# Research Papers — 2026-01 · bundle page 38 of 189 > Full-text markdown of papers #1851–1900 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-01/page/38 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/37.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/39.md) ### Endogenous Reprompting: Self-Evolving Cognitive Alignment for Unified Multimodal Models Authors: Zhenchen Tang, Songlin Yang, Zichuan Wang, Bo Peng, Yang Li, Beibei Dong, Jing Dong Published: 2026-01-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.20305v1 Unified Multimodal Models (UMMs) exhibit strong understanding, yet this capability often fails to effectively guide generation. We identify this as a Cognitive Gap: the model lacks the understanding of how to enhance its own generation process. To bridge this gap, we propose Endogenous Reprompting, a mechanism that transforms the model's understanding from a passive encoding process into an explicit generative reasoning step by generating self-aligned descriptors during generation. To achieve this, we introduce SEER (Self-Evolving Evaluator and Reprompter), a training framework that establishes a two-stage endogenous loop using only 300 samples from a compact proxy task, Visual Instruction Elaboration. First, Reinforcement Learning with Verifiable Rewards (RLVR) activates the model's latent evaluation ability via curriculum learning, producing a high-fidelity endogenous reward signal. Second, Reinforcement Learning with Model-rewarded Thinking (RLMT) leverages this signal to optimize the generative reasoning policy. Experiments show that SEER consistently outperforms state-of-the-art baselines in evaluation accuracy, reprompting efficiency, and generation quality, without sacrificing general multimodal capabilities. --- ### Structure-constrained Language-informed Diffusion Model for Unpaired Low-dose Computed Tomography Angiography Reconstruction Authors: Genyuan Zhang, Zihao Wang, Zhifan Gao, Lei Xu, Zhen Zhou, Haijun Yu, Jianjia Zhang, Xiujian Liu, Weiwei Zhang, Shaoyu Wang, Huazhu Fu, Fenglin Liu et al. Published: 2026-01-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.20304v1 The application of iodinated contrast media (ICM) improves the sensitivity and specificity of computed tomography (CT) for a wide range of clinical indications. However, overdose of ICM can cause problems such as kidney damage and life-threatening allergic reactions. Deep learning methods can generate CT images of normal-dose ICM from low-dose ICM, reducing the required dose while maintaining diagnostic power. However, existing methods are difficult to realize accurate enhancement with incompletely paired images, mainly because of the limited ability of the model to recognize specific structures. To overcome this limitation, we propose a Structure-constrained Language-informed Diffusion Model (SLDM), a unified medical generation model that integrates structural synergy and spatial intelligence. First, the structural prior information of the image is effectively extracted to constrain the model inference process, thus ensuring structural consistency in the enhancement process. Subsequently, semantic supervision strategy with spatial intelligence is introduced, which integrates the functions of visual perception and spatial reasoning, thus prompting the model to achieve accurate enhancement. Finally, the subtraction angiography enhancement module is applied, which serves to improve the contrast of the ICM agent region to suitable interval for observation. Qualitative analysis of visual comparison and quantitative results of several metrics demonstrate the effectiveness of our method in angiographic reconstruction for low-dose contrast medium CT angiography. --- ### Physically Guided Visual Mass Estimation from a Single RGB Image Authors: Sungjae Lee, Junhan Jeong, Yeonjoo Hong, Kwang In Kim Published: 2026-01-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.20303v2 Estimating object mass from visual input is challenging because mass depends jointly on geometric volume and material-dependent density, neither of which is directly observable from RGB appearance. Consequently, mass prediction from pixels is ill-posed and therefore benefits from physically meaningful representations to constrain the space of plausible solutions. We propose a physically structured framework for single-image mass estimation that addresses this ambiguity by aligning visual cues with the physical factors governing mass. From a single RGB image, we recover object-centric three-dimensional geometry via monocular depth estimation to inform volume and extract coarse material semantics using a vision-language model to guide density-related reasoning. These geometry, semantic, and appearance representations are fused through an instance-adaptive gating mechanism, and two physically guided latent factors (volume- and density-related) are predicted through separate regression heads under mass-only supervision. Experiments on image2mass and ABO-500 show that the proposed method consistently outperforms state-of-the-art methods. --- ### Bridging the Applicator Gap with Data-Doping:Dual-Domain Learning for Precise Bladder Segmentation in CT-Guided Brachytherapy Authors: Suresh Das, Siladittya Manna, Sayantari Ghosh Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20302v1 Performance degradation due to covariate shift remains a major challenge for deep learning models in medical image segmentation. An open question is whether samples from a shifted distribution can effectively support learning when combined with limited target domain data. We investigate this problem in the context of bladder segmentation in CT guided gynecological brachytherapy, a critical task for accurate dose optimization and organ at risk sparing. While CT scans without brachytherapy applicators (no applicator: NA) are widely available, scans with applicators inserted (with applicator: WA) are scarce and exhibit substantial anatomical deformation and imaging artifacts, making automated segmentation particularly difficult. We propose a dual domain learning strategy that integrates NA and WA CT data to improve robustness and generalizability under covariate shift. Using a curated assorted dataset, we show that NA data alone fail to capture the anatomical and artifact related characteristics of WA images. However, introducing a modest proportion of WA data into a predominantly NA training set leads to significant performance improvements. Through systematic experiments across axial, coronal, and sagittal planes using multiple deep learning architectures, we demonstrate that doping only 10 to 30 percent WA data achieves segmentation performance comparable to models trained exclusively on WA data. The proposed approach attains Dice similarity coefficients of up to 0.94 and Intersection over Union scores of up to 0.92, indicating effective domain adaptation and improved clinical reliability. This study highlights the value of integrating anatomically similar but distribution shifted datasets to overcome data scarcity and enhance deep learning based segmentation for brachytherapy treatment planning. --- ### Towards Compact and Robust DNNs via Compression-aware Sharpness Minimization Authors: Jialuo He, Huangxun Chen Published: 2026-01-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.20301v1 Sharpness-Aware Minimization (SAM) has recently emerged as an effective technique for improving DNN robustness to input variations. However, its interplay with the compactness requirements of on-device DNN deployments remains less explored. Simply pruning a SAM-trained model can undermine robustness, since flatness in the continuous parameter space does not necessarily translate to robustness under the discrete structural changes induced by pruning. Conversely, applying SAM after pruning may be fundamentally constrained by architectural limitations imposed by an early, robustness-agnostic pruning pattern. To address this gap, we propose Compression-aware ShArpness Minimization (C-SAM), a framework that shifts sharpness-aware learning from parameter perturbations to mask perturbations. By explicitly perturbing pruning masks during training, C-SAM promotes a flatter loss landscape with respect to model structure, enabling the discovery of pruning patterns that simultaneously optimize model compactness and robustness to input variations. Extensive experiments on CelebA-HQ, Flowers-102, and CIFAR-10-C across ResNet-18, GoogLeNet, and MobileNet-V2 show that C-SAM consistently achieves higher certified robustness than strong baselines, with improvements of up to 42%, while maintaining task accuracy comparable to the corresponding unpruned models. --- ### MiLorE-SSL: Scaling Multilingual Capabilities in Self-Supervised Models without Forgetting Authors: Jing Xu, Minglin Wu, Xueyuan Chen, Xixin Wu, Helen Meng Published: 2026-01-28 Categories: cs.CL, eess.AS arXiv: https://arxiv.org/abs/2601.20300v1 Self-supervised learning (SSL) has greatly advanced speech representation learning, but multilingual SSL models remain constrained to languages encountered during pretraining. Retraining from scratch to incorporate new languages is computationally expensive, while sequential training without migitation strategies often leads to catastrophic forgetting. To address this, we propose MiLorE-SSL, a lightweight framework that combines LoRA modules with a soft mixture-of-experts (MoE) mechanism for efficient continual multilingual training. LoRA provides efficient low-rank adaptation, while soft MoE promotes flexible expert sharing across languages, reducing cross-lingual interference. To further mitigate forgetting, we introduce limited replay data from existing languages, avoiding reliance on large historical corpora. Experiments on ML-SUPERB demonstrate that MiLorE-SSL achieves strong performance in new languages and improves the ability in existing ones with only 2.14% trainable parameters. --- ### Truthfulness Despite Weak Supervision: