# Research Papers — 2025-12 · bundle page 83 of 175

> Full-text markdown of papers #4101–4150 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/83
[← Previous bundle](https://aipod.dev/papers/archive/2025-12/page/82.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-12/page/84.md)

### Image Diffusion Preview with Consistency Solver

Authors: Fu-Yun Wang, Hao Zhou, Liangzhe Yuan, Sanghyun Woo, Boqing Gong, Bohyung Han, Ming-Hsuan Yang, Han Zhang, Yukun Zhu, Ting Liu, Long Zhao
Published: 2025-12-15
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2512.13592v2

The slow inference process of image diffusion models significantly degrades interactive user experiences. To address this, we introduce Diffusion Preview, a novel paradigm employing rapid, low-step sampling to generate preliminary outputs for user evaluation, deferring full-step refinement until the preview is deemed satisfactory. Existing acceleration methods, including training-free solvers and post-training distillation, struggle to deliver high-quality previews or ensure consistency between previews and final outputs. We propose ConsistencySolver derived from general linear multistep methods, a lightweight, trainable high-order solver optimized via Reinforcement Learning, that enhances preview quality and consistency. Experimental results demonstrate that ConsistencySolver significantly improves generation quality and consistency in low-step scenarios, making it ideal for efficient preview-and-refine workflows. Notably, it achieves FID scores on-par with Multistep DPM-Solver using 47% fewer steps, while outperforming distillation baselines. Furthermore, user studies indicate our approach reduces overall user interaction time by nearly 50% while maintaining generation quality. Code is available at https://github.com/G-U-N/consolver.

---

### ReFusion: A Diffusion Large Language Model with Parallel Autoregressive Decoding

Authors: Jia-Nan Li, Jian Guan, Wei Wu, Chongxuan Li
Published: 2025-12-15
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.13586v2

Autoregressive models (ARMs) are hindered by slow sequential inference. While masked diffusion models (MDMs) offer a parallel alternative, they suffer from critical drawbacks: high computational overhead from precluding Key-Value (KV) caching, and incoherent generation arising from learning dependencies over an intractable space of token combinations. To address these limitations, we introduce \textsc{ReFusion}, a novel masked diffusion model that integrates sequence reorganization into the causal attention framework. By elevating parallel decoding from the token level to a higher slot level, \textsc{ReFusion} interleaves inter-slot diffusion-based selection with intra-slot autoregressive infilling, while reordering newly generated slots ahead of the remaining masks after each iteration. Consequently, this design simultaneously unlocks full KV cache reuse and reduces learning complexity from an intractable token combination space to a manageable slot-level permutation space. Extensive experiments on seven diverse benchmarks show that \textsc{ReFusion} not only overwhelmingly surpasses prior MDMs with a 34\% performance gain and an over 18$\times$ speedup on average, but also bridges the performance gap to strong ARMs while maintaining a 2.33$\times$ average speedup.

---

### DP-CSGP: Differentially Private Stochastic Gradient Push with Compressed Communication

Authors: Zehan Zhu, Heng Zhao, Yan Huang, Joey Tianyi Zhou, Shouling Ji, Jinming Xu
Published: 2025-12-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13583v1

In this paper, we propose a Differentially Private Stochastic Gradient Push with Compressed communication (termed DP-CSGP) for decentralized learning over directed graphs. Different from existing works, the proposed algorithm is designed to maintain high model utility while ensuring both rigorous differential privacy (DP) guarantees and efficient communication. For general non-convex and smooth objective functions, we show that the proposed algorithm achieves a tight utility bound of $\mathcal{O}\left( \sqrt{d\log \left( \frac{1}δ \right)}/(\sqrt{n}Jε) \right)$ ($J$ and $d$ are the number of local samples and the dimension of decision variables, respectively) with $\left(ε, δ\right)$-DP guarantee for each node, matching that of decentralized counterparts with exact communication. Extensive experiments on benchmark tasks show that, under the same privacy budget, DP-CSGP achieves comparable model accuracy with significantly lower communication cost than existing decentralized counterparts with exact communication.

---

### Reproducing and Dissecting Denoising Language Models for Speech Recognition

Authors: Dorian Koch, Albert Zeyer, Nick Rossenbach, Ralf Schlüter, Hermann Ney
Published: 2025-12-15
Categories: cs.NE
arXiv: https://arxiv.org/abs/2512.13576v1

Denoising language models (DLMs) have been proposed as a powerful alternative to traditional language models (LMs) for automatic speech recognition (ASR), motivated by their ability to use bidirectional context and adapt to a specific ASR model's error patterns. However, the complexity of the DLM training pipeline has hindered wider investigation. This paper presents the first independent, large-scale empirical study of DLMs. We build and release a complete, reproducible pipeline to systematically investigate the impact of key design choices. We evaluate dozens of configurations across multiple axes, including various data augmentation techniques (e.g., SpecAugment, dropout, mixup), different text-to-speech systems, and multiple decoding strategies. Our comparative analysis in a common subword vocabulary setting demonstrates that DLMs outperform traditional LMs, but only after a distinct compute tipping point. While LMs are more efficient at lower budgets, DLMs scale better with longer training, mirroring behaviors observed in diffusion language models. However, we observe smaller improvements than those reported in prior character-based work, which indicates that the DLM's performance is conditional on factors such as the vocabulary. Our analysis reveals that a key factor for improving performance is to condition the DLM on richer information from the ASR's hypothesis space, rather than just a single best guess. To this end, we introduce DLM-sum, a novel method for decoding from multiple ASR hypotheses, which consistently outperforms the previously proposed DSR decoding method. We believe our findings and public pipeline provide a crucial foundation for the community to better understand, improve, and build upon this promising class of models. The code is publicly available at https://github.com/rwth-i6/2025-denoising-lm/.

---

### MMhops-R1: Multimodal Multi-hop Reasoning

Authors: Tao Zhang, Ziqi Zhang, Zongyang Ma, Yuxin Chen, Bing Li, Chunfeng Yuan, Guangting Wang, Fengyun Rao, Ying Shan, Weiming Hu
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13573v2

The ability to perform multi-modal multi-hop reasoning by iteratively integrating information across various modalities and external knowledge is critical for addressing complex real-world challenges. However, existing Multi-modal Large Language Models (MLLMs) are predominantly limited to single-step reasoning, as existing benchmarks lack the complexity needed to evaluate and drive multi-hop abilities. To bridge this gap, we introduce MMhops, a novel, large-scale benchmark designed to systematically evaluate and foster multi-modal multi-hop reasoning. MMhops dataset comprises two challenging task formats, Bridging and Comparison, which necessitate that models dynamically construct complex reasoning chains by integrating external knowledge. To tackle the challenges posed by MMhops, we propose MMhops-R1, a novel multi-modal Retrieval-Augmented Generation (mRAG) framework for dynamic reasoning. Our framework utilizes reinforcement learning to optimize the model for autonomously planning reasoning paths, formulating targeted queries, and synthesizing multi-level information. Comprehensive experiments demonstrate that MMhops-R1 significantly outperforms strong baselines on MMhops, highlighting that dynamic planning and multi-modal knowledge integration are crucial for complex reasoning. Moreover, MMhops-R1 demonstrates strong generalization to tasks requiring fixed-hop reasoning, underscoring the robustness of our dynamic planning approach. In conclusion, our work contributes a challenging new benchmark and a powerful baseline model, and we will release the associated code, data, and weights to catalyze future research in this critical area.

---

### Superposition as Lossy Compression: Measure with Sparse Autoencoders and Connect to Adversarial Vulnerability

Authors: Leonard Bereska, Zoe Tzifa-Kratira, Reza Samavi, Efstratios Gavves
Published: 2025-12-15
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2512.13568v1

Neural networks achieve remarkable performance through superposition: encoding multiple features as overlapping directions in activation space rather than dedicating individual neurons to each feature. This challenges interpretability, yet we lack principled methods to measure superposition. We present an information-theoretic framework measuring a neural representation's effective degrees of freedom. We apply Shannon entropy to sparse autoencoder activations to compute the number of effective features as the minimum neurons needed for interference-free encoding. Equivalently, this measures how many "virtual neurons" the network simulates through superposition. When networks encode more effective features than actual neurons, they must accept interference as the price of compression. Our metric strongly correlates with ground truth in toy models, detects minimal superposition in algorithmic tasks, and reveals systematic reduction under dropout. Layer-wise patterns mirror intrinsic dimensionality studies on Pythia-70M. The metric also captures developmental dynamics, detecting sharp feature consolidation during grokking. Surprisingly, adversarial training can increase effective features while improving robustness, contradicting the hypothesis that superposition causes vulnerability. Instead, the effect depends on task complexity and network capacity: simple tasks with ample capacity allow feature expansion (abundance regime), while complex tasks or limited capacity force reduction (scarcity regime). By defining superposition as lossy compression, this work enables principled measurement of how neural networks organize information under computational constraints, connecting superposition to adversarial robustness.

