multi-stage pipeline to estimate poses: detect and segment the object, propose an initial pose, and then refine it. Under occlusion, however, the early-stage of such pipelines are prone to errors, which can propagate through the sequential processing, and consequently degrade the performance. To remedy this shortcoming, we propose four novel extensions to model-based 6D pose estimation methods: (i) a dynamic non-uniform dense sampling strategy that focuses computation on visible regions, reducing occlusion-induced errors; (ii) a multi-hypothesis inference mechanism that retains several confidence-ranked pose candidates, mitigating brittle single-path failures; (iii) iterative refinement to progressively improve pose accuracy; and (iv) series of occlusion-focused training augmentations that strengthen robustness and generalization. Furthermore, we propose a new weighted by visibility metric for evaluation under occlusion to minimize the bias in the existing protocols. Via extensive empirical evaluations, we show that our proposed approach achieves more than 5% improvement in accuracy on ICBIN and more than 2% on BOP dataset benchmarks, while achieving approximately 3 times faster inference.

---

### AquaSentinel: Next-Generation AI System Integrating Sensor Networks for Urban Underground Water Pipeline Anomaly Detection via Collaborative MoE-LLM Agent Architecture

Authors: Qiming Guo, Bishal Khatri, Wenbo Sun, Jinwen Tang, Hua Zhang, Wenlu Wang
Published: 2025-11-19
Categories: cs.CE, cs.AI
arXiv: https://arxiv.org/abs/2511.15870v1

Underground pipeline leaks and infiltrations pose significant threats to water security and environmental safety. Traditional manual inspection methods provide limited coverage and delayed response, often missing critical anomalies. This paper proposes AquaSentinel, a novel physics-informed AI system for real-time anomaly detection in urban underground water pipeline networks. We introduce four key innovations: (1) strategic sparse sensor deployment at high-centrality nodes combined with physics-based state augmentation to achieve network-wide observability from minimal infrastructure; (2) the RTCA (Real-Time Cumulative Anomaly) detection algorithm, which employs dual-threshold monitoring with adaptive statistics to distinguish transient fluctuations from genuine anomalies; (3) a Mixture of Experts (MoE) ensemble of spatiotemporal graph neural networks that provides robust predictions by dynamically weighting model contributions; (4) causal flow-based leak localization that traces anomalies upstream to identify source nodes and affected pipe segments. Our system strategically deploys sensors at critical network junctions and leverages physics-based modeling to propagate measurements to unmonitored nodes, creating virtual sensors that enhance data availability across the entire network. Experimental evaluation using 110 leak scenarios demonstrates that AquaSentinel achieves 100% detection accuracy. This work advances pipeline monitoring by demonstrating that physics-informed sparse sensing can match the performance of dense deployments at a fraction of the cost, providing a practical solution for aging urban infrastructure.

---

### The Subtle Art of Defection: Understanding Uncooperative Behaviors in LLM based Multi-Agent Systems

Authors: Devang Kulshreshtha, Wanyu Du, Raghav Jain, Srikanth Doss, Hang Su, Sandesh Swamy, Yanjun Qi
Published: 2025-11-19
Categories: cs.MA, cs.CL
arXiv: https://arxiv.org/abs/2511.15862v2

This paper introduces a novel framework for simulating and analyzing how uncooperative behaviors can destabilize or collapse LLM-based multi-agent systems. Our framework includes two key components: (1) a game theory-based taxonomy of uncooperative agent behaviors, addressing a notable gap in the existing literature; and (2) a structured, multi-stage simulation pipeline that dynamically generates and refines uncooperative behaviors as agents' states evolve. We evaluate the framework via a collaborative resource management setting, measuring system stability using metrics such as survival time and resource overuse rate. Empirically, our framework achieves 96.7% accuracy in generating realistic uncooperative behaviors, validated by human evaluations. Our results reveal a striking contrast: cooperative agents maintain perfect system stability (100% survival over 12 rounds with 0% resource overuse), while any uncooperative behavior can trigger rapid system collapse within 1 to 7 rounds. We also evaluate LLM-based defense methods, finding they detect some uncooperative behaviors, but some behaviors remain largely undetectable. These gaps highlight how uncooperative agents degrade collective outcomes and underscore the need for more resilient multi-agent systems.