Evaluating and Training LLMs Using Peer Prediction Authors: Tianyi Alex Qiu, Micah Carroll, Cameron Allen Published: 2026-01-28 Categories: cs.LG, cs.AI, cs.CL, cs.GT arXiv: https://arxiv.org/abs/2601.20299v1 The evaluation and post-training of large language models (LLMs) rely on supervision, but strong supervision for difficult tasks is often unavailable, especially when evaluating frontier models. In such cases, models are demonstrated to exploit evaluations built on such imperfect supervision, leading to deceptive results. However, underutilized in LLM research, a wealth of mechanism design research focuses on game-theoretic incentive compatibility, i.e., eliciting honest and informative answers with weak supervision. Drawing from this literature, we introduce the peer prediction method for model evaluation and post-training. It rewards honest and informative answers over deceptive and uninformative ones, using a metric based on mutual predictability and without requiring ground truth labels. We demonstrate the method's effectiveness and resistance to deception, with both theoretical guarantees and empirical validation on models with up to 405B parameters. We show that training an 8B model with peer prediction-based reward recovers most of the drop in truthfulness due to prior malicious finetuning, even when the reward is produced by a 0.135B language model with no finetuning. On the evaluation front, in contrast to LLM-as-a-Judge which requires strong and trusted judges, we discover an inverse scaling property in peer prediction, where, surprisingly, resistance to deception is strengthened as the capability gap between the experts and participants widens, enabling reliable evaluation of strong models with weak supervision. In particular, LLM-as-a-Judge become worse than random guess when facing deceptive models 5-20x the judge's size, while peer prediction thrives when such gaps are large, including in cases with over 100x size difference. --- ### Artifact-Aware Evaluation for High-Quality Video Generation Authors: Chen Zhu, Jiashu Zhu, Yanxun Li, Meiqi Wu, Bingze Song, Chubin Chen, Jiahong Wu, Xiangxiang Chu, Yangang Wang Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20297v1 With the rapid advancement of video generation techniques, evaluating and auditing generated videos has become increasingly crucial. Existing approaches typically offer coarse video quality scores, lacking detailed localization and categorization of specific artifacts. In this work, we introduce a comprehensive evaluation protocol focusing on three key aspects affecting human perception: Appearance, Motion, and Camera. We define these axes through a taxonomy of 10 prevalent artifact categories reflecting common generative failures observed in video generation. To enable robust artifact detection and categorization, we introduce GenVID, a large-scale dataset of 80k videos generated by various state-of-the-art video generation models, each carefully annotated for the defined artifact categories. Leveraging GenVID, we develop DVAR, a Dense Video Artifact Recognition framework for fine-grained identification and classification of generative artifacts. Extensive experiments show that our approach significantly improves artifact detection accuracy and enables effective filtering of low-quality content. --- ### Cheap2Rich: A Multi-Fidelity Framework for Data Assimilation and System Identification of Multiscale Physics -- Rotating Detonation Engines Authors: Yuxuan Bao, Jan Zajac, Megan Powers, Venkat Raman, J. Nathan Kutz Published: 2026-01-28 Categories: cs.LG, cs.AI, cs.CE, math.DS arXiv: https://arxiv.org/abs/2601.20295v1 Bridging the sim2real gap between computationally inexpensive models and complex physical systems remains a central challenge in machine learning applications to engineering problems, particularly in multi-scale settings where reduced-order models typically capture only dominant dynamics. In this work, we present Cheap2Rich, a multi-scale data assimilation framework that reconstructs high-fidelity state spaces from sparse sensor histories by combining a fast low-fidelity prior with learned, interpretable discrepancy corrections. We demonstrate the performance on rotating detonation engines (RDEs), a challenging class of systems that couple detonation-front propagation with injector-driven unsteadiness, mixing, and stiff chemistry across disparate scales. Our approach successfully reconstructs high-fidelity RDE states from sparse measurements while isolating physically meaningful discrepancy dynamics associated with injector-driven effects. The results highlight a general multi-fidelity framework for data assimilation and system identification in complex multi-scale systems, enabling rapid design exploration and real-time monitoring and control while providing interpretable discrepancy dynamics. Code for this project is is available at: github.com/kro0l1k/Cheap2Rich. --- ### A Learning-based Framework for Spatial Impulse Response Compensation in 3D Photoacoustic Computed Tomography Authors: Kaiyi Yang, Seonyeong Park, Gangwon Jeong, Hsuan-Kai Huang, Alexander A. Oraevsky, Umberto Villa, Mark A. Anastasio Published: 2026-01-28 Categories: cs.LG, eess.SP, physics.med-ph arXiv: https://arxiv.org/abs/2601.20291v1 Photoacoustic computed tomography (PACT) is a promising imaging modality that combines the advantages of optical contrast with ultrasound detection. Utilizing ultrasound transducers with larger surface areas can improve detection sensitivity. However, when computationally efficient analytic reconstruction methods that neglect the spatial impulse responses (SIRs) of the transducer are employed, the spatial resolution of the reconstructed images will be compromised. Although optimization-based reconstruction methods can explicitly account for SIR effects, their computational cost is generally high, particularly in three-dimensional (3D) applications. To address the need for accurate but rapid 3D PACT image reconstruction, this study presents a framework for establishing a learned SIR compensation method that operates in the data domain. The learned compensation method maps SIR-corrupted PACT measurement data to compensated data that would have been recorded by idealized point-like transducers. Subsequently, the compensated data can be used with a computationally efficient reconstruction method that neglects SIR effects. Two variants of the learned compensation model are investigated that employ a U-Net model and a specifically designed, physics-inspired model, referred to as Deconv-Net. A fast and analytical training data generation procedure is also a component of the presented framework. The framework is rigorously validated in virtual imaging studies, demonstrating resolution improvement and robustness to noise variations, object complexity, and sound speed heterogeneity. When applied to in-vivo breast imaging data, the learned compensation models revealed fine structures that had been obscured by SIR-induced artifacts. To our knowledge, this is the first demonstration of learned SIR compensation in 3D PACT imaging. --- ### A Source-Free Approach for Domain Adaptation via Multiview Image Transformation and Latent Space Consistency Authors: Debopom Sutradhar, Md. Abdur Rahman, Mohaimenul Azam Khan Raiaan, Reem E. Mohamed, Sami Azam Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20284v1 Domain adaptation (DA) addresses the challenge of transferring knowledge from a source domain to a target domain where image data distributions may differ. Existing DA methods often require access to source domain data, adversarial training, or complex pseudo-labeling techniques, which are computationally expensive. To address these challenges, this paper introduces a novel source-free domain adaptation method. It is the first approach to use multiview augmentation and latent space consistency techniques to learn domain-invariant features directly from the target domain. Our method eliminates the need for source-target alignment or pseudo-label refinement by learning transferable representations solely from the target domain by enforcing consistency between multiple augmented views in the latent space. Additionally, the method ensures consistency in the learned features by generating multiple augmented views of target domain data and minimizing the distance between their feature representations in the latent space. We also introduce a ConvNeXt-based encoder and design a loss function that combines classification and consistency objectives to drive effective adaptation directly from the target domain. The proposed model achieves an average classification accuracy of 90. 72\%, 84\%, and 97. 