---

### A Nonparametric Statistics Approach to Feature Selection in Deep Neural Networks with Theoretical Guarantees

Authors: Junye Du, Zhenghao Li, Zhutong Gu, Long Feng
Published: 2025-12-15
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2512.13565v1

This paper tackles the problem of feature selection in a highly challenging setting: $\mathbb{E}(y | \boldsymbol{x}) = G(\boldsymbol{x}_{\mathcal{S}_0})$, where $\mathcal{S}_0$ is the set of relevant features and $G$ is an unknown, potentially nonlinear function subject to mild smoothness conditions. Our approach begins with feature selection in deep neural networks, then generalizes the results to H{ö}lder smooth functions by exploiting the strong approximation capabilities of neural networks. Unlike conventional optimization-based deep learning methods, we reformulate neural networks as index models and estimate $\mathcal{S}_0$ using the second-order Stein's formula. This gradient-descent-free strategy guarantees feature selection consistency with a sample size requirement of $n = Ω(p^2)$, where $p$ is the feature dimension. To handle high-dimensional scenarios, we further introduce a screening-and-selection mechanism that achieves nonlinear selection consistency when $n = Ω(s \log p)$, with $s$ representing the sparsity level. Additionally, we refit a neural network on the selected features for prediction and establish performance guarantees under a relaxed sparsity assumption. Extensive simulations and real-data analyses demonstrate the strong performance of our method even in the presence of complex feature interactions.

---

### Memory in the Age of AI Agents

Authors: Yuyang Hu, Shichun Liu, Yanwei Yue, Guibin Zhang, Boyang Liu, Fangyi Zhu, Jiahang Lin, Honglin Guo, Shihan Dou, Zhiheng Xi, Senjie Jin, Jiejun Tan et al.
Published: 2025-12-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2512.13564v2

Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of contemporary agent memory systems. This work aims to provide an up-to-date landscape of current agent memory research. We begin by clearly delineating the scope of agent memory and distinguishing it from related concepts such as LLM memory, retrieval augmented generation (RAG), and context engineering. We then examine agent memory through the unified lenses of forms, functions, and dynamics. From the perspective of forms, we identify three dominant realizations of agent memory, namely token-level, parametric, and latent memory. From the perspective of functions, we propose a finer-grained taxonomy that distinguishes factual, experiential, and working memory. From the perspective of dynamics, we analyze how memory is formed, evolved, and retrieved over time. To support practical development, we compile a comprehensive summary of memory benchmarks and open-source frameworks. Beyond consolidation, we articulate a forward-looking perspective on emerging research frontiers, including memory automation, reinforcement learning integration, multimodal memory, multi-agent memory, and trustworthiness issues. We hope this survey serves not only as a reference for existing work, but also as a conceptual foundation for rethinking memory as a first-class primitive in the design of future agentic intelligence.

---

### 3D Human-Human Interaction Anomaly Detection

Authors: Shun Maeda, Chunzhi Gu, Koichiro Kamide, Katsuya Hotta, Shangce Gao, Chao Zhang
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13560v1

Human-centric anomaly detection (AD) has been primarily studied to specify anomalous behaviors in a single person. However, as humans by nature tend to act in a collaborative manner, behavioral anomalies can also arise from human-human interactions. Detecting such anomalies using existing single-person AD models is prone to low accuracy, as these approaches are typically not designed to capture the complex and asymmetric dynamics of interactions. In this paper, we introduce a novel task, Human-Human Interaction Anomaly Detection (H2IAD), which aims to identify anomalous interactive behaviors within collaborative 3D human actions. To address H2IAD, we then propose Interaction Anomaly Detection Network (IADNet), which is formalized with a Temporal Attention Sharing Module (TASM). Specifically, in designing TASM, we share the encoded motion embeddings across both people such that collaborative motion correlations can be effectively synchronized. Moreover, we notice that in addition to temporal dynamics, human interactions are also characterized by spatial configurations between two people. We thus introduce a Distance-Based Relational Encoding Module (DREM) to better reflect social cues in H2IAD. The normalizing flow is eventually employed for anomaly scoring. Extensive experiments on human-human motion benchmarks demonstrate that IADNet outperforms existing Human-centric AD baselines in H2IAD.

---

### Verifying Rumors via Stance-Aware Structural Modeling

Authors: Gibson Nkhata, Uttamasha Anjally Oyshi, Quan Mai, Susan Gauch
Published: 2025-12-15
Categories: cs.CL, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2512.13559v1

Verifying rumors on social media is critical for mitigating the spread of false information. The stances of conversation replies often provide important cues to determine a rumor's veracity. However, existing models struggle to jointly capture semantic content, stance information, and conversation strructure, especially under the sequence length constraints of transformer-based encoders. In this work, we propose a stance-aware structural modeling that encodes each post in a discourse with its stance signal and aggregates reply embedddings by stance category enabling a scalable and semantically enriched representation of the entire thread. To enhance structural awareness, we introduce stance distribution and hierarchical depth as covariates, capturing stance imbalance and the influence of reply depth. Extensive experiments on benchmark datasets demonstrate that our approach significantly outperforms prior methods in the ability to predict truthfulness of a rumor. We also demonstrate that our model is versatile for early detection and cross-platfrom generalization.

---

### PrahokBART: A Pre-trained Sequence-to-Sequence Model for Khmer Natural Language Generation

Authors: Hour Kaing, Raj Dabre, Haiyue Song, Van-Hien Tran, Hideki Tanaka, Masao Utiyama
Published: 2025-12-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2512.13552v1

This work introduces {\it PrahokBART}, a compact pre-trained sequence-to-sequence model trained from scratch for Khmer using carefully curated Khmer and English corpora. We focus on improving the pre-training corpus quality and addressing the linguistic issues of Khmer, which are ignored in existing multilingual models, by incorporating linguistic components such as word segmentation and normalization. We evaluate PrahokBART on three generative tasks: machine translation, text summarization, and headline generation, where our results demonstrate that it outperforms mBART50, a strong multilingual pre-trained model. Additionally, our analysis provides insights into the impact of each linguistic module and evaluates how effectively our model handles space during text generation, which is crucial for the naturalness of texts in Khmer.

---

### A Class of Accelerated Fixed-Point-Based Methods with Delayed Inexact Oracles and Its Applications

Authors: Nghia Nguyen-Trung, Quoc Tran-Dinh
Published: 2025-12-15
Categories: math.OC, stat.ML
arXiv: https://arxiv.org/abs/2512.13547v1

In this paper, we develop a novel accelerated fixed-point-based framework using delayed inexact oracles to approximate a fixed point of a nonexpansive operator (or equivalently, a root of a co-coercive operator), a central problem in scientific computing. Our approach leverages both Nesterov's acceleration technique and the Krasnosel'skii-Mann (KM) iteration, while accounting for delayed inexact oracles, a key mechanism in asynchronous algorithms. We also introduce a unified approximate error condition for delayed inexact oracles, which can cover various practical scenarios. Under mild conditions and appropriate parameter updates, we establish both $\mathcal{O}(1/k^2)$ non-asymptotic and $o(1/k^2)$ asymptotic convergence rates in expectation for the squared norm of residual. Our rate significantly improves the $\mathcal{O}(1/k)$ rates in classical KM-type methods, including their asynchronous variants. We also establish $o(1/k^2)$ almost sure convergence rates and the almost sure convergence of iterates to a solution of the problem. Within our framework, we instantiate three settings for the underlying operator: (i) a deterministic universal delayed oracle; (ii) a stochastic delayed oracle; and (iii) a finite-sum structure with asynchronous updates. For each case, we instantiate our framework to obtain a concrete algorithmic variant for which our convergence results still apply, and whose iteration complexity depends linearly on the maximum delay. Finally, we verify our algorithms and theoretical results through two numerical examples on both matrix game and shallow neural network training problems.