---

### A Crowdsourced Study of ChatBot Influence in Value-Driven Decision Making Scenarios

Authors: Anthony Wise, Xinyi Zhou, Martin Reimann, Anind Dey, Leilani Battle
Published: 2025-11-19
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2511.15857v1

Similar to social media bots that shape public opinion, healthcare and financial decisions, LLM-based ChatBots like ChatGPT can persuade users to alter their behavior. Unlike prior work that persuades via overt-partisan bias or misinformation, we test whether framing alone suffices. We conducted a crowdsourced study, where 336 participants interacted with a neutral or one of two value-framed ChatBots while deciding to alter US defense spending. In this single policy domain with controlled content, participants exposed to value-framed ChatBots significantly changed their budget choices relative to the neutral control. When the frame misaligned with their values, some participants reinforced their original preference, revealing a potentially replicable backfire effect, originally considered rare in the literature. These findings suggest that value-framing alone lowers the barrier for manipulative uses of LLMs, revealing risks distinct from overt bias or misinformation, and clarifying risks to countering misinformation.

---

### GLOBE: Accurate and Generalizable PDE Surrogates using Domain-Inspired Architectures and Equivariances

Authors: Peter Sharpe
Published: 2025-11-19
Categories: cs.LG, physics.flu-dyn
arXiv: https://arxiv.org/abs/2511.15856v3

We introduce GLOBE, a new neural surrogate for homogeneous PDEs that draws inductive bias from boundary-element methods and equivariant ML. GLOBE represents solutions as superpositions of learnable Green's-function-like kernels evaluated from boundary faces to targets, composed across multiscale branches and communication hyperlayers. The architecture is translation-, rotation-, and parity-equivariant; discretization-invariant in the fine-mesh limit; and units-invariant via rigorous nondimensionalization. An explicit far-field decay envelope stabilizes extrapolation, boundary-to-boundary hyperlayer communication mediates long-range coupling, and the all-to-all boundary-to-target evaluation yields a global receptive field that respects PDE information flow, even for elliptic PDEs. On AirFRANS (steady incompressible RANS over NACA airfoils), GLOBE achieves substantial accuracy improvements. On the "Full" split, it reduces mean-squared error by roughly 200x on all fields relative to the dataset's reference baselines, and roughly 50x relative to the next-best-performing model. In the "Scarce" split, it achieves over 100x lower error on velocity and pressure fields and over 600x lower error on surface pressure than Transolver. Qualitative results show sharp near-wall gradients, coherent wakes, and limited errors under modest extrapolation in Reynolds number and angle of attack. In addition to this accuracy, the model is quite compact (117k parameters), and fields can be evaluated at arbitrary points during inference. We also demonstrate the ability to train and predict with non-watertight meshes, which has strong practical implications. These results show that rigorous physics- and domain-inspired inductive biases can achieve large gains in accuracy, generalizability, and practicality for ML-based PDE surrogates for industrial computer-aided engineering (CAE).

---

### discretize_distributions: Efficient Quantization of Gaussian Mixtures with Guarantees in Wasserstein Distance

Authors: Steven Adams, Elize Alwash, Luca Laurenti
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.15854v1

We present discretize_distributions, a Python package that efficiently constructs discrete approximations of Gaussian mixture distributions and provides guarantees on the approximation error in Wasserstein distance. The package implements state-of-the-art quantization methods for Gaussian mixture models and extends them to improve scalability. It further integrates complementary quantization strategies such as sigma-point methods and provides a modular interface that supports custom schemes and integration into control and verification pipelines for cyber-physical systems. We benchmark the package on various examples, including high-dimensional, large, and degenerate Gaussian mixtures, and demonstrate that discretize_distributions produces accurate approximations at low computational cost.

---

### The Ensemble Kalman Inversion Race

Authors: Rebecca Gjini, Matthias Morzfeld, Oliver R. A. Dunbar, Tapio Schneider
Published: 2025-11-19
Categories: physics.data-an, stat.ML
arXiv: https://arxiv.org/abs/2511.15853v2