12\% in Office-31, Office-Home and Office-Caltech datasets, respectively. Further evaluations confirm that our study improves existing methods by an average classification accuracy increment of +1.23\%, +7.26\%, and +1.77\% on the respective datasets. --- ### One Word is Enough: Minimal Adversarial Perturbations for Neural Text Ranking Authors: Tanmay Karmakar, Sourav Saha, Debapriyo Majumdar, Surjyanee Halder Published: 2026-01-28 Categories: cs.IR, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2601.20283v1 Neural ranking models (NRMs) achieve strong retrieval effectiveness, yet prior work has shown they are vulnerable to adversarial perturbations. We revisit this robustness question with a minimal, query-aware attack that promotes a target document by inserting or substituting a single, semantically aligned word - the query center. We study heuristic and gradient-guided variants, including a white-box method that identifies influential insertion points. On TREC-DL 2019/2020 with BERT and monoT5 re-rankers, our single-word attacks achieve up to 91% success while modifying fewer than two tokens per document on average, achieving competitive rank and score boosts with far fewer edits under a comparable white-box setup to ensure fair evaluation against PRADA. We also introduce new diagnostic metrics to analyze attack sensitivity beyond aggregate success rates. Our analysis reveals a Goldilocks zone in which mid-ranked documents are most vulnerable. These findings demonstrate practical risks and motivate future defenses for robust neural ranking. --- ### Memory Retrieval in Transformers: Insights from The Encoding Specificity Principle Authors: Viet Hung Dinh, Ming Ding, Youyang Qu, Kanchana Thilakarathna Published: 2026-01-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.20282v1 While explainable artificial intelligence (XAI) for large language models (LLMs) remains an evolving field with many unresolved questions, increasing regulatory pressures have spurred interest in its role in ensuring transparency, accountability, and privacy-preserving machine unlearning. Despite recent advances in XAI have provided some insights, the specific role of attention layers in transformer based LLMs remains underexplored. This study investigates the memory mechanisms instantiated by attention layers, drawing on prior research in psychology and computational psycholinguistics that links Transformer attention to cue based retrieval in human memory. In this view, queries encode the retrieval context, keys index candidate memory traces, attention weights quantify cue trace similarity, and values carry the encoded content, jointly enabling the construction of a context representation that precedes and facilitates memory retrieval. Guided by the Encoding Specificity Principle, we hypothesize that the cues used in the initial stage of retrieval are instantiated as keywords. We provide converging evidence for this keywords-as-cues hypothesis. In addition, we isolate neurons within attention layers whose activations selectively encode and facilitate the retrieval of context-defining keywords. Consequently, these keywords can be extracted from identified neurons and further contribute to downstream applications such as unlearning. --- ### The Forecast After the Forecast: A Post-Processing Shift in Time Series Authors: Daojun Liang, Qi Li, Yinglong Wang, Jing Chen, Hu Zhang, Xiaoxiao Cui, Qizheng Wang, Shuo Li Published: 2026-01-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.20280v1 Time series forecasting has long been dominated by advances in model architecture, with recent progress driven by deep learning and hybrid statistical techniques. However, as forecasting models approach diminishing returns in accuracy, a critical yet underexplored opportunity emerges: the strategic use of post-processing. In this paper, we address the last-mile gap in time-series forecasting, which is to improve accuracy and uncertainty without retraining or modifying a deployed backbone. We propose $δ$-Adapter, a lightweight, architecture-agnostic way to boost deployed time series forecasters without retraining. $δ$-Adapter learns tiny, bounded modules at two interfaces: input nudging (soft edits to covariates) and output residual correction. We provide local descent guarantees, $O(δ)$ drift bounds, and compositional stability for combined adapters. Meanwhile, it can act as a feature selector by learning a sparse, horizon-aware mask over inputs to select important features, thereby improving interpretability. In addition, it can also be used as a distribution calibrator to measure uncertainty. Thus, we introduce a Quantile Calibrator and a Conformal Corrector that together deliver calibrated, personalized intervals with finite-sample coverage. Our experiments across diverse backbones and datasets show that $δ$-Adapter improves accuracy and calibration with negligible compute and no interface changes. --- ### Hallucination Begins Where Saliency Drops Authors: Xiaofeng Zhang, Yuanchao Zhu, Chaochen Gu, Xiaosong Yuan, Qiyan Zhao, Jiawei Cao, Feilong Tang, Sinan Fan, Yaomin Shen, Chen Shen, Hao Tang Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20279v1 Recent studies have examined attention dynamics in large vision-language models (LVLMs) to detect hallucinations. However, existing approaches remain limited in reliably distinguishing hallucinated from factually grounded outputs, as they rely solely on forward-pass attention patterns and neglect gradient-based signals that reveal how token influence propagates through the network. To bridge this gap, we introduce LVLMs-Saliency, a gradient-aware diagnostic framework that quantifies the visual grounding strength of each output token by fusing attention weights with their input gradients. Our analysis uncovers a decisive pattern: hallucinations frequently arise when preceding output tokens exhibit low saliency toward the prediction of the next token, signaling a breakdown in contextual memory retention. Leveraging this insight, we propose a dual-mechanism inference-time framework to mitigate hallucinations: (1) Saliency-Guided Rejection Sampling (SGRS), which dynamically filters candidate tokens during autoregressive decoding by rejecting those whose saliency falls below a context-adaptive threshold, thereby preventing coherence-breaking tokens from entering the output sequence; and (2) Local Coherence Reinforcement (LocoRE), a lightweight, plug-and-play module that strengthens attention from the current token to its most recent predecessors, actively counteracting the contextual forgetting behavior identified by LVLMs-Saliency. Extensive experiments across multiple LVLMs demonstrate that our method significantly reduces hallucination rates while preserving fluency and task performance, offering a robust and interpretable solution for enhancing model reliability. Code is available at: https://github.com/zhangbaijin/LVLMs-Saliency --- ### Beyond the Needle's Illusion: Decoupled Evaluation of Evidence Access and Use under Semantic Interference at 326M-Token Scale Authors: Tianwei Lin, Zuyi Zhou, Xinda Zhao, Chenke Wang, Xiaohong Li, Yu Chen, Chuanrui Hu, Jian Pei, Yafeng Deng Published: 2026-01-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.20276v1 Long-context LLM agents must access the right evidence from large environments and use it faithfully. However, the popular Needle-in-a-Haystack (NIAH) evaluation mostly measures benign span localization. The needle is near-unique, and the haystack is largely irrelevant. We introduce EverMemBench-S (EMB-S), an adversarial NIAH-style benchmark built on a 326M-token MemoryBank. While the full MemoryBank spans 326M tokens for retrieval-based (RAG) evaluation, we evaluate native long-context models only at scales that fit within each model's context window (up to 1M tokens in this work) to ensure a fair comparison. EMB-S pairs queries with collision-tested near-miss hard negatives and gold evidence sets spanning one or more documents, validated via human screening and LLM verification. We also propose a decoupled diagnostic protocol that reports evidence access (document-ID localization) separately from end-to-end QA quality under full-context prompting. This enables consistent diagnosis for both native long-context prompting and retrieval pipelines. Across a reference-corpus ladder from domain-isolated 64K contexts to a globally shared 326M-token environment, we observe a clear reality gap. Systems that saturate benign NIAH degrade sharply in evidence access under semantic interference. These results indicate that semantic discrimination, not context length alone, is the dominant bottleneck for long-context memory at scale. --- ### RusLICA: A Russian-Language Platform for Automated Linguistic Inquiry and Category Analysis Authors: Elina Sigdel, Anastasia Panfilova Published: 2026-01-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.20275v1 Defining psycholinguistic characteristics in written texts is a task gaining increasing attention from researchers. One of the most widely used tools in the current field is Linguistic Inquiry and Word Count (LIWC) that originally was developed to analyze English texts and translated into multiple languages. Our approach offers the adaptation of LIWC methodology for the Russian language, considering its grammatical and cultural specificities. The suggested approach comprises 96 categories, integrating syntactic, morphological, lexical, general statistical features, and results of predictions obtained using pre-trained language models (LMs) for text analysis. Rather than applying direct translation to existing thesauri, we built the dictionary specifically for the Russian language based on the content from several lexicographic resources, semantic dictionaries and corpora. The paper describes the process of mapping lemmas to 42 psycholinguistic categories and the implementation of the analyzer as part of RusLICA web service. --- ### SwiftFusion: Scalable Sequence Parallelism for Distributed Inference of Diffusion Transformers on GPUs Authors: Jiacheng Yang, Jun Wu, Yaoyao Ding, Zhiying Xu, Yida Wang, Gennady Pekhimenko Published: 2026-01-28 Categories: cs.DC, cs.CV arXiv: https://arxiv.org/abs/2601.20273v2 Diffusion Transformers (DiTs) have gained increasing adoption in high-quality image and video generation. As demand for higher-resolution images and longer videos increases, single-GPU inference becomes inefficient due to increased latency and large activation sizes. Current frameworks employ sequence parallelism (SP) techniques such as Ulysses Attention and Ring Attention to scale inference. However, these implementations have three primary limitations: (1) suboptimal communication patterns for network topologies on modern GPU machines, (2) latency bottlenecks from all-to-all operations in inter-machine communication, and (3) GPU sender-receiver synchronization and computation overheads from using two-sided communication libraries. To address these issues, we present StreamFusion, a topology-aware efficient DiT serving engine. StreamFusion incorporates three key innovations: (1) a topology-aware sequence parallelism technique that accounts for inter- and intra-machine bandwidth differences, (2) Torus Attention, a novel SP technique