---

### Pancakes: Consistent Multi-Protocol Image Segmentation Across Biomedical Domains

Authors: Marianne Rakic, Siyu Gai, Etienne Chollet, John V. Guttag, Adrian V. Dalca
Published: 2025-12-15
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2512.13534v1

A single biomedical image can be meaningfully segmented in multiple ways, depending on the desired application. For instance, a brain MRI can be segmented according to tissue types, vascular territories, broad anatomical regions, fine-grained anatomy, or pathology, etc. Existing automatic segmentation models typically either (1) support only a single protocol, the one they were trained on, or (2) require labor-intensive manual prompting to specify the desired segmentation. We introduce Pancakes, a framework that, given a new image from a previously unseen domain, automatically generates multi-label segmentation maps for multiple plausible protocols, while maintaining semantic consistency across related images. Pancakes introduces a new problem formulation that is not currently attainable by existing foundation models. In a series of experiments on seven held-out datasets, we demonstrate that our model can significantly outperform existing foundation models in producing several plausible whole-image segmentations, that are semantically coherent across images.

---

### Adaptive Sampling for Hydrodynamic Stability

Authors: Anshima Singh, David J. Silvester
Published: 2025-12-15
Categories: physics.flu-dyn, cs.LG
arXiv: https://arxiv.org/abs/2512.13532v2

An adaptive sampling approach for efficient detection of bifurcation boundaries in parametrized fluid flow problems is presented herein. The study extends the machine-learning approach of Silvester~(J. Comput. Phys., 553 (2026), 114743), where a classifier network was trained on preselected simulation data to identify bifurcated and nonbifurcated flow regimes. In contrast, the proposed methodology introduces adaptivity through a flow-based deep generative model that automatically refines the sampling of the parameter space. The strategy has two components: a classifier network maps the flow parameters to a bifurcation probability, and a probability density estimation technique (KRnet) for the generation of new samples at each adaptive step. The classifier output provides a probabilistic measure of flow stability, and the Shannon entropy of these predictions is employed as an uncertainty indicator. KRnet is trained to approximate a probability density function that concentrates sampling in regions of high entropy, thereby directing computational effort towards the evolving bifurcation boundary. This coupling between classification and generative modeling establishes a feedback-driven adaptive learning process analogous to error-indicator based refinement in contemporary partial differential equation solution strategies. Starting from a uniform parameter distribution, the new approach achieves accurate bifurcation boundary identification with significantly fewer Navier--Stokes simulations, providing a scalable foundation for high-dimensional stability analysis.

---

### Actively Learning Joint Contours of Multiple Computer Experiments

Authors: Shih-Ni Prim, Kevin R. Quinlan, Paul Hawkins, Jagadeesh Movva, Annie S. Booth
Published: 2025-12-15
Categories: stat.ME, cs.LG
arXiv: https://arxiv.org/abs/2512.13530v2

Contour location---the process of sequentially training a surrogate model to identify the design inputs that result in a pre-specified response value from a single computer experiment---is a well-studied active learning problem. Here, we tackle a related but distinct problem: identifying the input configuration that returns pre-specified values of multiple computer experiments simultaneously. Motivated by computer experiments of the rotational torques acting upon a vehicle in flight, we aim to identify stable flight conditions that result in zero torque forces. We propose a ``joint contour location'' (jCL) scheme that strikes a strategic balance between exploring the multiple response surfaces while exploiting learning of the intersecting contours. Rather than working exploration and exploitation into a single acquisition function, we devise two distinct acquisition schemes with a decision rule to choose between the two, which also provides a natural stopping criterion if no solution is present. We employ traditional Gaussian processes (GPs), multitask GPs, and deep GPs, but our jCL procedure is applicable to any surrogate that can provide posterior predictive distributions. Our jCL designs significantly outperform existing (single response) CL strategies, optimization-based alternatives, and previous strategies for targeting joint contours, enabling us to efficiently locate the optimal configurations for our motivating computer experiments.

---

### Enhancing lithological interpretation from petrophysical well log of IODP expedition 390/393 using machine learning

Authors: Raj Sahu, Saumen Maiti
Published: 2025-12-15
Categories: physics.geo-ph, cs.LG
arXiv: https://arxiv.org/abs/2512.13529v1

Enhanced lithological interpretation from well logs plays a key role in geological resource exploration and mapping, as well as in geo-environmental modeling studies. Core and cutting information is useful for making sound interpretations of well logs; however, these are rarely collected at each depth due to high costs. Moreover, well log interpretation using traditional methods is constrained by poor borehole conditions. Traditional statistical methods are mostly linear, often failing to discriminate between lithology and rock facies, particularly when dealing with overlapping well log signals characterized by the structural and compositional variation of rock types. In this study, we develop multiple supervised and unsupervised machine learning algorithms to jointly analyze multivariate well log data from Integrated Ocean Drilling Program (IODP) expeditions 390 and 393 for enhanced lithological interpretations. Among the algorithms, Logistic Regression, Decision Trees, Gradient Boosting, Support Vector Machines (SVM), k-Nearest Neighbors (KNN), and Multi-Layer Perceptron (MLP) neural network models, the Decision Tree and Gradient Boosting models outperformed the others, achieving an accuracy of 0.9950 and an F1-score of 0.9951. While unsupervised machine learning (ML) provides the foundation for cluster information that inherently supports the classification algorithm, supervised ML is applied to devise a data-driven lithology clustering mechanism for IODP datasets. The joint ML-based method developed here has the potential to be further explored for analyzing other well log datasets from the world's oceans.

---

### Async Control: Stress-testing Asynchronous Control Measures for LLM Agents

Authors: Asa Cooper Stickland, Jan Michelfeit, Arathi Mani, Charlie Griffin, Ollie Matthews, Tomek Korbak, Rogan Inglis, Oliver Makins, Alan Cooney
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.13526v1

LLM-based software engineering agents are increasingly used in real-world development tasks, often with access to sensitive data or security-critical codebases. Such agents could intentionally sabotage these codebases if they were misaligned. We investigate asynchronous monitoring, in which a monitoring system reviews agent actions after the fact. Unlike synchronous monitoring, this approach does not impose runtime latency, while still attempting to disrupt attacks before irreversible harm occurs. We treat monitor development as an adversarial game between a blue team (who design monitors) and a red team (who create sabotaging agents). We attempt to set the game rules such that they upper bound the sabotage potential of an agent based on Claude 4.1 Opus. To ground this game in a realistic, high-stakes deployment scenario, we develop a suite of 5 diverse software engineering environments that simulate tasks that an agent might perform within an AI developer's internal infrastructure. Over the course of the game, we develop an ensemble monitor that achieves a 6% false negative rate at 1% false positive rate on a held out test environment. Then, we estimate risk of sabotage at deployment time by extrapolating from our monitor's false negative rate. We describe one simple model for this extrapolation, present a sensitivity analysis, and describe situations in which the model would be invalid. Code is available at: https://github.com/UKGovernmentBEIS/async-control.

---

### A Deep Learning Model of Mental Rotation Informed by Interactive VR Experiments

Authors: Raymond Khazoum, Daniela Fernandes, Aleksandr Krylov, Qin Li, Stephane Deny
Published: 2025-12-15
Categories: q-bio.NC, cs.LG
arXiv: https://arxiv.org/abs/2512.13517v2

Mental rotation -- the ability to compare objects seen from different viewpoints -- is a fundamental example of mental simulation and spatial world modeling in humans. Here we propose a mechanistic model of human mental rotation, leveraging recent advances in deep, equivariant, and neuro-symbolic learning. Our model consists of three stacked components: (1) an equivariant neural encoder, producing 3D spatial representations of objects from images, (2) a neuro-symbolic object encoder, deriving symbolic objects descriptions from these spatial representations, and (3) a neural decision agent, comparing these symbolic descriptions to prescribe rotation simulations in 3D latent space via a recurrent pathway. Our model design is guided by the existing experimental literature on mental rotation, which we complemented with experiments in VR where participants could at times manipulate the objects to compare. Our model captures well the performance, response times and behavior of participants in our and others' experiments, and through ablation studies we demonstrate the necessity of each component. Our work adds to a recent collection of deep neural models of human spatial reasoning, further demonstrating the potency of integrating deep, equivariant, and symbolic representations to model the human mind.

