Papers

Zero-shot Neural Architecture Search removes the prohibitive cost of traditional NAS, but its search process is typically based on the evolutionary algorithm (EA); lacking an explicit model of the objective, it often resorts to a near-random search through mutation. Bayesian Optimization offers a principled alternative by modeling the objective and aggregating information across iterations, but scales poorly to the high-dimensional, discrete, graph-structured spaces of modern NAS, restricting its use to only small networks. In this paper, we bring Bayesian Optimization to zero-shot NAS for large-scale architectures by learning a latent space via a Variational Autoencoder trained to reconstruct a novel prefix encoding of architectures and propose a proxy scalarization that combines several zero-shot proxies into a single Bayesian Optimization objective. After only 10,000 iterations of the proposed search algorithm (8 hours on a single GPU), our method found a network architecture which under the given model parameter count constraints achieves state-of-the-art results on three separate tasks – image classification, object detection and semantic segmentation.

Vision-language models (VLMs) achieve strong zero-shot transferability but remain vulnerable to target-domain shifts at inference time. Test-time adaptation (TTA) offers a practical remedy, yet most existing VLM-TTA methods follow a prediction-side adaptation paradigm. They use test samples to adjust logits, prototypes, caches, priors, or feature statistics, often incurring additional computational overhead. In this paper, we take a different perspective and reframe VLM-TTA as candidate verification rather than prediction adjustment. We propose Test-Time Correction (TTC), a hypothesis-based correction framework guided by a simple principle: hypothesize, reconstruct, correct. Given a test feature and its top-k candidate labels, TTC treats each candidate label as a hypothesis, reconstructs the feature within the corresponding latent subspace stored in a memory bank, and measures the resulting divergence shift. This shift quantifies how much the candidate subspace and its relations to other candidates change after the hypothetical insertion of the test feature. A correct candidate hypothesis induces only a small shift, whereas an incorrect one perturbs the subspace more strongly. TTC therefore corrects the prediction by selecting the candidate with the minimum aggregated divergence shift. This training-free candidate-verification mechanism avoids iterative optimization and provides a favorable accuracy-efficiency trade-off. Across five TTA settings and 15 benchmark datasets, including zero-shot classification, domain generalization, few-shot classification, base-to-novel generalization, and cross-dataset evaluation, TTC consistently improves accuracy over state-of-the-art VLM-TTA methods while achieving up to 2x speedup, over 3x lower CPU memory usage, and up to 1.4x lower GPU memory usage than the lowest-memory training-free baseline.

Accurately quantifying the predictive uncertainty or improving model calibration plays an important role in medical image classification, in particular in melanoma diagnosis, where accurate uncertainty quantification can have significant implications for patient care. One of the methods for calibration improvement is data augmentation. In addition, data augmentation as a method for synthetically increasing the size of the dataset has been proven to improve the performance of models trained on imbalanced datasets. However, the impact of data augmentation, as a transformation of a part of the original data, on calibration of models trained on imbalanced datasets, in particular in melanoma classification is under-explored. We train neural networks on SIIM-ISIC 2020 melanoma classification dataset under two conditions: with and without data augmentation, and compare the differences in AUC and expected calibration error (ECE) in both scenarios. Our results shows improvements in uncertainty calibration using different augmentation methods.

Melanoma is the deadliest type of skin cancer, whose early diagnosis is crucial for patients' survival. Image classification using deep learning models has shown promising results for melanoma diagnosis. However, the performance of these models on the melanoma datasets such as SIIM-ISIC melanoma classification dataset is a challenge due to the class imbalance. One of the methods to deal with this challenge is using loss function modifications. In this work, we have investigated the effect of different loss functions on the performance of deep neural networks. We trained these networks using focal loss, logit-adjusted softmax cross-entropy (CE) loss, and weighted softmax CE loss, and we report different metrics for evaluating performance and uncertainty calibration. Our results suggest that focal loss delivers a good combination of performance in terms of AUC and uncertainty calibration in terms of expected calibration error (ECE) simultaneously.