enabling overlapping of inter-machine all-to-all operations with computation, and (3) a one-sided communication implementation that minimizes GPU sender-receiver synchronization and computation overheads. Our experiments demonstrate that StreamFusion outperforms the state-of-the-art approach by an average of $1.35\times$ (up to $1.77\times$). --- ### Eliciting Least-to-Most Reasoning for Phishing URL Detection Authors: Holly Trikilis, Pasindu Marasinghe, Fariza Rashid, Suranga Seneviratne Published: 2026-01-28 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2601.20270v1 Phishing continues to be one of the most prevalent attack vectors, making accurate classification of phishing URLs essential. Recently, large language models (LLMs) have demonstrated promising results in phishing URL detection. However, their reasoning capabilities that enabled such performance remain underexplored. To this end, in this paper, we propose a Least-to-Most prompting framework for phishing URL detection. In particular, we introduce an "answer sensitivity" mechanism that guides Least-to-Most's iterative approach to enhance reasoning and yield higher prediction accuracy. We evaluate our framework using three URL datasets and four state-of-the-art LLMs, comparing against a one-shot approach and a supervised model. We demonstrate that our framework outperforms the one-shot baseline while achieving performance comparable to that of the supervised model, despite requiring significantly less training data. Furthermore, our in-depth analysis highlights how the iterative reasoning enabled by Least-to-Most, and reinforced by our answer sensitivity mechanism, drives these performance gains. Overall, we show that this simple yet powerful prompting strategy consistently outperforms both one-shot and supervised approaches, despite requiring minimal training or few-shot guidance. Our experimental setup can be found in our Github repository github.sydney.edu.au/htri0928/least-to-most-phishing-detection. --- ### Empirical Likelihood-Based Fairness Auditing: Distribution-Free Certification and Flagging Authors: Jie Tang, Chuanlong Xie, Xianli Zeng, Lixing Zhu Published: 2026-01-28 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2601.20269v2 Machine learning models in high-stakes applications, such as recidivism prediction and automated personnel selection, often exhibit systematic performance disparities across sensitive subpopulations, raising critical concerns regarding algorithmic bias. Fairness auditing addresses these risks through two primary functions: certification, which verifies adherence to fairness constraints; and flagging, which isolates specific demographic groups experiencing disparate treatment. However, existing auditing techniques are frequently limited by restrictive distributional assumptions or prohibitive computational overhead. We propose a novel empirical likelihood-based (EL) framework that constructs robust statistical measures for model performance disparities. Unlike traditional methods, our approach is non-parametric; the proposed disparity statistics follow asymptotically chi-square or mixed chi-square distributions, ensuring valid inference without assuming underlying data distributions. This framework uses a constrained optimization profile that admits stable numerical solutions, facilitating both large-scale certification and efficient subpopulation discovery. Empirically, the EL methods outperform bootstrap-based approaches, yielding coverage rates closer to nominal levels while reducing computational latency by several orders of magnitude. We demonstrate the practical utility of this framework on the COMPAS dataset, where it successfully flags intersectional biases, specifically identifying a significantly higher positive prediction rate for African-American males under 25 and a systemic under-prediction for Caucasian females relative to the population mean. --- ### Robust SDE Parameter Estimation Under Missing Time Information Setting Authors: Long Van Tran, Truyen Tran, Phuoc Nguyen Published: 2026-01-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.20268v1 Recent advances in stochastic differential equations (SDEs) have enabled robust modeling of real-world dynamical processes across diverse domains, such as finance, health, and systems biology. However, parameter estimation for SDEs typically relies on accurately timestamped observational sequences. When temporal ordering information is corrupted, missing, or deliberately hidden (e.g., for privacy), existing estimation methods often fail. In this paper, we investigate the conditions under which temporal order can be recovered and introduce a novel framework that simultaneously reconstructs temporal information and estimates SDE parameters. Our approach exploits asymmetries between forward and backward processes, deriving a score-matching criterion to infer the correct temporal order between pairs of observations. We then recover the total order via a sorting procedure and estimate SDE parameters from the reconstructed sequence using maximum likelihood. Finally, we conduct extensive experiments on synthetic and real-world datasets to demonstrate the effectiveness of our method, extending parameter estimation to settings with missing temporal order and broadening applicability in sensitive domains. --- ### Reversible Efficient Diffusion for Image Fusion Authors: Xingxin Xu, Bing Cao, DongDong Li, Qinghua Hu, Pengfei Zhu Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20260v1 Multi-modal image fusion aims to consolidate complementary information from diverse source images into a unified representation. The fused image is expected to preserve fine details and maintain high visual fidelity. While diffusion models have demonstrated impressive generative capabilities in image generation, they often suffer from detail loss when applied to image fusion tasks. This issue arises from the accumulation of noise errors inherent in the Markov process, leading to inconsistency and degradation in the fused results. However, incorporating explicit supervision into end-to-end training of diffusion-based image fusion introduces challenges related to computational efficiency. To address these limitations, we propose the Reversible Efficient Diffusion (RED) model - an explicitly supervised training framework that inherits the powerful generative capability of diffusion models while avoiding the distribution estimation. --- ### C2:Cross learning module enhanced decision transformer with Constraint-aware loss for auto-bidding Authors: Jinren Ding, Xuejian Xu, Shen Jiang, Zhitong Hao, Jinhui Yang, Peng Jiang Published: 2026-01-28 Categories: cs.LG, cs.GT arXiv: https://arxiv.org/abs/2601.20257v2 Decision Transformer (DT) shows promise for generative auto-bidding by capturing temporal dependencies, but suffers from two critical limitations: insufficient cross-correlation modeling among state, action, and return-to-go (RTG) sequences, and indiscriminate learning of optimal/suboptimal behaviors. To address these, we propose C2, a novel framework enhancing DT with two core innovations: (1) a Cross Learning Block (CLB) via cross-attention to strengthen inter-sequence correlation modeling; (2) a Constraint-aware Loss (CL) incorporating budget and Cost-Per-Acquisition (CPA) constraints for selective learning of optimal trajectories. Extensive offline evaluations on the AuctionNet dataset demonstrate consistent performance gains (up to 3.2% over state-of-the-art method) across diverse budget settings; ablation studies verify the complementary synergy of CLB and CL, confirming C2's superiority in auto-bidding. The code for reproducing our results is available at: https://github.com/Dingjinren/C2. --- ### SoftHateBench: Evaluating Moderation Models Against Reasoning-Driven, Policy-Compliant Hostility Authors: Xuanyu Su, Diana Inkpen, Nathalie Japkowicz Published: 2026-01-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.20256v1 Online hate on social media ranges from overt slurs and threats (\emph{hard hate speech}) to \emph{soft hate speech}: discourse that appears reasonable on the surface but uses framing and value-based arguments to steer audiences toward blaming or excluding a target group. We hypothesize that current moderation systems, largely optimized for surface toxicity cues, are not robust to this reasoning-driven hostility, yet existing benchmarks do not measure this gap systematically. We introduce \textbf{\textsc{SoftHateBench}}, a generative benchmark that produces soft-hate variants while preserving the underlying hostile standpoint. To generate soft hate, we integrate the \emph{Argumentum Model of Topics} (AMT) and \emph{Relevance Theory} (RT) in a unified framework: AMT provides the backbone argument structure for rewriting an explicit hateful standpoint into a seemingly neutral discussion while preserving the stance, and RT guides generation to keep the AMT chain logically coherent. The benchmark spans \textbf{7} sociocultural domains and \textbf{28} target groups, comprising \textbf{4,745} soft-hate instances. Evaluations across encoder-based detectors, general-purpose LLMs, and safety models show a consistent drop from hard to soft tiers: systems that detect explicit hostility often fail when the same stance is conveyed through subtle, reasoning-based language. \textcolor{red}{\textbf{Disclaimer.} Contains offensive examples used solely for research.