---

### Fine-tuned LLM-based Code Migration Framework

Authors: Oleg Grynets, Vasyl Lyashkevych, Dmytro Baran, Maksym Orliansky, Taras Zelenyy, Markiian Leshchyshyn
Published: 2025-12-15
Categories: cs.SE, cs.CL, cs.LO
arXiv: https://arxiv.org/abs/2512.13515v3

The study presents the outcomes of research and experimental validation in the domain of automated codebase migration, with a focus on addressing challenges in transitioning SQL-based systems. The proposed method for migration essentially appears as a framework that leverages the best aspects of traditional software engineering techniques and provides an iterative, scalable, precise and efficient solution for modern database transformations. The central piece of the approach is the integration of a fine-tuned Large Language Model to address critical issues in SQL code conversion, such as syntax mapping, resolving discrepancies between Oracle PL/SQL and PostgreSQL, and optimising database elements such as stored procedures, triggers, views, and overall database logic. Thus, the method involves a trade-off between fine-tuning and prompt engineering. Special attention is given to a fine-tuning approach, which enhances the adaptability and compatibility with migration requirements across the entire database. According to the achieved results, fine-tuning plays a very important role. The study employs targeted evaluation methodologies along with computational metrics to measure the success of iterative conversion cycles. Core innovations include automated SQL feature detection, semi-supervised error analysis and integration of Subject Matter Experts feedback within a systematic migration workflow. The methodology achieves significant reductions in Syntax Error Rates, enhances feature alignment throughout migration iterations, and leverages dataset sampling to ensure continual improvement. By embedding GAI into the migration process, the framework facilitates precise feature mapping, semi-automated error resolution, and data-driven optimisation loops, improving workflow efficiency.

---

### Adapting MLLMs for Nuanced Video Retrieval

Authors: Piyush Bagad, Andrew Zisserman
Published: 2025-12-15
Categories: cs.CV, cs.IR
arXiv: https://arxiv.org/abs/2512.13511v4

Our objective is to build an embedding model that captures the nuanced relationship between a search query and candidate videos. We cover three aspects of nuanced retrieval: (i) temporal, (ii) negation, and (iii) multimodal. For temporal nuance, we consider chiral actions that need distinguishing between temporally opposite actions like "opening a door" vs. "closing a door". For negation, we consider queries with negators such as "not", "none" that allow a user to specify what they do not want. For multimodal nuance, we consider the task of composed retrieval where the query comprises a video along with a text edit instruction. The goal is to develop a unified embedding model that handles such nuances effectively. To that end, we repurpose a Multimodal Large Language Model (MLLM) trained to generate text into an embedding model. We fine-tune it with a contrastive loss on text alone with carefully sampled hard negatives that instill the desired nuances in the learned embedding space. Despite the text-only training, our method achieves state of the art performance on all benchmarks for nuanced video retrieval. We also analyze how this improvement is achieved, and show that text-only training reduces the modality gap between text and video embeddings leading to better organization of the embedding space.

---

### MedCEG: Reinforcing Verifiable Medical Reasoning with Critical Evidence Graph

Authors: Linjie Mu, Yannian Gu, Zhongzhen Huang, Yakun Zhu, Shaoting Zhang, Xiaofan Zhang
Published: 2025-12-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2512.13510v1

Large language models with reasoning capabilities have demonstrated impressive performance across a wide range of domains. In clinical applications, a transparent, step-by-step reasoning process provides physicians with strong evidence to support decision-making. While reinforcement learning has effectively enhanced reasoning performance in medical contexts, the clinical reliability of these reasoning processes remains limited because their accuracy and validity are often overlooked during training. To address this gap, we propose MedCEG, a framework that augments medical language models with clinically valid reasoning pathways by explicitly supervising the reasoning process through a Critical Evidence Graph (CEG). We curate a dataset of challenging clinical cases and algorithmically construct a CEG for each sample to represent a high-quality verifiable reasoning pathway. To guide the reasoning process, we introduce a Clinical Reasoning Procedure Reward, which evaluates Node Coverage, Structural Correctness, and Chain Completeness, thereby providing a holistic assessment of reasoning quality. Experimental results show that MedCEG surpasses existing methods in performance while producing clinically valid reasoning chains, representing a solid advancement in reliable medical AI reasoning. The code and models are available at https://github.com/LinjieMu/MedCEG.

---

### Seedance 1.5 pro: A Native Audio-Visual Joint Generation Foundation Model

Authors: Team Seedance, Heyi Chen, Siyan Chen, Xin Chen, Yanfei Chen, Ying Chen, Zhuo Chen, Feng Cheng, Tianheng Cheng, Xinqi Cheng, Xuyan Chi, Jian Cong et al.
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13507v3

Recent strides in video generation have paved the way for unified audio-visual generation. In this work, we present Seedance 1.5 pro, a foundational model engineered specifically for native, joint audio-video generation. Leveraging a dual-branch Diffusion Transformer architecture, the model integrates a cross-modal joint module with a specialized multi-stage data pipeline, achieving exceptional audio-visual synchronization and superior generation quality. To ensure practical utility, we implement meticulous post-training optimizations, including Supervised Fine-Tuning (SFT) on high-quality datasets and Reinforcement Learning from Human Feedback (RLHF) with multi-dimensional reward models. Furthermore, we introduce an acceleration framework that boosts inference speed by over 10X. Seedance 1.5 pro distinguishes itself through precise multilingual and dialect lip-syncing, dynamic cinematic camera control, and enhanced narrative coherence, positioning it as a robust engine for professional-grade content creation. Seedance 1.5 pro is now accessible on Volcano Engine at https://console.volcengine.com/ark/region:ark+cn-beijing/experience/vision?type=GenVideo.

---

### Learning under Distributional Drift: Prequential Reproducibility as an Intrinsic Statistical Resource

Authors: Sofiya Zaichyk
Published: 2025-12-15
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2512.13506v4

Statistical learning under distributional drift remains poorly characterized, especially in closed-loop settings where learning alters the data-generating law. We introduce an intrinsic drift budget $C_T$ that quantifies cumulative information-geometric motion of the data distribution along the realized learner-environment trajectory, measured in Fisher-Rao distance. The budget separates exogenous environmental change from policy-sensitive feedback induced by the learner's actions. This gives a rate-based characterization of prequential reproducibility: when performance on the realized stream is used to predict one-step-ahead performance under the next distribution, the drift contribution enters through the average motion rate $C_T/T$, not through cumulative drift alone. We prove a drift-feedback bound of order $T^{-1/2}+C_T/T$, up to controlled second-order remainder terms, and establish a matching sharpness lower bound for the same prequential reproducibility gap on a canonical regular subclass. Thus the dependence on the average Fisher-Rao motion rate is tight up to constants: $C_T/T$ is sufficient for upper control and unavoidable on regular hard subclasses. We further prove an information-theoretic indistinguishability result showing that order-$C/T$ effects on the one-step-ahead target need not be identifiable from the realized performance stream alone. Finally, we show that fixed monitoring channels induce contracted observable Fisher motion, and experiments, including a misspecified real-data feedback setting, indicate that appropriately chosen channels can retain risk-relevant drift signal when the intrinsic data-generating law is unavailable. The resulting theory treats exogenous drift, adaptive data analysis, and performative feedback as different sources of Fisher-Rao motion along the same learner-environment trajectory.

---

### Defending the Hierarchical Result Models of Precedential Constraint

Authors: Henry Prakken, Wijnand van Woerkom
Published: 2025-12-15
Categories: cs.AI
arXiv: https://arxiv.org/abs/2512.13505v1

In recent years, hierarchical case-based-reasoning models of precedential constraint have been proposed. In various papers, Trevor Bench-Capon criticised these models on the grounds that they would give incorrect outcomes in some cases. In particular, the models would not account for the possibility that intermediate factors are established with different strengths by different base-level factors. In this paper we respond to these criticisms for van Woerkom's result-based hierarchical models. We argue that in some examples Bench-Capon seems to interpret intermediate factors as dimensions, and that applying van Woerkom's dimension-based version of the hierarchical result model to these examples avoids Bench-Capon's criticisms.

---

### Behavior-Aware and Generalizable Defense Against Black-Box Adversarial Attacks for ML-Based IDS

Authors: Sabrine Ennaji, Elhadj Benkhelifa, Luigi Vincenzo Mancini
Published: 2025-12-15
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2512.13501v1

Machine learning based intrusion detection systems are increasingly targeted by black box adversarial attacks, where attackers craft evasive inputs using indirect feedback such as binary outputs or behavioral signals like response time and resource usage. While several defenses have been proposed, including input transformation, adversarial training, and surrogate detection, they often fall short in practice. Most are tailored to specific attack types, require internal model access, or rely on static mechanisms that fail to generalize across evolving attack strategies. Furthermore, defenses such as input transformation can degrade intrusion detection system performance, making them unsuitable for real time deployment. To address these limitations, we propose Adaptive Feature Poisoning, a lightweight and proactive defense mechanism designed specifically for realistic black box scenarios. Adaptive Feature Poisoning assumes that probing can occur silently and continuously, and introduces dynamic and context aware perturbations to selected traffic features, corrupting the attacker feedback loop without impacting detection capabilities. The method leverages traffic profiling, change point detection, and adaptive scaling to selectively perturb features that an attacker is likely exploiting, based on observed deviations. We evaluate Adaptive Feature Poisoning against multiple realistic adversarial attack strategies, including silent probing, transferability based attacks, and decision boundary based attacks. The results demonstrate its ability to confuse attackers, degrade attack effectiveness, and preserve detection performance. By offering a generalizable, attack agnostic, and undetectable defense, Adaptive Feature Poisoning represents a significant step toward practical and robust adversarial resilience in machine learning based intrusion detection systems.