Ensemble Kalman methods were initially developed to solve nonlinear data assimilation problems in oceanography but are now popular in applications far beyond their original use cases. Of particular interest is climate model calibration. As hybrid physics and machine-learning models advance, the number of parameters and complexity of parameterizations in climate models will continue to grow. To fully realize these advances, we must move from laborious hand-tuning to calibration-driven model development in rapid iteration cycles. Thus, robust calibration of these parameters plays an increasingly important role. We focus on learning climate model parameters by minimizing the misfit between modeled and observed climate statistics in an idealized setting. Ensemble Kalman methods are a natural choice for this problem because they are derivative-free, scalable to high dimensions, and robust to noise caused by statistical observations. Given the many variants of ensemble methods proposed, an important question is: Which ensemble Kalman method should be used for climate model calibration? To answer this question, we perform systematic numerical experiments to explore the relative computational efficiencies of several ensemble Kalman methods. The numerical experiments involve statistical observations of Lorenz-type models of increasing complexity, frequently used to represent simplified atmospheric systems, with some featuring neural network parameterizations. For each test problem, several ensemble Kalman methods and a derivative-based method "race" to reach a specified accuracy, and we measure the computational cost required to achieve the desired accuracy. We investigate how prior information and the parameter or data dimensions define the computational costs of the various methods.

---

### Step-Audio-R1 Technical Report

Authors: Fei Tian, Xiangyu Tony Zhang, Yuxin Zhang, Haoyang Zhang, Yuxin Li, Daijiao Liu, Yayue Deng, Donghang Wu, Jun Chen, Liang Zhao, Chengyuan Yao, Hexin Liu et al.
Published: 2025-11-19
Categories: cs.AI, cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2511.15848v2

Recent advances in reasoning models have demonstrated remarkable success in text and vision domains through extended chain-of-thought deliberation. However, a perplexing phenomenon persists in audio language models: they consistently perform better with minimal or no reasoning, raising a fundamental question - can audio intelligence truly benefit from deliberate thinking? We introduce Step-Audio-R1, the first audio reasoning model that successfully unlocks reasoning capabilities in the audio domain. Through our proposed Modality-Grounded Reasoning Distillation (MGRD) framework, Step-Audio-R1 learns to generate audio-relevant reasoning chains that genuinely ground themselves in acoustic features rather than hallucinating disconnected deliberations. Our model exhibits strong audio reasoning capabilities, surpassing Gemini 2.5 Pro and achieving performance comparable to the state-of-the-art Gemini 3 Pro across comprehensive audio understanding and reasoning benchmarks spanning speech, environmental sounds, and music. These results demonstrate that reasoning is a transferable capability across modalities when appropriately anchored, transforming extended deliberation from a liability into a powerful asset for audio intelligence. By establishing the first successful audio reasoning model, Step-Audio-R1 opens new pathways toward building truly multimodal reasoning systems that think deeply across all sensory modalities.

---

### Transparent Early ICU Mortality Prediction with Clinical Transformer and Per-Case Modality Attribution

Authors: Alexander Bakumenko, Janine Hoelscher, Hudson Smith
Published: 2025-11-19
Categories: cs.LG
arXiv: https://arxiv.org/abs/2511.15847v1

Early identification of intensive care patients at risk of in-hospital mortality enables timely intervention and efficient resource allocation. Despite high predictive performance, existing machine learning approaches lack transparency and robustness, limiting clinical adoption. We present a lightweight, transparent multimodal ensemble that fuses physiological time-series measurements with unstructured clinical notes from the first 48 hours of an ICU stay. A logistic regression model combines predictions from two modality-specific models: a bidirectional LSTM for vitals and a finetuned ClinicalModernBERT transformer for notes. This traceable architecture allows for multilevel interpretability: feature attributions within each modality and direct per-case modality attributions quantifying how vitals and notes influence each decision. On the MIMIC-III benchmark, our late-fusion ensemble improves discrimination over the best single model (AUPRC 0.565 vs. 0.526; AUROC 0.891 vs. 0.876) while maintaining well-calibrated predictions. The system remains robust through a calibrated fallback when a modality is missing. These results demonstrate competitive performance with reliable, auditable risk estimates and transparent, predictable operation, which together are crucial for clinical use.