} --- ### HE-SNR: Uncovering Latent Logic via Entropy for Guiding Mid-Training on SWE-bench Authors: Yueyang Wang, Jiawei Fu, Baolong Bi, Xili Wang, Xiaoqing Liu Published: 2026-01-28 Categories: cs.LG, cs.CL, cs.SE arXiv: https://arxiv.org/abs/2601.20255v3 SWE-bench has emerged as the premier benchmark for evaluating Large Language Models on complex software engineering tasks. While these capabilities are fundamentally acquired during the mid-training phase and subsequently elicited during Supervised Fine-Tuning (SFT), there remains a critical deficit in metrics capable of guiding mid-training effectively. Standard metrics such as Perplexity (PPL) are compromised by the "Long-Context Tax" and exhibit weak correlation with downstream SWE performance. In this paper, we bridge this gap by first introducing a rigorous data filtering strategy. Crucially, we propose the Entropy Compression Hypothesis, redefining intelligence not by scalar Top-1 compression, but by the capacity to structure uncertainty into Entropy-Compressed States of low orders ("reasonable hesitation"). Grounded in this fine-grained entropy analysis, we formulate a novel metric, HE-SNR (High-Entropy Signal-to-Noise Ratio). We validate our approach on models with up to 560B parameters across different context windows (32K/128K). This work provides both the theoretical foundation and practical tools for optimizing the latent potential of LLMs in complex engineering domains. --- ### Automated Benchmark Generation from Domain Guidelines Informed by Bloom's Taxonomy Authors: Si Chen, Le Huy Khiem, Annalisa Szymanski, Ronald Metoyer, Ting Hua, Nitesh V. Chawla Published: 2026-01-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.20253v1 Open-ended question answering (QA) evaluates a model's ability to perform contextualized reasoning beyond factual recall. This challenge is especially acute in practice-based domains, where knowledge is procedural and grounded in professional judgment, while most existing LLM benchmarks depend on pre-existing human exam datasets that are often unavailable in such settings. We introduce a framework for automated benchmark generation from expert-authored guidelines informed by Bloom's Taxonomy. It converts expert practices into implicit violation-based scenarios and expands them into auto-graded multiple-choice questions (MCQs) and multi-turn dialogues across four cognitive levels, enabling deterministic, reproducible, and scalable evaluation. Applied to three applied domains: teaching, dietetics, and caregiving, we find differences between model and human-like reasoning: LLMs sometimes perform relatively better on higher-order reasoning (Analyze) but fail more frequently on lower-level items (Remember). We produce large-scale, psychometrically informed benchmarks that surface these non-intuitive model behaviors and enable evaluation of contextualized reasoning in real-world settings. --- ### Efficient Evaluation of LLM Performance with Statistical Guarantees Authors: Skyler Wu, Yash Nair, Emmanuel J. Candès Published: 2026-01-28 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2601.20251v3 Exhaustively evaluating many large language models (LLMs) on a large suite of benchmarks is expensive. We cast benchmarking as finite-population inference and, under a fixed query budget, seek tight confidence intervals (CIs) for model accuracy with valid frequentist coverage. We propose Factorized Active Querying (FAQ), which (a) leverages historical information through a Bayesian factor model; (b) adaptively selects questions using a hybrid variance-reduction/active-learning sampling policy; and (c) maintains validity through Proactive Active Inference -- a finite-population extension of active inference (Zrnic & Candès, 2024) that enables direct question selection while preserving coverage. With negligible overhead cost, FAQ delivers up to $5\times$ effective sample size gains over strong baselines on two benchmark suites, across varying historical-data missingness levels: this means that it matches the CI width of uniform sampling while using up to $5\times$ fewer queries. We release our source code and our curated datasets to support reproducible evaluation and future research. --- ### Order-Optimal Sample Complexity of Rectified Flows Authors: Hari Krishna Sahoo, Mudit Gaur, Vaneet Aggarwal Published: 2026-01-28 Categories: cs.LG, cs.AI, cs.IT, stat.ML arXiv: https://arxiv.org/abs/2601.20250v1 Recently, flow-based generative models have shown superior efficiency compared to diffusion models. In this paper, we study rectified flow models, which constrain transport trajectories to be linear from the base distribution to the data distribution. This structural restriction greatly accelerates sampling, often enabling high-quality generation with a single Euler step. Under standard assumptions on the neural network classes used to parameterize the velocity field and data distribution, we prove that rectified flows achieve sample complexity $\tilde{O}(\varepsilon^{-2})$. This improves on the best known $O(\varepsilon^{-4})$ bounds for flow matching model and matches the optimal rate for mean estimation. Our analysis exploits the particular structure of rectified flows: because the model is trained with a squared loss along linear paths, the associated hypothesis class admits a sharply controlled localized Rademacher complexity. This yields the improved, order-optimal sample complexity and provides a theoretical explanation for the strong empirical performance of rectified flow models. --- ### BLENDER: Blended Text Embeddings and Diffusion Residuals for Intra-Class Image Synthesis in Deep Metric Learning Authors: Jan Niklas Kolf, Ozan Tezcan, Justin Theiss, Hyung Jun Kim, Wentao Bao, Bhargav Bhushanam, Khushi Gupta, Arun Kejariwal, Naser Damer, Fadi Boutros Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20246v1 The rise of Deep Generative Models (DGM) has enabled the generation of high-quality synthetic data. When used to augment authentic data in Deep Metric Learning (DML), these synthetic samples enhance intra-class diversity and improve the performance of downstream DML tasks. We introduce BLenDeR, a diffusion sampling method designed to increase intra-class diversity for DML in a controllable way by leveraging set-theory inspired union and intersection operations on denoising residuals. The union operation encourages any attribute present across multiple prompts, while the intersection extracts the common direction through a principal component surrogate. These operations enable controlled synthesis of diverse attribute combinations within each class, addressing key limitations of existing generative approaches. Experiments on standard DML benchmarks demonstrate that BLenDeR consistently outperforms state-of-the-art baselines across multiple datasets and backbones. Specifically, BLenDeR achieves 3.7% increase in Recall@1 on CUB-200 and a 1.8% increase on Cars-196, compared to state-of-the-art baselines under standard experimental settings. --- ### How AI Impacts Skill Formation Authors: Judy Hanwen Shen, Alex Tamkin Published: 2026-01-28 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2601.20245v2 AI assistance produces significant productivity gains across professional domains, particularly for novice workers. Yet how this assistance affects the development of skills required to effectively supervise AI remains unclear. Novice workers who rely heavily on AI to complete unfamiliar tasks may compromise their own skill acquisition in the process. We conduct randomized experiments to study how developers gained mastery of a new asynchronous programming library with and without the assistance of AI. We find that AI use impairs conceptual understanding, code reading, and debugging abilities, without delivering significant efficiency gains on average. Participants who fully delegated coding tasks showed some productivity improvements, but at the cost of learning the library. We identify six distinct AI interaction patterns, three of which involve cognitive engagement and preserve learning outcomes even when participants receive AI assistance. Our findings suggest that AI-enhanced productivity is not a shortcut to competence and AI assistance should be carefully adopted into workflows to preserve skill formation -- particularly in safety-critical domains. --- ### MALLOC: Benchmarking the Memory-aware Long Sequence Compression for Large Sequential Recommendation Authors: Qihang Yu, Kairui Fu, Zhaocheng Du, Yuxuan Si, Kaiyuan Li, Weihao Zhao, Zhicheng Zhang, Jieming Zhu, Quanyu Dai, Zhenhua Dong, Shengyu Zhang, Kun Kuang et al. Published: 2026-01-28 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2601.20234v2 The scaling law, which indicates that model performance improves with increasing dataset and model capacity, has fueled a growing trend in expanding recommendation models in both industry and academia. However, the advent of large-scale recommenders also brings significantly higher computational costs, particularly under the long-sequence dependencies inherent in the user intent of recommendation systems. Current approaches often rely on pre-storing the intermediate states of the past behavior for each user, thereby reducing the quadratic re-computation cost for the following requests. Despite their effectiveness, these methods often treat memory merely as a medium for acceleration, without adequately considering the space overhead it introduces. This presents a critical challenge in real-world recommendation systems with billions of users, each of whom might initiate thousands of interactions and require massive memory for state storage. Fortunately, there have been several memory management strategies examined for compression in LLM, while most have not been evaluated on the recommendation task. To mitigate this gap, we introduce MALLOC, a comprehensive benchmark for memory-aware long sequence compression. MALLOC presents a comprehensive investigation and systematic classification of memory management techniques applicable to large sequential recommendations. These techniques are integrated into state-of-the-art recommenders, enabling a reproducible and accessible evaluation platform. Through extensive experiments across accuracy, efficiency, and complexity, we demonstrate the holistic reliability of MALLOC in advancing large-scale recommendation. Code is available at https://anonymous.4open.science/r/MALLOC. --- ### Visual Prompt-Agnostic Evolution Authors: Junze Wang, Lei Fan, Dezheng Zhang, Weipeng Jing, Donglin