---

### On-Device Continual Learning for Unsupervised Visual Anomaly Detection in Dynamic Manufacturing

Authors: Haoyu Ren, Kay Koehle, Kirill Dorofeev, Darko Anicic
Published: 2025-12-15
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2512.13497v1

In modern manufacturing, Visual Anomaly Detection (VAD) is essential for automated inspection and consistent product quality. Yet, increasingly dynamic and flexible production environments introduce key challenges: First, frequent product changes in small-batch and on-demand manufacturing require rapid model updates. Second, legacy edge hardware lacks the resources to train and run large AI models. Finally, both anomalous and normal training data are often scarce, particularly for newly introduced product variations. We investigate on-device continual learning for unsupervised VAD with localization, extending the PatchCore to incorporate online learning for real-world industrial scenarios. The proposed method leverages a lightweight feature extractor and an incremental coreset update mechanism based on k-center selection, enabling rapid, memory-efficient adaptation from limited data while eliminating costly cloud retraining. Evaluations on an industrial use case are conducted using a testbed designed to emulate flexible production with frequent variant changes in a controlled environment. Our method achieves a 12% AUROC improvement over the baseline, an 80% reduction in memory usage, and faster training compared to batch retraining. These results confirm that our method delivers accurate, resource-efficient, and adaptive VAD suitable for dynamic and smart manufacturing.

---

### Soul: Breathe Life into Digital Human for High-fidelity Long-term Multimodal Animation

Authors: Jiangning Zhang, Junwei Zhu, Zhenye Gan, Donghao Luo, Chuming Lin, Feifan Xu, Xu Peng, Jianlong Hu, Yuansen Liu, Yijia Hong, Weijian Cao, Han Feng et al.
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13495v1

We propose a multimodal-driven framework for high-fidelity long-term digital human animation termed $\textbf{Soul}$, which generates semantically coherent videos from a single-frame portrait image, text prompts, and audio, achieving precise lip synchronization, vivid facial expressions, and robust identity preservation. We construct Soul-1M, containing 1 million finely annotated samples with a precise automated annotation pipeline (covering portrait, upper-body, full-body, and multi-person scenes) to mitigate data scarcity, and we carefully curate Soul-Bench for comprehensive and fair evaluation of audio-/text-guided animation methods. The model is built on the Wan2.2-5B backbone, integrating audio-injection layers and multiple training strategies together with threshold-aware codebook replacement to ensure long-term generation consistency. Meanwhile, step/CFG distillation and a lightweight VAE are used to optimize inference efficiency, achieving an 11.4$\times$ speedup with negligible quality loss. Extensive experiments show that Soul significantly outperforms current leading open-source and commercial models on video quality, video-text alignment, identity preservation, and lip-synchronization accuracy, demonstrating broad applicability in real-world scenarios such as virtual anchors and film production. Project page at https://zhangzjn.github.io/projects/Soul/

---

### SkipCat: Rank-Maximized Low-Rank Compression of Large Language Models via Shared Projection and Block Skipping

Authors: Yu-Chen Lu, Sheng-Feng Yu, Hui-Hsien Weng, Pei-Shuo Wang, Yu-Fang Hu, Liang Hung-Chun, Hung-Yueh Chiang, Kai-Chiang Wu
Published: 2025-12-15
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2512.13494v1

Large language models (LLM) have achieved remarkable performance across a wide range of tasks. However, their substantial parameter sizes pose significant challenges for deployment on edge devices with limited computational and memory resources. Low-rank compression is a promising approach to address this issue, as it reduces both computational and memory costs, making LLM more suitable for resource-constrained environments. Nonetheless, naïve low-rank compression methods require a significant reduction in the retained rank to achieve meaningful memory and computation savings. For a low-rank model, the ranks need to be reduced by more than half to yield efficiency gains. Such aggressive truncation, however, typically results in substantial performance degradation. To address this trade-off, we propose SkipCat, a novel low-rank compression framework that enables the use of higher ranks while achieving the same compression rates. First, we introduce an intra-layer shared low-rank projection method, where multiple matrices that share the same input use a common projection. This reduces redundancy and improves compression efficiency. Second, we propose a block skipping technique that omits computations and memory transfers for selected sub-blocks within the low-rank decomposition. These two techniques jointly enable our compressed model to retain more effective ranks under the same compression budget. Experimental results show that, without any additional fine-tuning, our method outperforms previous low-rank compression approaches by 7% accuracy improvement on zero-shot tasks under the same compression rate. These results highlight the effectiveness of our rank-maximized compression strategy in preserving model performance under tight resource constraints.

---

### Transform Trained Transformer for Accelerating Native 4K Video Generation

Authors: Jiangning Zhang, Junwei Zhu, Teng Hu, Yabiao Wang, Donghao Luo, Weijian Cao, Zhenye Gan, Xiaobin Hu, Zhucun Xue, Xiangtai Li, Chengjie Wang, Yong Liu
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13492v2

Native 4K (2176$\times$3840) video generation remains a critical challenge due to the quadratic computational explosion of full-attention as spatiotemporal resolution increases, making it difficult for models to strike a balance between efficiency and quality. This paper proposes a novel Transformer retrofit strategy termed T3 ($\textbf{T}$ransform $\textbf{T}$rained $\textbf{T}$ransformer) that, without altering the core architecture of full-attention pretrained models, significantly reduces compute requirements by optimizing their forward logic. Specifically, $\textbf{T3-Video}$ introduces a multi-scale weight-sharing window attention mechanism and, via hierarchical blocking together with an axis-preserving full-attention design, can effect an "attention pattern" transformation of a pretrained model using only modest compute and data. Results on $\textbf{4K-VBench}$ show that $\textbf{T3-Video}$ substantially outperforms existing approaches: while delivering performance improvements (+4.29$\uparrow$ VQA and +0.08$\uparrow$ VTC), it accelerates native 4K video generation by more than 10$\times$. Project page at https://zhangzjn.github.io/projects/T3-Video

---

### From Zipf's Law to Neural Scaling through Heaps' Law and Hilberg's Hypothesis

Authors: Łukasz Dębowski
Published: 2025-12-15
Categories: cs.IT, cs.LG, math.ST
arXiv: https://arxiv.org/abs/2512.13491v3

We inspect the deductive connection between the neural scaling law and Zipf's law -- two statements discussed in machine learning and quantitative linguistics. The neural scaling law describes how the cross entropy rate of a foundation model -- such as a large language model -- changes with respect to the amount of training tokens, parameters, and compute. By contrast, Zipf's law posits that the distribution of tokens exhibits a power law tail. Whereas similar claims have been made in more specific settings, we show that the neural scaling law is a consequence of Zipf's law under certain broad assumptions that we reveal systematically. The derivation steps are as follows: We derive Heaps' law on the vocabulary growth from Zipf's law, Hilberg's hypothesis on the entropy scaling from Heaps' law, and the neural scaling from Hilberg's hypothesis. We illustrate these inference steps by a toy example of the Santa Fe process that satisfies all four statistical laws.

---

### SIGMA: An AI-Empowered Training Stack on Early-Life Hardware

Authors: Lei Qu, Lianhai Ren, Peng Cheng, Rui Gao, Ruizhe Wang, Tianyu Chen, Xiao Liu, Xingjian Zhang, Yeyun Gong, Yifan Xiong, Yucheng Ding, Yuting Jiang et al.
Published: 2025-12-15
Categories: cs.DC, cs.CL
arXiv: https://arxiv.org/abs/2512.13488v1

An increasing variety of AI accelerators is being considered for large-scale training. However, enabling large-scale training on early-life AI accelerators faces three core challenges: frequent system disruptions and undefined failure modes that undermine reliability; numerical errors and training instabilities that threaten correctness and convergence; and the complexity of parallelism optimization combined with unpredictable local noise that degrades efficiency. To address these challenges, SIGMA is an open-source training stack designed to improve the reliability, stability, and efficiency of large-scale distributed training on early-life AI hardware. The core of this initiative is the LUCIA TRAINING PLATFORM (LTP), the system optimized for clusters with early-life AI accelerators. Since its launch in March 2025, LTP has significantly enhanced training reliability and operational productivity. Over the past five months, it has achieved an impressive 94.45% effective cluster accelerator utilization, while also substantially reducing node recycling and job-recovery times. Building on the foundation of LTP, the LUCIA TRAINING FRAMEWORK (LTF) successfully trained SIGMA-MOE, a 200B MoE model, using 2,048 AI accelerators. This effort delivered remarkable stability and efficiency outcomes, achieving 21.08% MFU, state-of-the-art downstream accuracy, and encountering only one stability incident over a 75-day period. Together, these advances establish SIGMA, which not only tackles the critical challenges of large-scale training but also establishes a new benchmark for AI infrastructure and platform innovation, offering a robust, cost-effective alternative to prevailing established accelerator stacks and significantly advancing AI capabilities and scalability. The source code of SIGMA is available at https://github.com/microsoft/LuciaTrainingPlatform.