---

### The Loss of Control Playbook: Degrees, Dynamics, and Preparedness

Authors: Charlotte Stix, Annika Hallensleben, Alejandro Ortega, Matteo Pistillo
Published: 2025-11-19
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2511.15846v5

This research report addresses the absence of an actionable definition for Loss of Control (LoC) in AI systems by developing a novel taxonomy and preparedness framework. Despite increasing policy and research attention, existing LoC definitions vary significantly in scope and timeline, hindering effective LoC assessment and mitigation. To address this issue, we draw from an extensive literature review and propose a graded LoC taxonomy, based on the metrics of severity and persistence, that distinguishes between Deviation, Bounded LoC, and Strict LoC. We model pathways toward a societal state of vulnerability in which sufficiently advanced AI systems have acquired or could acquire the means to cause Bounded or Strict LoC once a catalyst, either misalignment or pure malfunction, materializes. We argue that this state becomes increasingly likely over time, absent strategic intervention, and propose a strategy to avoid reaching a state of vulnerability. Rather than focusing solely on intervening on AI capabilities and propensities potentially relevant for LoC or on preventing potential catalysts, we introduce a complementary framework that emphasizes three extrinsic factors: Deployment context, Affordances, and Permissions (the DAP framework). Compared to work on intrinsic factors and catalysts, this framework has the unfair advantage of being actionable today. Finally, we put forward a plan to maintain preparedness and prevent the occurrence of LoC outcomes should a state of societal vulnerability be reached, focusing on governance measures (threat modeling, deployment policies, emergency response) and technical controls (pre-deployment testing, control measures, monitoring) that could maintain a condition of perennial suspension.

---

### Attention-Based Feature Online Conformal Prediction for Time Series

Authors: Meiyi Zhu, Caili Guo, Chunyan Feng, Osvaldo Simeone
Published: 2025-11-19
Categories: cs.LG, cs.IT, eess.SP
arXiv: https://arxiv.org/abs/2511.15838v1

Online conformal prediction (OCP) wraps around any pre-trained predictor to produce prediction sets with coverage guarantees that hold irrespective of temporal dependencies or distribution shifts. However, standard OCP faces two key limitations: it operates in the output space using simple nonconformity (NC) scores, and it treats all historical observations uniformly when estimating quantiles. This paper introduces attention-based feature OCP (AFOCP), which addresses both limitations through two key innovations. First, AFOCP operates in the feature space of pre-trained neural networks, leveraging learned representations to construct more compact prediction sets by concentrating on task-relevant information while suppressing nuisance variation. Second, AFOCP incorporates an attention mechanism that adaptively weights historical observations based on their relevance to the current test point, effectively handling non-stationarity and distribution shifts. We provide theoretical guarantees showing that AFOCP maintains long-term coverage while provably achieving smaller prediction intervals than standard OCP under mild regularity conditions. Extensive experiments on synthetic and real-world time series datasets demonstrate that AFOCP consistently reduces the size of prediction intervals by as much as $88\%$ as compared to OCP, while maintaining target coverage levels, validating the benefits of both feature-space calibration and attention-based adaptive weighting.

---

### EfficientSAM3: Progressive Hierarchical Distillation for Video Concept Segmentation from SAM1, 2, and 3

Authors: Chengxi Zeng, Yuxuan Jiang, Aaron Zhang
Published: 2025-11-19
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2511.15833v1

The Segment Anything Model 3 (SAM3) advances visual understanding with Promptable Concept Segmentation (PCS) across images and videos, but its unified architecture (shared vision backbone, DETR-style detector, dense-memory tracker) remains prohibitive for on-device use. We present EfficientSAM3, a family of efficient models built on Progressive Hierarchical Distillation (PHD) that transfers capability from SAM3 to lightweight students in three stages: (1) Encoder Distillation aligns image features via prompt-in-the-loop training on SA-1B; (2) Temporal Memory Distillation replaces dense memory with a compact Perceiver-based module trained on SA-V to compress and retrieve spatiotemporal features efficiently; and (3) End-to-End Fine-Tuning refines the full pipeline on the official SAM3 PCS data to preserve concept-level performance. PHD yields a spectrum of student variants using RepViT, TinyViT, and EfficientViT backbones, enabling on-device concept segmentation and tracking while maintaining high fidelity to teacher behavior. We benchmark on popular VOS datasets, and compare with varies of releated work, achieing strong performance-efficiency trade-offs.