Di, Yang Song, Sidong Liu, Cong Cong Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20232v2 Visual Prompt Tuning (VPT) adapts a frozen Vision Transformer (ViT) to downstream tasks by inserting a small number of learnable prompt tokens into the token sequence at each layer. However, we observe that existing VPT variants often suffer from unstable training dynamics, characterized by gradient oscillations. A layer-wise analysis reveals that shallow-layer prompts tend to stagnate early, while deeper-layer prompts exhibit high-variance oscillations, leading to cross-layer mismatch. These issues slow convergence and degrade final performance. To address these challenges, we propose Prompt-Agnostic Evolution ($\mathtt{PAE}$), which strengthens vision prompt tuning by explicitly modeling prompt dynamics. From a frequency-domain perspective, we initialize prompts in a task-aware direction by uncovering and propagating frequency shortcut patterns that the backbone inherently exploits for recognition. To ensure coherent evolution across layers, we employ a shared Koopman operator that imposes a global linear transformation instead of uncoordinated, layer-specific updates. Finally, inspired by Lyapunov stability theory, we introduce a regularizer that constrains error amplification during evolution. Extensive experiments show that $\mathtt{PAE}$ accelerates convergence with an average $1.41\times$ speedup and improves accuracy by 1-3% on 25 datasets across multiple downstream tasks. Beyond performance, $\mathtt{PAE}$ is prompt-agnostic and lightweight, and it integrates seamlessly with diverse VPT variants without backbone modification or inference-time changes. --- ### Certificate-Guided Pruning for Stochastic Lipschitz Optimization Authors: Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma Published: 2026-01-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.20231v1 We study black-box optimization of Lipschitz functions under noisy evaluations. Existing adaptive discretization methods implicitly avoid suboptimal regions but do not provide explicit certificates of optimality or measurable progress guarantees. We introduce \textbf{Certificate-Guided Pruning (CGP)}, which maintains an explicit \emph{active set} $A_t$ of potentially optimal points via confidence-adjusted Lipschitz envelopes. Any point outside $A_t$ is certifiably suboptimal with high probability, and under a margin condition with near-optimality dimension $α$, we prove $\Vol(A_t)$ shrinks at a controlled rate yielding sample complexity $\tildeO(\varepsilon^{-(2+α)})$. We develop three extensions: CGP-Adaptive learns $L$ online with $O(\log T)$ overhead; CGP-TR scales to $d > 50$ via trust regions with local certificates; and CGP-Hybrid switches to GP refinement when local smoothness is detected. Experiments on 12 benchmarks ($d \in (2, 100)$) show CGP variants match or exceed strong baselines while providing principled stopping criteria via certificate volume. --- ### Unit-Based Agent for Semi-Cascaded Full-Duplex Dialogue Systems Authors: Haoyuan Yu, Yuxuan Chen, Minjie Cai Published: 2026-01-28 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2601.20230v2 Full-duplex voice interaction is crucial for natural human computer interaction. We present a framework that decomposes complex dialogue into minimal conversational units, enabling the system to process each unit independently and predict when to transit to the next. This framework is instantiated as a semi-cascaded full-duplex dialogue system built around a multimodal large language model, supported by auxiliary modules such as voice activity detection (VAD) and text-to-speech (TTS) synthesis. The resulting system operates in a train-free, plug-and-play manner. Experiments on the HumDial dataset demonstrate the effectiveness of our framework, which ranks second among all teams on the test set of the Human-like Spoken Dialogue Systems Challenge (Track 2: Full-Duplex Interaction). Code is available at the GitHub repository https://github.com/yu-haoyuan/fd-badcat. --- ### Proactive SFC Provisioning with Forecast-Driven DRL in Data Centers Authors: Parisa Fard Moshiri, Poonam Lohan, Burak Kantarci, Emil Janulewicz Published: 2026-01-28 Categories: cs.NI, cs.LG arXiv: https://arxiv.org/abs/2601.20229v1 Service Function Chaining (SFC) requires efficient placement of Virtual Network Functions (VNFs) to satisfy diverse service requirements while maintaining high resource utilization in Data Centers (DCs). Conventional static resource allocation often leads to overprovisioning or underprovisioning due to the dynamic nature of traffic loads and application demands. To address this challenge, we propose a hybrid forecast-driven Deep reinforcement learning (DRL) framework that combines predictive intelligence with SFC provisioning. Specifically, we leverage DRL to generate datasets capturing DC resource utilization and service demands, which are then used to train deep learning forecasting models. Using Optuna-based hyperparameter optimization, the best-performing models, Spatio-Temporal Graph Neural Network, Temporal Graph Neural Network, and Long Short-Term Memory, are combined into an ensemble to enhance stability and accuracy. The ensemble predictions are integrated into the DC selection process, enabling proactive placement decisions that consider both current and future resource availability. Experimental results demonstrate that the proposed method not only sustains high acceptance ratios for resource-intensive services such as Cloud Gaming and VoIP but also significantly improves acceptance ratios for latency-critical categories such as Augmented Reality increases from 30% to 50%, while Industry 4.0 improves from 30% to 45%. Consequently, the prediction-based model achieves significantly lower E2E latencies of 20.5%, 23.8%, and 34.8% reductions for VoIP, Video Streaming, and Cloud Gaming, respectively. This strategy ensures more balanced resource allocation, and reduces contention. --- ### Quantum statistics from classical simulations via generative Gibbs sampling Authors: Weizhou Wang, Xuanxi Zhang, Jonathan Weare, Aaron R. Dinner Published: 2026-01-28 Categories: physics.chem-ph, cond-mat.stat-mech, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2601.20228v1 Accurate simulation of nuclear quantum effects is essential for molecular modeling but expensive using path integral molecular dynamics (PIMD). We present GG-PI, a ring-polymer-based framework that combines generative modeling of the single-bead conditional density with Gibbs sampling to recover quantum statistics from classical simulation data. GG-PI uses inexpensive standard classical simulations or existing data for training and allows transfer across temperatures without retraining. On standard test systems, GG-PI significantly reduces wall clock time compared to PIMD. Our approach extends easily to a wide range of problems with similar Markov structure. --- ### ProFlow: Zero-Shot Physics-Consistent Sampling via Proximal Flow Guidance Authors: Zichao Yu, Ming Li, Wenyi Zhang, Difan Zou, Weiguo Gao Published: 2026-01-28 Categories: cs.LG, cs.AI, math.NA arXiv: https://arxiv.org/abs/2601.20227v1 Inferring physical fields from sparse observations while strictly satisfying partial differential equations (PDEs) is a fundamental challenge in computational physics. Recently, deep generative models offer powerful data-driven priors for such inverse problems, yet existing methods struggle to enforce hard physical constraints without costly retraining or disrupting the learned generative prior. Consequently, there is a critical need for a sampling mechanism that can reconcile strict physical consistency and observational fidelity with the statistical structure of the pre-trained prior. To this end, we present ProFlow, a proximal guidance framework for zero-shot physics-consistent sampling, defined as inferring solutions from sparse observations using a fixed generative prior without task-specific retraining. The algorithm employs a rigorous two-step scheme that alternates between: (\romannumeral1) a terminal optimization step, which projects the flow prediction onto the intersection of the physically and observationally consistent sets via proximal minimization; and (\romannumeral2) an interpolation step, which maps the refined state back to the generative trajectory to maintain consistency with the learned flow probability path. This procedure admits a Bayesian interpretation as a sequence of local maximum a posteriori (MAP) updates. Comprehensive benchmarks on Poisson, Helmholtz, Darcy, and viscous Burgers' equations demonstrate that ProFlow achieves superior physical and observational consistency, as well as more accurate distributional statistics, compared to state-of-the-art diffusion- and flow-based baselines. --- ### Parametric and Generative Forecasts of EPEX Day-Ahead Energy Market Curves Authors: Julian Gutierrez, Redouane Silvente Published: 2026-01-28 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2601.20226v3 We propose two methodologies for modelling aggregated supply and demand curves in the EPEX SPOT Day-Ahead market, emphasizing generative models as a way to recover distributional variability. The first is a low-dimensional parametric representation that yields deterministic point forecasts; the second is a high-dimensional order-level representation that samples from a conditional distribution of plausible curves. Both model the full curve structure, enabling the analysis of price sensitivity, volume sensitivity, and price impact. The parametric representation uses plateau levels, elastic-region boundaries, and polynomial coefficients, forecast with eXtreme Gradient Boosting. The main contribution is the generative representation, which uses price arrivals and volume-increment marks and is implemented with conditional Denoising Diffusion Probabilistic Models. Using French EPEX data from 2021 to 2024, we evaluate both approaches through curve reconstruction