---

### Advancing Bangla Machine Translation Through Informal Datasets

Authors: Ayon Roy, Risat Rahaman, Sadat Shibly, Udoy Saha Joy, Abdulla Al Kafi, Farig Yousuf Sadeque
Published: 2025-12-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2512.13487v1

Bangla is the sixth most widely spoken language globally, with approximately 234 million native speakers. However, progress in open-source Bangla machine translation remains limited. Most online resources are in English and often remain untranslated into Bangla, excluding millions from accessing essential information. Existing research in Bangla translation primarily focuses on formal language, neglecting the more commonly used informal language. This is largely due to the lack of pairwise Bangla-English data and advanced translation models. If datasets and models can be enhanced to better handle natural, informal Bangla, millions of people will benefit from improved online information access. In this research, we explore current state-of-the-art models and propose improvements to Bangla translation by developing a dataset from informal sources like social media and conversational texts. This work aims to advance Bangla machine translation by focusing on informal language translation and improving accessibility for Bangla speakers in the digital world.

---

### Real-Time AI-Driven Milling Digital Twin Towards Extreme Low-Latency

Authors: Wenyi Liu, R. Sharma, W. "Grace" Guo, J. Yi, Y. B. Guo
Published: 2025-12-15
Categories: eess.SY, cs.LG
arXiv: https://arxiv.org/abs/2512.13482v1

Digital twin (DT) enables smart manufacturing by leveraging real-time data, AI models, and intelligent control systems. This paper presents a state-of-the-art analysis on the emerging field of DTs in the context of milling. The critical aspects of DT are explored through the lens of virtual models of physical milling, data flow from physical milling to virtual model, and feedback from virtual model to physical milling. Live data streaming protocols and virtual modeling methods are highlighted. A case study showcases the transformative capability of a real-time machine learning-driven live DT of tool-work contact in a milling process. Future research directions are outlined to achieve the goals of Industry 4.0 and beyond.

---

### neuralFOMO: Can LLMs Handle Being Second Best? Measuring Envy-Like Preferences in Multi-Agent Settings

Authors: Arnav Ramamoorthy, Shrey Dhorajiya, Ojas Pungalia, Rashi Upadhyay, Abhishek Mishra, Abhiram H, Tejasvi Alladi, Sujan Yenuganti, Dhruv Kumar
Published: 2025-12-15
Categories: cs.AI, cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2512.13481v2

Envy shapes competitiveness and cooperation in human groups, yet its role in large language model interactions remains largely unexplored. As LLMs increasingly operate in multi-agent settings, it is important to examine whether they exhibit envy-like preferences under social comparison. We evaluate LLM behavior across two scenarios: (1) a point-allocation game testing sensitivity to relative versus absolute payoff, and (2) comparative evaluations across general and contextual settings. To ground our analysis in psychological theory, we adapt four established psychometric questionnaires spanning general, domain-specific, workplace, and sibling-based envy. Our results reveal heterogeneous envy-like patterns across models and contexts, with some models sacrificing personal gain to reduce a peer's advantage, while others prioritize individual maximization. These findings highlight competitive dispositions as a design and safety consideration for multi-agent LLM systems.

---

### Element-wise Modulation of Random Matrices for Efficient Neural Layers

Authors: Maksymilian Szorc
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.13480v1

Fully connected layers are a primary source of memory and computational overhead in deep neural networks due to their dense, often redundant parameterization. While various compression techniques exist, they frequently introduce complex engineering trade-offs or degrade model performance. We propose the Parametrized Random Projection (PRP) layer, a novel approach that decouples feature mixing from adaptation by utilizing a fixed random matrix modulated by lightweight, learnable element-wise parameters. This architecture drastically reduces the trainable parameter count to a linear scale while retaining reliable accuracy across various benchmarks. The design serves as a stable, computationally efficient solution for architectural scaling and deployment in resource-limited settings.

---

### NRR-Core: Non-Resolution Reasoning as a Computational Framework for Contextual Identity and Ambiguity Preservation

Authors: Kei Saito
Published: 2025-12-15
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2512.13478v13

Language-processing systems that optimize for a single resolved output risk losing ambiguity. With incomplete context, competing interpretations may be compressed prematurely. We specify Non-Resolution Reasoning (NRR) as an explicit retention-commitment interface for preserving context-indexed alternatives until evidence supports commitment. NRR organizes context-indexed alternatives, independently active weights, declared retention and commitment operations, and non-destructive output projection around three principles: Context-indexed Non-Identity, Approximate Identity, and Non-Resolution. It specifies a retained state and candidate operator vocabulary, and proposes Multi-Vector Embeddings, Non-Collapsing Attention, and Contextual Identity Tracking as implementable architectural realizations. In a reproducible synthetic two-turn task, one gated Multi-Vector-Embedding instantiation maintains high output entropy before disambiguating context arrives ($H = 0.91$ bits, near the $1.0$-bit maximum), while a controlled single-embedding baseline has low entropy ($H = 0.15$ bits); both tested systems resolve correctly after context arrives. Thus, high pre-context output uncertainty and accurate later resolution can coexist in the tested gated configuration. This result does not validate the full NRR architecture or matched-parameter superiority; the specification, proposed components, and demonstrated behavior remain distinct contribution layers. NRR targets premature commitment, not commitment itself: alternatives can remain available while evidence is incomplete, and commitment occurs at explicit output or action gates. The question is not whether AI should resolve ambiguity, but when, how, and under whose control. Implementation: https://github.com/kei-saito-research/nrr-core. Series hub: https://github.com/kei-saito-research/nrr-series-hub.

---

### Scaling Laws for Code: Every Programming Language Matters

Authors: Jian Yang, Shawn Guo, Lin Jing, Wei Zhang, Aishan Liu, Chuan Hao, Zhoujun Li, Wayne Xin Zhao, Xianglong Liu, Weifeng Lv, Bryan Dai
Published: 2025-12-15
Categories: cs.CL
arXiv: https://arxiv.org/abs/2512.13472v1

Code large language models (Code LLMs) are powerful but costly to train, with scaling laws predicting performance from model size, data, and compute. However, different programming languages (PLs) have varying impacts during pre-training that significantly affect base model performance, leading to inaccurate performance prediction. Besides, existing works focus on language-agnostic settings, neglecting the inherently multilingual nature of modern software development. Therefore, it is first necessary to investigate the scaling laws of different PLs, and then consider their mutual influences to arrive at the final multilingual scaling law. In this paper, we present the first systematic exploration of scaling laws for multilingual code pre-training, conducting over 1000+ experiments (Equivalent to 336,000+ H800 hours) across multiple PLs, model sizes (0.2B to 14B parameters), and dataset sizes (1T tokens). We establish comprehensive scaling laws for code LLMs across multiple PLs, revealing that interpreted languages (e.g., Python) benefit more from increased model size and data than compiled languages (e.g., Rust). The study demonstrates that multilingual pre-training provides synergistic benefits, particularly between syntactically similar PLs. Further, the pre-training strategy of the parallel pairing (concatenating code snippets with their translations) significantly enhances cross-lingual abilities with favorable scaling properties. Finally, a proportion-dependent multilingual scaling law is proposed to optimally allocate training tokens by prioritizing high-utility PLs (e.g., Python), balancing high-synergy pairs (e.g., JavaScript-TypeScript), and reducing allocation to fast-saturating languages (Rust), achieving superior average performance across all PLs compared to uniform distribution under the same compute budget.