---

### UniFit: Towards Universal Virtual Try-on with MLLM-Guided Semantic Alignment

Authors: Wei Zhang, Yeying Jin, Xin Li, Yan Zhang, Xiaofeng Cong, Cong Wang, Fengcai Qiao, zhichao Lian
Published: 2025-11-19
Categories: cs.CV
arXiv: https://arxiv.org/abs/2511.15831v2

Image-based virtual try-on (VTON) aims to synthesize photorealistic images of a person wearing specified garments. Despite significant progress, building a universal VTON framework that can flexibly handle diverse and complex tasks remains a major challenge. Recent methods explore multi-task VTON frameworks guided by textual instructions, yet they still face two key limitations: (1) semantic gap between text instructions and reference images, and (2) data scarcity in complex scenarios. To address these challenges, we propose UniFit, a universal VTON framework driven by a Multimodal Large Language Model (MLLM). Specifically, we introduce an MLLM-Guided Semantic Alignment Module (MGSA), which integrates multimodal inputs using an MLLM and a set of learnable queries. By imposing a semantic alignment loss, MGSA captures cross-modal semantic relationships and provides coherent and explicit semantic guidance for the generative process, thereby reducing the semantic gap. Moreover, by devising a two-stage progressive training strategy with a self-synthesis pipeline, UniFit is able to learn complex tasks from limited data. Extensive experiments show that UniFit not only supports a wide range of VTON tasks, including multi-garment and model-to-model try-on, but also achieves state-of-the-art performance. The source code and pretrained models are available at https://github.com/zwplus/UniFit.

---

### Mini Amusement Parks (MAPs): A Testbed for Modelling Business Decisions

Authors: Stéphane Aroca-Ouellette, Ian Berlot-Attwell, Panagiotis Lymperopoulos, Abhiramon Rajasekharan, Tongqi Zhu, Herin Kang, Kaheer Suleman, Sam Pasupalak
Published: 2025-11-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.15830v3

Despite rapid progress in artificial intelligence, current systems struggle with the interconnected challenges that define real-world decision making. Practical domains such as business management require open-ended optimization, actively learning environment dynamics from sparse experience, planning over long horizons in stochastic settings, and reasoning over spatial information. Yet no existing human--AI benchmarks assess how well agents integrate these challenges in a grounded decision-making context. To this end, we introduce Mini Amusement Parks (MAPs), an amusement-park simulator designed to evaluate an agent's ability to model its environment, anticipate long-term consequences under uncertainty, and strategically operate a complex business. We provide expert human performance and a comprehensive evaluation of state-of-the-art agents, finding experts outperform these systems by 11.4x on easy mode and 15.3x on medium mode. Our analysis reveals persistent weaknesses in long-horizon planning, sample-efficient learning, spatial reasoning, and modelling uncertainty. By unifying these challenges within a single environment, MAPs offers a new foundation for benchmarking agents capable of adaptable decision making. Code: https://github.com/Skyfall-Research/MAPs

---

### IMACT-CXR: An Interactive Multi-Agent Conversational Tutoring System for Chest X-Ray Interpretation

Authors: Tuan-Anh Le, Anh Mai Vu, David Yang, Akash Awasthi, Hien Van Nguyen
Published: 2025-11-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.15825v2

IMACT-CXR is an interactive multi-agent conversational tutor that helps trainees interpret chest X-rays by unifying spatial annotation, gaze analysis, knowledge retrieval, and image-grounded reasoning in a single AutoGen-based workflow. The tutor simultaneously ingests learner bounding boxes, gaze samples, and free-text observations. Specialized agents evaluate localization quality, generate Socratic coaching, retrieve PubMed evidence, suggest similar cases from REFLACX, and trigger NV-Reason-CXR-3B for vision-language reasoning when mastery remains low or the learner explicitly asks. Bayesian Knowledge Tracing (BKT) maintains skill-specific mastery estimates that drive both knowledge reinforcement and case similarity retrieval. A lung-lobe segmentation module derived from a TensorFlow U-Net enables anatomically aware gaze feedback, and safety prompts prevent premature disclosure of ground-truth labels. We describe the system architecture, implementation highlights, and integration with the REFLACX dataset for real DICOM cases. IMACT-CXR demonstrates responsive tutoring flows with bounded latency, precise control over answer leakage, and extensibility toward live residency deployment. Preliminary evaluation shows improved localization and diagnostic reasoning compared to baselines.