and a price-maker storage optimization problem. The parametric implementation provides a deterministic reference, while the diffusion-based implementation produces distributions of plausible curves and achieves higher realized profits and smaller gaps to an oracle benchmark in the storage application. --- ### Feature Projection Learning for Better Vision-Language Reasoning Authors: Yi Zhang, Weicheng Lin, Liang-Jie Zhang Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20224v1 Vision-Language Pre-Trained models, notably CLIP, that utilize contrastive learning have proven highly adept at extracting generalizable visual features. To inherit the well-learned knowledge of VLP models for downstream tasks, several approaches aim to adapt them efficiently with limited supervision. However, these methods either suffer from limited performance, excessive learnable parameters, or extended training times, all of which hinder their effectiveness in adapting the CLIP model to downstream tasks. In this work, we propose a simple yet efficient and effective method called \textit{\textbf{F}eature \textbf{P}rojection \textbf{L}earning(FPL)} to address these problems. Specifically, we develop a projection model that projects class prototype features into the query image feature space and reconstructs the query image feature map. The negative average squared reconstruction error is used as the class score. In this way, we transform the classification problem into a feature projection problem. The final output of this method is a combination of the prediction from the projection model and the original pre-trained CLIP. Comprehensive empirical evaluations confirm that FPL delivers superior accuracy, surpassing the current state-of-the-art methods by a substantial margin. --- ### Scaling Medical Reasoning Verification via Tool-Integrated Reinforcement Learning Authors: Hang Zhang, Ruheng Wang, Yuelyu Ji, Mingu Kwak, Xizhi Wu, Chenyu Li, Li Zhang, Wenqi Shi, Yifan Peng, Yanshan Wang Published: 2026-01-28 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.20221v1 Large language models have achieved strong performance on medical reasoning benchmarks, yet their deployment in clinical settings demands rigorous verification to ensure factual accuracy. While reward models offer a scalable approach for reasoning trace verification, existing methods face two limitations: they produce only scalar reward values without explicit justification, and they rely on single-pass retrieval that precludes adaptive knowledge access as verification unfolds. We introduce $\method$, an agentic framework that addresses these limitations by training medical reasoning verifiers to iteratively query external medical corpora during evaluation. Our approach combines tool-augmented verification with an iterative reinforcement learning paradigm that requires only trace-level supervision, alongside an adaptive curriculum mechanism that dynamically adjusts training data distribution. Across four medical reasoning benchmarks, $\method$ achieves substantial gains over existing methods, improving MedQA accuracy by 23.5% and MedXpertQA by 32.0% relative to the base generator in particular. Crucially, $\method$ demonstrates an $\mathbf{8\times}$ reduction in sampling budget requirement compared to prior reward model baselines. These findings establish that grounding verification in dynamically retrieved evidence offers a principled path toward more reliable medical reasoning systems. --- ### DenseGRPO: From Sparse to Dense Reward for Flow Matching Model Alignment Authors: Haoyou Deng, Keyu Yan, Chaojie Mao, Xiang Wang, Yu Liu, Changxin Gao, Nong Sang Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20218v2 Recent GRPO-based approaches built on flow matching models have shown remarkable improvements in human preference alignment for text-to-image generation. Nevertheless, they still suffer from the sparse reward problem: the terminal reward of the entire denoising trajectory is applied to all intermediate steps, resulting in a mismatch between the global feedback signals and the exact fine-grained contributions at intermediate denoising steps. To address this issue, we introduce \textbf{DenseGRPO}, a novel framework that aligns human preference with dense rewards, which evaluates the fine-grained contribution of each denoising step. Specifically, our approach includes two key components: (1) we propose to predict the step-wise reward gain as dense reward of each denoising step, which applies a reward model on the intermediate clean images via an ODE-based approach. This manner ensures an alignment between feedback signals and the contributions of individual steps, facilitating effective training; and (2) based on the estimated dense rewards, a mismatch drawback between the uniform exploration setting and the time-varying noise intensity in existing GRPO-based methods is revealed, leading to an inappropriate exploration space. Thus, we propose a reward-aware scheme to calibrate the exploration space by adaptively adjusting a timestep-specific stochasticity injection in the SDE sampler, ensuring a suitable exploration space at all timesteps. Extensive experiments on multiple standard benchmarks demonstrate the effectiveness of the proposed DenseGRPO and highlight the critical role of the valid dense rewards in flow matching model alignment. --- ### An Accounting Identity for Algorithmic Fairness Authors: Hadi Elzayn, Jacob Goldin Published: 2026-01-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.20217v1 We derive an accounting identity for predictive models that links accuracy with common fairness criteria. The identity shows that for globally calibrated models, the weighted sums of miscalibration within groups and error imbalance across groups is equal to a "total unfairness budget." For binary outcomes, this budget is the model's mean-squared error times the difference in group prevalence across outcome classes. The identity nests standard impossibility results as special cases, while also describing inherent tradeoffs when one or more fairness measures are not perfectly satisfied. The results suggest that accuracy and fairness are best viewed as complements in binary prediction tasks: increasing accuracy necessarily shrinks the total unfairness budget and vice-versa. Experiments on benchmark data confirm the theory and show that many fairness interventions largely substitute between fairness violations, and when they reduce accuracy they tend to expand the total unfairness budget. The results extend naturally to prediction tasks with non-binary outcomes, illustrating how additional outcome information can relax fairness incompatibilities and identifying conditions under which the binary-style impossibility does and does not extend to regression tasks. --- ### Spark: Strategic Policy-Aware Exploration via Dynamic Branching for Long-Horizon Agentic Learning Authors: Jinyang Wu, Shuo Yang, Changpeng Yang, Yuhao Shen, Shuai Zhang, Zhengqi Wen, Jianhua Tao Published: 2026-01-28 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2601.20209v2 Reinforcement learning has empowered large language models to act as intelligent agents, yet training them for long-horizon tasks remains challenging due to the scarcity of high-quality trajectories, especially under limited resources. Existing methods typically scale up rollout sizes and indiscriminately allocate computational resources among intermediate steps. Such attempts inherently waste substantial computation budget on trivial steps while failing to guarantee sample quality. To address this, we propose \textbf{Spark} (\textbf{S}trategic \textbf{P}olicy-\textbf{A}ware explo\textbf{R}ation via \textbf{K}ey-state dynamic branching), a novel framework that selectively branches at critical decision states for resource-efficient exploration. Our key insight is to activate adaptive branching exploration at critical decision points to probe promising trajectories, thereby achieving precise resource allocation that prioritizes sampling quality over blind coverage. This design leverages the agent's intrinsic decision-making signals to reduce dependence on human priors, enabling the agent to autonomously expand exploration and achieve stronger generalization. Experiments across diverse tasks (e.g., embodied planning), demonstrate that \textsc{Spark} achieves superior success rates with significantly fewer training samples, exhibiting robust generalization even in unseen scenarios. Our code and checkpoints are available at https://github.com/jinyangwu/SPARK. --- ### TRACER: Texture-Robust Affordance Chain-of-Thought for Deformable-Object Refinement Authors: Wanjun Jia, Kang Li, Fan Yang, Mengfei Duan, Wenrui Chen, Yiming Jiang, Hui Zhang, Kailun Yang, Zhiyong Li, Yaonan Wang Published: 2026-01-28 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2601.20208v1 The central challenge in robotic manipulation of deformable objects lies in aligning high-level semantic instructions with physical interaction points under complex appearance and texture variations. Due to near-infinite degrees of freedom, complex dynamics, and heterogeneous patterns, existing vision-based affordance prediction methods often suffer from boundary overflow and fragmented functional regions. To address these issues, we propose TRACER, a Texture-Robust Affordance Chain-of-thought with dEformable-object Refinement framework, which establishes a cross-hierarchical mapping from hierarchical semantic reasoning to appearance-robust and physically consistent functional region refinement. Specifically, a Tree-structured Affordance Chain-of-Thought (TA-CoT) is formulated to decompose high-level task intentions into hierarchical sub-task semantics, providing consistent guidance across various execution stages. To ensure spatial integrity, a Spatial-Constrained Boundary Refinement (SCBR) mechanism is introduced to suppress prediction spillover, guiding the perceptual response to converge toward authentic interaction