Engineering design often starts from a 2D sketch that fixes style and proportions, yet the subsequent 3D shape optimization relies on learned generative priors to keep the geometry valid. However, these priors are agnostic to the sketch: while they admit a valid design by correcting a drifted proposal back to its training distribution, they often correct it towards the high-density region, ignoring the specified design. We introduce AnchorGen, a rectified-flow framework trained unconditionally on the concatenated shape and sketch latents of paired data. The learned manifold represents the joint distribution of shape-sketch pairs, so constraining the sketch component restricts the iterate to the sub-manifold of shapes consistent with a target style. Since training employs no conditioning signal, the constraint is imposed at inference: gradient descent optimizes the shape latent to minimize a differentiable drag surrogate, while constraining the sketch latent to remain close to the target sketch via a token-wise cosine penalty. A single model thereby supports design-preserving optimization, dimensionally explicit design edits, and sketch-only synthesis.

We investigate CLIP for zero-shot gender estimation from full-face and periocular images. Three CLIP backbones are evaluated on 11,299 frontal images from Adience using image-text similarity with male/female prompts, achieving 95.54% full-face accuracy without task-specific training. For periocular, zero-shot predictions are strongly biased towards males, primarily due to a misaligned decision boundary. Threshold alignment substantially reduces this bias, reaching 85.29% accuracy. Linear SVMs trained on CLIP features provide only marginal gains, with a best periocular accuracy of 86.17%, approximately 2.8% above previous Adience results in the literature. Nevertheless, the gap with full-face performance confirms the greater difficulty of periocular gender estimation

Cone-beam computed tomography (CBCT) is widely used in image-guided radiotherapy, but scatter, beam hardening, noise, truncation, and other artifacts limit image quality and CT number accuracy. Paired CBCT and CT data are difficult to obtain clinically because of motion, anatomical changes, and acquisition mismatch. We present RefineCBCT, an unpaired CBCT refinement framework that uses pseudo-label guidance and short-step diffusion to reduce artifacts while preserving patient-specific anatomy. RefineCBCT was trained and evaluated on unpaired CBCT and planning CT data from public LUNG TCIA and PELVIC TCIA datasets and compared with representative GAN and diffusion based methods. On LUNG TCIA, it achieved the best results across all metrics, with MAE 19.411, RMSE 62.758, PSNR 30.845 dB, and SSIM 0.931. On PELVIC TCIA, it achieved the best MAE, PSNR, and SSIM, with values of 14.905, 36.671 dB, and 0.876. The refined images showed fewer streaking and shading artifacts, clearer anatomical boundaries, and improved soft tissue uniformity, with line profile and ROI analyses showing closer agreement with planning CT. These results suggest that RefineCBCT provides efficient and effective CBCT refinement under clinically realistic unpaired training conditions and may support more reliable CBCT use in image-guided radiotherapy workflows. Code is publicly available on GitHub, and the evaluated datasets are available from The Cancer Imaging Archive.

Semantic segmentation of street panoramas can support detailed descriptions of urban environments, yet small datasets and unequal training costs make model selection difficult. This paper presents the system used for a first place submission to the PalmCity challenge in the leaderboard snapshot dated 5 October 2026. Nine pretrained segmentation systems are compared using approximately equal computation budgets. The candidates include DeepLabV3+, SegFormer, UPerNet, Mask2Former, DINOv3 with a linear decoder, and an Encoder only Mask Transformer using DINOv3. The two leading candidates are trained independently with three random seeds and longer budgets. Equal averaging of class probabilities from the three Encoder only Mask Transformer models, evaluated at three image scales with horizontal reflection, produces 60.95% mean intersection over union and 71.16% mean F1 on the 84 image public validation split. The submitted predictions receive 57.08% mean intersection over union and 67.96% mean F1 on the hidden test leaderboard. Producing all 249 test masks takes 251.49 seconds including model initialization and provenance checks on one NVIDIA RTX 5090. Peak allocated GPU memory is 2.70 GiB. The study reports all eligible models, all inference variants, class level errors, source conditions, and reproducibility checks, providing a documented challenge workflow with existing architectures.