---

### PoseAnything: Universal Pose-guided Video Generation with Part-aware Temporal Coherence

Authors: Ruiyan Wang, Teng Hu, Kaihui Huang, Zihan Su, Ran Yi, Lizhuang Ma
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13465v1

Pose-guided video generation refers to controlling the motion of subjects in generated video through a sequence of poses. It enables precise control over subject motion and has important applications in animation. However, current pose-guided video generation methods are limited to accepting only human poses as input, thus generalizing poorly to pose of other subjects. To address this issue, we propose PoseAnything, the first universal pose-guided video generation framework capable of handling both human and non-human characters, supporting arbitrary skeletal inputs. To enhance consistency preservation during motion, we introduce Part-aware Temporal Coherence Module, which divides the subject into different parts, establishes part correspondences, and computes cross-attention between corresponding parts across frames to achieve fine-grained part-level consistency. Additionally, we propose Subject and Camera Motion Decoupled CFG, a novel guidance strategy that, for the first time, enables independent camera movement control in pose-guided video generation, by separately injecting subject and camera motion control information into the positive and negative anchors of CFG. Furthermore, we present XPose, a high-quality public dataset containing 50,000 non-human pose-video pairs, along with an automated pipeline for annotation and filtering. Extensive experiments demonstrate that Pose-Anything significantly outperforms state-of-the-art methods in both effectiveness and generalization.

---

### DP-EMAR: A Differentially Private Framework for Autonomous Model Weight Repair in Federated IoT Systems

Authors: Chethana Prasad Kabgere, Shylaja S S
Published: 2025-12-15
Categories: cs.LG, math.NA
arXiv: https://arxiv.org/abs/2512.13460v1

Federated Learning (FL) enables decentralized model training without sharing raw data, but model weight distortion remains a major challenge in resource constrained IoT networks. In multi tier Federated IoT (Fed-IoT) systems, unstable connectivity and adversarial interference can silently alter transmitted parameters, degrading convergence. We propose DP-EMAR, a differentially private, error model based autonomous repair framework that detects and reconstructs transmission induced distortions during FL aggregation. DP-EMAR estimates corruption patterns and applies adaptive correction before privacy noise is added, enabling reliable in network repair without violating confidentiality. By integrating Differential Privacy (DP) with Secure Aggregation (SA), the framework distinguishes DP noise from genuine transmission errors. Experiments on heterogeneous IoT sensor and graph datasets show that DP-EMAR preserves convergence stability and maintains near baseline performance under communication corruption while ensuring strict (epsilon, delta)-DP guarantees. The framework enhances robustness, communication efficiency, and trust in privacy preserving Federated IoT learning.

---

### SSAS: Cross-subject EEG-based Emotion Recognition through Source Selection with Adversarial Strategy

Authors: Yici Liu, Qi Wei Oung, Hoi Leong Lee
Published: 2025-12-15
Categories: cs.LG, cs.AI, eess.SP
arXiv: https://arxiv.org/abs/2512.13458v2

Electroencephalographic (EEG) signals have long been applied in the field of affective brain-computer interfaces (aBCIs). Cross-subject EEG-based emotion recognition has demonstrated significant potential in practical applications due to its suitability across diverse people. However, most studies on cross-subject EEG-based emotion recognition neglect the presence of inter-individual variability and negative transfer phenomena during model training. To address this issue, a cross-subject EEG-based emotion recognition through source selection with adversarial strategy is introduced in this paper. The proposed method comprises two modules: the source selection network (SS) and the adversarial strategies network (AS). The SS uses domain labels to reverse-engineer the training process of domain adaptation. Its key idea is to disrupt class separability and magnify inter-domain differences, thereby raising the classification difficulty and forcing the model to learn domain-invariant yet emotion-relevant representations. The AS gets the source domain selection results and the pretrained domain discriminators from SS. The pretrained domain discriminators compute a novel loss aimed at enhancing the performance of domain classification during adversarial training, ensuring the balance of adversarial strategies. This paper provides theoretical insights into the proposed method and achieves outstanding performance on two EEG-based emotion datasets, SEED and SEED-IV. The code can be found at https://github.com/liuyici/SSAS.

---

### Test-Time Modification: Inverse Domain Transformation for Robust Perception

Authors: Arpit Jadon, Joshua Niemeijer, Yuki M. Asano
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13454v4

Generative foundation models contain broad visual knowledge and can produce diverse image variations, making them particularly promising for advancing domain generalization tasks. They can be used for training data augmentation, but synthesizing comprehensive target-domain variations remains slow, expensive, and incomplete. We propose an alternative: using diffusion models at test time to map target images back to the source distribution where the downstream model was trained. This approach requires only a source domain description, preserves the task model, and eliminates large-scale synthetic data generation. We demonstrate consistent improvements across segmentation, detection, and classification tasks under challenging environmental shifts in real-to-real domain generalization scenarios with unknown target distributions. Our analysis spans multiple generative and downstream models, including an ensemble variant for enhanced robustness. The method improves BDD100K-Night-Det mAP@50 from 10.2 to 31.8, ImageNet-R top-1 from 36.1 to 60.8, and DarkZurich mIoU from 28.6 to 46.3.

---

### XNNTab -- Interpretable Neural Networks for Tabular Data using Sparse Autoencoders

Authors: Khawla Elhadri, Jörg Schlötterer, Christin Seifert
Published: 2025-12-15
Categories: cs.LG
arXiv: https://arxiv.org/abs/2512.13442v3

In data-driven applications relying on tabular data, where interpretability is key, machine learning models such as decision trees and linear regression are applied. Although neural networks can provide higher predictive performance, they are not used because of their blackbox nature. In this work, we present XNNTab, a neural architecture that combines the expressiveness of neural networks and interpretability. XNNTab first learns highly non-linear feature representations, which are decomposed into monosemantic features using a sparse autoencoder (SAE). These features are then assigned human-interpretable concepts, making the overall model prediction intrinsically interpretable. XNNTab outperforms interpretable predictive models, and achieves comparable performance to its non-interpretable counterparts.

---

### Large language models are not about natural language

Authors: Johan J. Bolhuis, Andrea Moro, Stephen Crain, Sandiway Fong
Published: 2025-12-15
Categories: cs.CL, q-bio.NC
arXiv: https://arxiv.org/abs/2512.13441v1

Large Language Models are useless for linguistics, as they are probabilistic models that require a vast amount of data to analyse externalized strings of words. In contrast, human language is underpinned by a mind-internal computational system that recursively generates hierarchical thought structures. The language system grows with minimal external input and can readily distinguish between real language and impossible languages.

---

### IMILIA: interpretable multiple instance learning for inflammation prediction in IBD from H&E whole slide images

Authors: Thalyssa Baiocco-Rodrigues, Antoine Olivier, Reda Belbahri, Thomas Duboudin, Pierre-Antoine Bannier, Benjamin Adjadj, Katharina Von Loga, Nathan Noiry, Maxime Touzot, Hector Roux de Bezieux
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13440v1

As the therapeutic target for Inflammatory Bowel Disease (IBD) shifts toward histologic remission, the accurate assessment of microscopic inflammation has become increasingly central for evaluating disease activity and response to treatment. In this work, we introduce IMILIA (Interpretable Multiple Instance Learning for Inflammation Analysis), an end-to-end framework designed for the prediction of inflammation presence in IBD digitized slides stained with hematoxylin and eosin (H&E), followed by the automated computation of markers characterizing tissue regions driving the predictions. IMILIA is composed of an inflammation prediction module, consisting of a Multiple Instance Learning (MIL) model, and an interpretability module, divided in two blocks: HistoPLUS, for cell instance detection, segmentation and classification; and EpiSeg, for epithelium segmentation. IMILIA achieves a cross-validation ROC-AUC of 0.83 on the discovery cohort, and a ROC-AUC of 0.99 and 0.84 on two external validation cohorts. The interpretability module yields biologically consistent insights: tiles with higher predicted scores show increased densities of immune cells (lymphocytes, plasmocytes, neutrophils and eosinophils), whereas lower-scored tiles predominantly contain normal epithelial cells. Notably, these patterns were consistent across all datasets. Code and models to partially replicate the results on the public IBDColEpi dataset can be found at https://github.com/owkin/imilia.