manifolds. Furthermore, an Interactive Convergence Refinement Flow (ICRF) is developed to aggregate discrete pixels corrupted by appearance noise, significantly enhancing the spatial continuity and physical plausibility of the identified functional regions. Extensive experiments conducted on the Fine-AGDDO15 dataset and a real-world robotic platform demonstrate that TRACER significantly improves affordance grounding precision across diverse textures and patterns inherent to deformable objects. More importantly, it enhances the success rate of long-horizon tasks, effectively bridging the gap between high-level semantic reasoning and low-level physical execution. The source code and dataset will be made publicly available at https://github.com/Dikay1/TRACER. --- ### Towards Intelligent Urban Park Development Monitoring: LLM Agents for Multi-Modal Information Fusion and Analysis Authors: Zixuan Xiao, Chunguang Hu, Jun Ma Published: 2026-01-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.20206v1 As an important part of urbanization, the development monitoring of newly constructed parks is of great significance for evaluating the effect of urban planning and optimizing resource allocation. However, traditional change detection methods based on remote sensing imagery have obvious limitations in high-level and intelligent analysis, and thus are difficult to meet the requirements of current urban planning and management. In face of the growing demand for complex multi-modal data analysis in urban park development monitoring, these methods often fail to provide flexible analysis capabilities for diverse application scenarios. This study proposes a multi-modal LLM agent framework, which aims to make full use of the semantic understanding and reasoning capabilities of LLM to meet the challenges in urban park development monitoring. In this framework, a general horizontal and vertical data alignment mechanism is designed to ensure the consistency and effective tracking of multi-modal data. At the same time, a specific toolkit is constructed to alleviate the hallucination issues of LLM due to the lack of domain-specific knowledge. Compared to vanilla GPT-4o and other agents, our approach enables robust multi-modal information fusion and analysis, offering reliable and scalable solutions tailored to the diverse and evolving demands of urban park development monitoring. --- ### Hyperparameter Transfer with Mixture-of-Expert Layers Authors: Tianze Jiang, Blake Bordelon, Cengiz Pehlevan, Boris Hanin Published: 2026-01-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.20205v3 Mixture-of-Experts (MoE) layers have emerged as an important tool in scaling up modern neural networks by decoupling total trainable parameters from activated parameters in the forward pass for each token. However, sparse MoEs add complexity to training due to (i) new trainable parameters (router weights) that, like all other parameter groups, require hyperparameter (HP) tuning; (ii) new architecture scale dimensions (number of and size of experts) that must be chosen and potentially taken large. To make HP selection cheap and reliable, we propose a new parameterization for transformer models with MoE layers when scaling model width, depth, number of experts, and expert (hidden) size. Our parameterization is justified by a novel dynamical mean-field theory (DMFT) analysis. When varying different model dimensions trained at a fixed token budget, we find empirically that our parameterization enables reliable HP transfer across models from 51M to over 2B total parameters. We further take HPs identified from sweeping small models on a short token horizon to train larger models on longer horizons and report performant model behaviors. --- ### Minimum-Cost Network Flow with Dual Predictions Authors: Zhiyang Chen, Hailong Yao, Xia Yin Published: 2026-01-28 Categories: cs.LG, cs.DS arXiv: https://arxiv.org/abs/2601.20203v1 Recent work has shown that machine-learned predictions can provably improve the performance of classic algorithms. In this work, we propose the first minimum-cost network flow algorithm augmented with a dual prediction. Our method is based on a classic minimum-cost flow algorithm, namely $\varepsilon$-relaxation. We provide time complexity bounds in terms of the infinity norm prediction error, which is both consistent and robust. We also prove sample complexity bounds for PAC-learning the prediction. We empirically validate our theoretical results on two applications of minimum-cost flow, i.e., traffic networks and chip escape routing, in which we learn a fixed prediction, and a feature-based neural network model to infer the prediction, respectively. Experimental results illustrate $12.74\times$ and $1.64\times$ average speedup on two applications. --- ### DeRaDiff: Denoising Time Realignment of Diffusion Models Authors: Ratnavibusena Don Shahain Manujith, Teoh Tze Tzun, Kenji Kawaguchi, Yang Zhang Published: 2026-01-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.20198v2 Recent advances align diffusion models with human preferences to increase aesthetic appeal and mitigate artifacts and biases. Such methods aim to maximize a conditional output distribution aligned with higher rewards whilst not drifting far from a pretrained prior. This is commonly enforced by KL (Kullback Leibler) regularization. As such, a central issue still remains: how does one choose the right regularization strength? Too high of a strength leads to limited alignment and too low of a strength leads to "reward hacking". This renders the task of choosing the correct regularization strength highly non-trivial. Existing approaches sweep over this hyperparameter by aligning a pretrained model at multiple regularization strengths and then choose the best strength. Unfortunately, this is prohibitively expensive. We introduce DeRaDiff, a denoising time realignment procedure that, after aligning a pretrained model once, modulates the regularization strength during sampling to emulate models trained at other regularization strengths without any additional training or finetuning. Extending decoding-time realignment from language to diffusion models, DeRaDiff operates over iterative predictions of continuous latents by replacing the reverse step reference distribution by a geometric mixture of an aligned and reference posterior, thus giving rise to a closed form update under common schedulers and a single tunable parameter, lambda, for on the fly control. Our experiments show that across multiple text image alignment and image-quality metrics, our method consistently provides a strong approximation for models aligned entirely from scratch at different regularization strengths. Thus, our method yields an efficient way to search for the optimal strength, eliminating the need for expensive alignment sweeps and thereby substantially reducing computational costs. --- ### Bias-Reduced Estimation of Finite Mixtures: An Application to Latent Group Structures in Panel Data Authors: Raphaël Langevin Published: 2026-01-28 Categories: stat.ME, cs.LG, econ.EM, stat.CO arXiv: https://arxiv.org/abs/2601.20197v2 Finite mixture models are widely used in econometric analyses to capture unobserved heterogeneity. This paper shows that maximum likelihood estimation of finite mixtures of parametric densities can suffer from substantial finite-sample bias in all parameters under mild regularity conditions. The bias arises from the influence of outliers in component densities with unbounded or large support and increases with the degree of overlap among mixture components. I show that maximizing the classification-mixture likelihood function, equipped with a consistent classifier, yields parameter estimates that are less biased than those obtained by standard maximum likelihood estimation (MLE). I then derive the asymptotic distribution of the resulting estimator and provide conditions under which oracle efficiency is achieved. Monte Carlo simulations show that conventional mixture MLE exhibits pronounced finite-sample bias, which diminishes as the sample size or the statistical distance between component densities tends to infinity. The simulations further show that the proposed estimation strategy generally outperforms standard MLE in finite samples in terms of both bias and mean squared errors under relatively weak assumptions. An empirical application to latent group panel structures using health administrative data shows that the proposed approach reduces out-of-sample prediction error by approximately 17.6% relative to the best results obtained from standard MLE procedures. --- ### Automated Marine Biofouling Assessment: Benchmarking Computer Vision and Multimodal LLMs on the Level of Fouling Scale Authors: Brayden Hamilton, Tim Cashmore, Peter Driscoll, Trevor Gee, Henry Williams Published: 2026-01-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.20196v1 Marine biofouling on vessel hulls poses major ecological, economic, and biosecurity risks. Traditional survey methods rely on diver inspections, which are hazardous and limited in scalability. This work investigates automated classification of biofouling severity on the Level of Fouling (LoF) scale using both custom computer vision models and large multimodal language models (LLMs). Convolutional neural networks, transformer-based segmentation, and zero-shot LLMs were evaluated on an expert-labelled dataset from the New Zealand Ministry for Primary Industries. Computer vision models showed high accuracy at extreme LoF categories but struggled with intermediate levels due to dataset imbalance and image framing. LLMs, guided by structured prompts and retrieval, achieved competitive performance without training and provided interpretable outputs. The results demonstrate complementary strengths across approaches and suggest that hybrid methods integrating segmentation coverage with LLM reasoning offer a promising pathway toward scalable and interpretable biofouling assessment. ---