---

### Atlas Gaussian processes on restricted domains and point clouds

Authors: Mu Niu, Yue Zhang, Ke Ye, Pokman Cheung, Yizhu Wang, Xiaochen Yang
Published: 2025-11-19
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2511.15822v1

In real-world applications, data often reside in restricted domains with unknown boundaries, or as high-dimensional point clouds lying on a lower-dimensional, nontrivial, unknown manifold. Traditional Gaussian Processes (GPs) struggle to capture the underlying geometry in such settings. Some existing methods assume a flat space embedded in a point cloud, which can be represented by a single latent chart (latent space), while others exhibit weak performance when the point cloud is sparse or irregularly sampled. The goal of this work is to address these challenges. The main contributions are twofold: (1) We establish the Atlas Brownian Motion (BM) framework for estimating the heat kernel on point clouds with unknown geometries and nontrivial topological structures; (2) Instead of directly using the heat kernel estimates, we construct a Riemannian corrected kernel by combining the global heat kernel with local RBF kernel and leading to the formulation of Riemannian-corrected Atlas Gaussian Processes (RC-AGPs). The resulting RC-AGPs are applied to regression tasks across synthetic and real-world datasets. These examples demonstrate that our method outperforms existing approaches in both heat kernel estimation and regression accuracy. It improves statistical inference by effectively bridging the gap between complex, high-dimensional observations and manifold-based inferences.

---

### Beyond Tsybakov: Model Margin Noise and $\mathcal{H}$-Consistency Bounds

Authors: Mehryar Mohri, Yutao Zhong
Published: 2025-11-19
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2511.15816v1

We introduce a new low-noise condition for classification, the Model Margin Noise (MM noise) assumption, and derive enhanced $\mathcal{H}$-consistency bounds under this condition. MM noise is weaker than Tsybakov noise condition: it is implied by Tsybakov noise condition but can hold even when Tsybakov fails, because it depends on the discrepancy between a given hypothesis and the Bayes-classifier rather than on the intrinsic distributional minimal margin (see Figure 1 for an illustration of an explicit example). This hypothesis-dependent assumption yields enhanced $\mathcal{H}$-consistency bounds for both binary and multi-class classification. Our results extend the enhanced $\mathcal{H}$-consistency bounds of Mao, Mohri, and Zhong (2025a) with the same favorable exponents but under a weaker assumption than the Tsybakov noise condition; they interpolate smoothly between linear and square-root regimes for intermediate noise levels. We also instantiate these bounds for common surrogate loss families and provide illustrative tables.

---

### Multidimensional scaling of two-mode three-way asymmetric dissimilarities: finding archetypal profiles and clustering

Authors: Mireia Mollar-Gumbau, Aleix Alcacer, Rafael Benitez, Vicente J. Bolos, Irene Epifanio
Published: 2025-11-19
Categories: stat.ME, stat.AP, stat.ML
arXiv: https://arxiv.org/abs/2511.15813v2

Multidimensional scaling visualizes dissimilarities among objects and reduces data dimensionality. While many methods address symmetric proximity data, asymmetric and especially three-way proximity data (capturing relationships across multiple occasions) remain underexplored. The h-plot enables the analysis of asymmetric and non-reflexive relationships by embedding dissimilarities in a Euclidean space, allowing further techniques like archetypoid analysis to identify representative extreme profiles. However, no existing methods extract archetypal profiles from three-way asymmetric proximity data. This work extends the h-plot methodology to three-way proximity data under both symmetric and asymmetric, conditional and unconditional frameworks. The proposed approach offers several advantages: intuitive interpretability through a unified Euclidean representation; an explicit, eigenvector-based analytical solution free from local minima; scale invariance under linear transformations; computational efficiency for large matrices; and a straightforward goodness-of-fit evaluation. Furthermore, it enables the identification of archetypal profiles and clustering structures for three-way asymmetric proximities. Its performance is compared with existing models for multidimensional scaling and clustering, and illustrated through a financial application. All data and code are provided to facilitate reproducibility.

---

### TopoReformer: Mitigating Adversarial Attacks Using Topological Purification in OCR Models

Authors: Bhagyesh Kumar, A S Aravinthakashan, Akshat Satyanarayan, Ishaan Gakhar, Ujjwal Verma
Published: 2025-11-19
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2511.15807v1