---

### From User Interface to Agent Interface: Efficiency Optimization of UI Representations for LLM Agents

Authors: Dezhi Ran, Zhi Gong, Yuzhe Guo, Mengzhou Wu, Yuan Cao, Haochuan Lu, Hengyu Zhang, Xia Zeng, Gang Cao, Liangchao Yao, Yuetang Deng, Wei Yang et al.
Published: 2025-12-15
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2512.13438v1

While Large Language Model (LLM) agents show great potential for automated UI navigation such as automated UI testing and AI assistants, their efficiency has been largely overlooked. Our motivating study reveals that inefficient UI representation creates a critical performance bottleneck. However, UI representation optimization, formulated as the task of automatically generating programs that transform UI representations, faces two unique challenges. First, the lack of Boolean oracles, which traditional program synthesis uses to decisively validate semantic correctness, poses a fundamental challenge to co-optimization of token efficiency and completeness. Second, the need to process large, complex UI trees as input while generating long, compositional transformation programs, making the search space vast and error-prone. Toward addressing the preceding limitations, we present UIFormer, the first automated optimization framework that synthesizes UI transformation programs by conducting constraint-based optimization with structured decomposition of the complex synthesis task. First, UIFormer restricts the program space using a domain-specific language (DSL) that captures UI-specific operations. Second, UIFormer conducts LLM-based iterative refinement with correctness and efficiency rewards, providing guidance for achieving the efficiency-completeness co-optimization. UIFormer operates as a lightweight plugin that applies transformation programs for seamless integration with existing LLM agents, requiring minimal modifications to their core logic. Evaluations across three UI navigation benchmarks spanning Android and Web platforms with five LLMs demonstrate that UIFormer achieves 48.7% to 55.8% token reduction with minimal runtime overhead while maintaining or improving agent performance. Real-world industry deployment at WeChat further validates the practical impact of UIFormer.

---

### Self-Supervised Ultrasound Representation Learning for Renal Anomaly Prediction in Prenatal Imaging

Authors: Youssef Megahed, Inok Lee, Robin Ducharme, Kevin Dick, Adrian D. C. Chan, Steven Hawken, Mark C. Walker
Published: 2025-12-15
Categories: eess.IV, cs.CV
arXiv: https://arxiv.org/abs/2512.13434v1

Prenatal ultrasound is the cornerstone for detecting congenital anomalies of the kidneys and urinary tract, but diagnosis is limited by operator dependence and suboptimal imaging conditions. We sought to assess the performance of a self-supervised ultrasound foundation model for automated fetal renal anomaly classification using a curated dataset of 969 two-dimensional ultrasound images. A pretrained Ultrasound Self-Supervised Foundation Model with Masked Autoencoding (USF-MAE) was fine-tuned for binary and multi-class classification of normal kidneys, urinary tract dilation, and multicystic dysplastic kidney. Models were compared with a DenseNet-169 convolutional baseline using cross-validation and an independent test set. USF-MAE consistently improved upon the baseline across all evaluation metrics in both binary and multi-class settings. USF-MAE achieved an improvement of about 1.87% (AUC) and 7.8% (F1-score) on the validation set, 2.32% (AUC) and 4.33% (F1-score) on the independent holdout test set. The largest gains were observed in the multi-class setting, where the improvement in AUC was 16.28% and 46.15% in F1-score. To facilitate model interpretability, Score-CAM visualizations were adapted for a transformer architecture and show that model predictions were informed by known, clinically relevant renal structures, including the renal pelvis in urinary tract dilation and cystic regions in multicystic dysplastic kidney. These results show that ultrasound-specific self-supervised learning can generate a useful representation as a foundation for downstream diagnostic tasks. The proposed framework offers a robust, interpretable approach to support the prenatal detection of renal anomalies and demonstrates the promise of foundation models in obstetric imaging.

---

### A Domain-Adapted Lightweight Ensemble for Resource-Efficient Few-Shot Plant Disease Classification

Authors: Anika Islam, Tasfia Tahsin, Zaarin Anjum, Md. Bakhtiar Hasan, Md. Hasanul Kabir
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13428v1

Accurate and timely identification of plant leaf diseases is essential for resilient and sustainable agriculture, yet most deep learning approaches rely on large annotated datasets and computationally intensive models that are unsuitable for data-scarce and resource-constrained environments. To address these challenges we present a few-shot learning approach within a lightweight yet efficient framework that combines domain-adapted MobileNetV2 and MobileNetV3 models as feature extractors, along with a feature fusion technique to generate robust feature representation. For the classification task, the fused features are passed through a Bi-LSTM classifier enhanced with attention mechanisms to capture sequential dependencies and focus on the most relevant features, thereby achieving optimal classification performance even in complex, real-world environments with noisy or cluttered backgrounds. The proposed framework was evaluated across multiple experimental setups, including both laboratory-controlled and field-captured datasets. On tomato leaf diseases from the PlantVillage dataset, it consistently improved performance across 1 to 15 shot scenarios, reaching 98.23+-0.33% at 15 shot, closely approaching the 99.98% SOTA benchmark achieved by a Transductive LSTM with attention, while remaining lightweight and mobile-friendly. Under real-world conditions using field images from the Dhan Shomadhan dataset, it maintained robust performance, reaching 69.28+-1.49% at 15-shot and demonstrating strong resilience to complex backgrounds. Notably, it also outperformed the previous SOTA accuracy of 96.0% on six diseases from PlantVillage, achieving 99.72% with only 15-shot learning. With a compact model size of approximately 40 MB and inference complexity of approximately 1.12 GFLOPs, this work establishes a scalable, mobile-ready foundation for precise plant disease diagnostics in data-scarce regions.

---

### MineTheGap: Automatic Mining of Biases in Text-to-Image Models

Authors: Noa Cohen, Nurit Spingarn-Eliezer, Inbar Huberman-Spiegelglas, Tomer Michaeli
Published: 2025-12-15
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2512.13427v1

Text-to-Image (TTI) models generate images based on text prompts, which often leave certain aspects of the desired image ambiguous. When faced with these ambiguities, TTI models have been shown to exhibit biases in their interpretations. These biases can have societal impacts, e.g., when showing only a certain race for a stated occupation. They can also affect user experience when creating redundancy within a set of generated images instead of spanning diverse possibilities. Here, we introduce MineTheGap - a method for automatically mining prompts that cause a TTI model to generate biased outputs. Our method goes beyond merely detecting bias for a given prompt. Rather, it leverages a genetic algorithm to iteratively refine a pool of prompts, seeking for those that expose biases. This optimization process is driven by a novel bias score, which ranks biases according to their severity, as we validate on a dataset with known biases. For a given prompt, this score is obtained by comparing the distribution of generated images to the distribution of LLM-generated texts that constitute variations on the prompt. Code and examples are available on the project's webpage.

---

### RecTok: Reconstruction Distillation along Rectified Flow

Authors: Qingyu Shi, Size Wu, Jinbin Bai, Kaidong Yu, Yujing Wang, Yunhai Tong, Xiangtai Li, Xuelong Li
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13421v2

Visual tokenizers play a crucial role in diffusion models. The dimensionality of latent space governs both reconstruction fidelity and the semantic expressiveness of the latent feature. However, a fundamental trade-off is inherent between dimensionality and generation quality, constraining existing methods to low-dimensional latent spaces. Although recent works have leveraged vision foundation models to enrich the semantics of visual tokenizers and accelerate convergence, high-dimensional tokenizers still underperform their low-dimensional counterparts. In this work, we propose RecTok, which overcomes the limitations of high-dimensional visual tokenizers through two key innovations: flow semantic distillation and reconstruction--alignment distillation. Our key insight is to make the forward flow in flow matching semantically rich, which serves as the training space of diffusion transformers, rather than focusing on the latent space as in previous works. Specifically, our method distills the semantic information in VFMs into the forward flow trajectories in flow matching. And we further enhance the semantics by introducing a masked feature reconstruction loss. Our RecTok achieves superior image reconstruction, generation quality, and discriminative performance. It achieves state-of-the-art results on the gFID-50K under both with and without classifier-free guidance settings, while maintaining a semantically rich latent space structure. Furthermore, as the latent dimensionality increases, we observe consistent improvements. Code and model are available at https://shi-qingyu.github.io/rectok.github.io.

---

### Learning to Generate Cross-Task Unexploitable Examples

Authors: Haoxuan Qu, Qiuchi Xiang, Yujun Cai, Yirui Wu, Majid Mirmehdi, Hossein Rahmani, Jun Liu
Published: 2025-12-15
Categories: cs.CV
arXiv: https://arxiv.org/abs/2512.13416v1

Unexploitable example generation aims to transform personal images into their unexploitable (unlearnable) versions before they are uploaded online, thereby preventing unauthorized exploitation of online personal images. Recently, this task has garnered significant research attention due to its critical relevance to personal data privacy. Yet, despite recent progress, existing methods for this task can still suffer from limited practical applicability, as they can fail to generate examples that are broadly unexploitable across different real-world computer vision tasks. To deal with this problem, in this work, we propose a novel Meta Cross-Task Unexploitable Example Generation (MCT-UEG) framework. At the core of our framework, to optimize the unexploitable example generator for effectively producing broadly unexploitable examples, we design a flat-minima-oriented meta training and testing scheme. Extensive experiments show the efficacy of our framework.

---