Adversarially perturbed images of text can cause sophisticated OCR systems to produce misleading or incorrect transcriptions from seemingly invisible changes to humans. Some of these perturbations even survive physical capture, posing security risks to high-stakes applications such as document processing, license plate recognition, and automated compliance systems. Existing defenses, such as adversarial training, input preprocessing, or post-recognition correction, are often model-specific, computationally expensive, and affect performance on unperturbed inputs while remaining vulnerable to unseen or adaptive attacks. To address these challenges, TopoReformer is introduced, a model-agnostic reformation pipeline that mitigates adversarial perturbations while preserving the structural integrity of text images. Topology studies properties of shapes and spaces that remain unchanged under continuous deformations, focusing on global structures such as connectivity, holes, and loops rather than exact distance. Leveraging these topological features, TopoReformer employs a topological autoencoder to enforce manifold-level consistency in latent space and improve robustness without explicit gradient regularization. The proposed method is benchmarked on EMNIST, MNIST, against standard adversarial attacks (FGSM, PGD, Carlini-Wagner), adaptive attacks (EOT, BDPA), and an OCR-specific watermark attack (FAWA).

---

### SURFing to the Fundamental Limit of Jet Tagging

Authors: Ian Pang, Darius A. Faroughy, David Shih, Ranit Das, Gregor Kasieczka
Published: 2025-11-19
Categories: hep-ph, cs.LG, hep-ex, physics.data-an
arXiv: https://arxiv.org/abs/2511.15779v1

Beyond the practical goal of improving search and measurement sensitivity through better jet tagging algorithms, there is a deeper question: what are their upper performance limits? Generative surrogate models with learned likelihood functions offer a new approach to this problem, provided the surrogate correctly captures the underlying data distribution. In this work, we introduce the SUrrogate ReFerence (SURF) method, a new approach to validating generative models. This framework enables exact Neyman-Pearson tests by training the target model on samples from another tractable surrogate, which is itself trained on real data. We argue that the EPiC-FM generative model is a valid surrogate reference for JetClass jets and apply SURF to show that modern jet taggers may already be operating close to the true statistical limit. By contrast, we find that autoregressive GPT models unphysically exaggerate top vs. QCD separation power encoded in the surrogate reference, implying that they are giving a misleading picture of the fundamental limit.

---

### Balancing Natural Language Processing Accuracy and Normalisation in Extracting Medical Insights

Authors: Paulina Tworek, Miłosz Bargieł, Yousef Khan, Tomasz Pełech-Pilichowski, Marek Mikołajczyk, Roman Lewandowski, Jose Sousa
Published: 2025-11-19
Categories: cs.AI
arXiv: https://arxiv.org/abs/2511.15778v1

Extracting structured medical insights from unstructured clinical text using Natural Language Processing (NLP) remains an open challenge in healthcare, particularly in non-English contexts where resources are scarce. This study presents a comparative analysis of NLP low-compute rule-based methods and Large Language Models (LLMs) for information extraction from electronic health records (EHR) obtained from the Voivodeship Rehabilitation Hospital for Children in Ameryka, Poland. We evaluate both approaches by extracting patient demographics, clinical findings, and prescribed medications while examining the effects of lack of text normalisation and translation-induced information loss. Results demonstrate that rule-based methods provide higher accuracy in information retrieval tasks, particularly for age and sex extraction. However, LLMs offer greater adaptability and scalability, excelling in drug name recognition. The effectiveness of the LLMs was compared with texts originally in Polish and those translated into English, assessing the impact of translation. These findings highlight the trade-offs between accuracy, normalisation, and computational cost when deploying NLP in healthcare settings. We argue for hybrid approaches that combine the precision of rule-based systems with the adaptability of LLMs, offering a practical path toward more reliable and resource-efficient clinical NLP in real-world hospitals.

---

### UniUltra: Interactive Parameter-Efficient SAM2 for Universal Ultrasound Segmentation

Authors: Yue Li, Qing Xu, Yixuan Zhang, Xiangjian He, Qian Zhang, Yuan Yao, Fiseha B. Tesema, Xin Chen, Ruili Wang, Zhen Chen, Chang Wen Chen
Published: 2025-11-19
Categories: eess.IV, cs.CV
arXiv: https://arxiv.org/abs/2511.15771v1

The Segment Anything Model 2 (SAM2) demonstrates remarkable universal segmentation capabilities on natural images. However, its performance on ultrasound images is significantly degraded due to domain disparities. This limitation raises two critical challenges: how to efficiently adapt SAM2 to ultrasound imaging while maintaining parameter efficiency, and how to deploy the adapted model effectively in resource-constrained clinical environments. To address these issues, we propose UniUltra for universal ultrasound segmentation. Specifically, we first introduce a novel context-edge hybrid adapter (CH-Adapter) that enhances fine-grained perception across diverse ultrasound imaging modalities while achieving parameter-efficient fine-tuning. To further improve clinical applicability, we develop a deep-supervised knowledge distillation (DSKD) technique that transfers knowledge from the large image encoder of the fine-tuned SAM2 to a super lightweight encoder, substantially reducing computational requirements without compromising performance. Extensive experiments demonstrate that UniUltra outperforms state-of-the-arts with superior generalization capabilities. Notably, our framework achieves competitive performance using only 8.91% of SAM2's parameters during fine-tuning, and the final compressed model reduces the parameter count by 94.08% compared to the original SAM2, making it highly suitable for practical clinical deployment. The source code is available at https://github.com/xq141839/UniUltra.

---

### TB or Not TB: Coverage-Driven Direct Preference Optimization for Verilog Stimulus Generation

Authors: Bardia Nadimi, Khashayar Filom, Deming Chen, Hao Zheng
Published: 2025-11-19
Categories: cs.LG, cs.AI, cs.PL
arXiv: https://arxiv.org/abs/2511.15767v1

With the rapid advancement of Large Language Models (LLMs), there is growing interest in applying them to hardware design and verification. Among these stages, design verification remains the most time-consuming and resource-intensive phase, where generating effective stimuli for the design under test (DUT) is both critical and labor-intensive. We present {\it TB or not TB}, a framework for automated stimulus generation using LLMs fine-tuned through Coverage-Driven Direct Preference Optimization (CD-DPO). To enable preference-based training, we introduce PairaNet, a dataset derived from PyraNet that pairs high- and low-quality testbenches labeled using simulation-derived coverage metrics. The proposed CD-DPO method integrates quantitative coverage feedback directly into the optimization objective, guiding the model toward generating stimuli that maximize verification coverage. Experiments on the CVDP CID12 benchmark show that {\it TB or not TB} outperforms both open-source and commercial baselines, achieving up to 77.27\% improvement in code coverage, demonstrating the effectiveness of Coverage-driven preference optimization for LLM-based hardware verification.

---

### Identifying the Supply Chain of AI for Trustworthiness and Risk Management in Critical Applications

Authors: Raymond K. Sheh, Karen Geappen
Published: 2025-11-19
Categories: cs.AI, cs.CR, cs.SE
arXiv: https://arxiv.org/abs/2511.15763v1

Risks associated with the use of AI, ranging from algorithmic bias to model hallucinations, have received much attention and extensive research across the AI community, from researchers to end-users. However, a gap exists in the systematic assessment of supply chain risks associated with the complex web of data sources, pre-trained models, agents, services, and other systems that contribute to the output of modern AI systems. This gap is particularly problematic when AI systems are used in critical applications, such as the food supply, healthcare, utilities, law, insurance, and transport. We survey the current state of AI risk assessment and management, with a focus on the supply chain of AI and risks relating to the behavior and outputs of the AI system. We then present a proposed taxonomy specifically for categorizing AI supply chain entities. This taxonomy helps stakeholders, especially those without extensive AI expertise, to "consider the right questions" and systematically inventory dependencies across their organization's AI systems. Our contribution bridges a gap between the current state of AI governance and the urgent need for actionable risk assessment and management of AI use in critical applications.

---

### A time for monsters: Organizational knowing after LLMs

Authors: Samer Faraj, Joel Perez Torrents, Saku Mantere, Anand Bhardwaj
Published: 2025-11-19
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2511.15762v1

Large Language Models (LLMs) are reshaping organizational knowing by unsettling the epistemological foundations of representational and practice-based perspectives. We conceptualize LLMs as Haraway-ian monsters, that is, hybrid, boundary-crossing entities that destabilize established categories while opening new possibilities for inquiry. Focusing on analogizing as a fundamental driver of knowledge, we examine how LLMs generate connections through large-scale statistical inference. Analyzing their operation across the dimensions of surface/deep analogies and near/far domains, we highlight both their capacity to expand organizational knowing and the epistemic risks they introduce. Building on this, we identify three challenges of living with such epistemic monsters: the transformation of inquiry, the growing need for dialogical vetting, and the redistribution of agency. By foregrounding the entangled dynamics of knowing-with-LLMs, the paper extends organizational theory beyond human-centered epistemologies and invites renewed attention to how knowledge is created, validated, and acted upon in the age of intelligent technologies.

---