Large Vision-Language Models (LVLMs) frequently suffer from object hallucination. Existing training-free interventions primarily manipulate attention weights, which indirectly affect the deep semantics reaching the final predictive layers. In this work, we shift our focus to the hidden state vectors extracted after self-attention and residual addition. Empirical analysis reveals that hallucinated tokens do not simply over-rely on linguistic priors; instead, they exhibit an anomalous contextual deviation, showing significantly lower similarities to both textual and visual contexts in intermediate layers. Motivated by this, we propose ROT, a layer-specific, training-free framework. ROT dynamically detects semantic deviation in the middle layers and applies a norm-preserving rotation to steer the hidden states back toward the local multimodal context plane spanned by the contexts. For subsequent layers, a representational smoothing mechanism is introduced to stabilize the calibrated trajectory. Extensive experiments on multiple benchmarks demonstrate that ROT consistently reduces hallucinations across various model architectures and scales, offering an efficient, geometry-driven solution for grounded generation.

Three-dimensional (3D) left ventricular (LV) reconstruction from sparse cardiac magnetic resonance (CMR) imaging remains challenging due to inter-slice misalignment and insufficient local spatial information between slices. Global aggregation of contour features may obscure local contour-to-surface relationships. We propose Local2Mesh, a spatially localized contour-to-mesh framework that deforms a template mesh to reconstruct 3D LV geometry from sparse 2D contours without 3D mesh annotations. The framework introduces geometry-aware alignment to correct inter-slice misalignment and a plane-aware Local Router that routes contour features to template vertices using vertex-to-plane distances. Local and global contour features then jointly guide graph-based template deformation for 3D LV reconstruction. Experiments on two public datasets, M&Ms-2 and ACDC, demonstrate superior geometric reconstruction and functional estimation over existing methods. Zero-shot transfer from M&Ms-2 to ACDC demonstrates strong cross-dataset generalization. Reconstructed meshes also improve disease classification over sparse contours, supporting their utility for downstream cardiac analysis. These results demonstrate that combining geometry-aware alignment with local contour-to-vertex modeling improves LV reconstruction from sparse 2D contours and supports downstream cardiac analysis. The code is available at https://github.com/hwu918945-alt/loca2mesh.

Deep learning has advanced chest X-ray (CXR) diagnosis, yet demographic biases in learned representations may contribute to performance disparities across intersectional groups. We propose a single-encoder framework combining dual-level decorrelation with prototype-guided cross-group contrastive learning to reduce demographic dependence while accounting for within-class variation. We further propose Demographic Representation Alignment Reduction (DRAR), a new metric that quantifies the reduction in demographic structure within disease representations. The framework is evaluated on four classification tasks using 34,809 CheXpert test images across eight intersectional groups, defined by age, sex and ethnicity. Compared with empirical risk minimization (ERM), our method reduces the mean equalized-odds gap from 15.41% to 10.86% and the AUC gap from 5.95% to 5.01%. Our method achieves a DRAR of 59.04% relative to ERM, with only a slight decrease in mean AUC. These results demonstrate that representation disentanglement can reduce demographic bias and improve intersectional fairness. Code is available at \url{https://github.com/06Yujie/Fair-Medical-Imaging}.

Recovering geometry, materials, and lighting from photographs is highly ambiguous when only static illumination is available. Active-lighting setups reduce the ambiguity but require dark rooms or specialized hardware. Instead, we synergize both static and flash lighting from casual indoor capture, with the flash on or off, each from independent viewpoints. The flash residual constrains albedo and the BRDF, while static lighting captures grazing-angle specular highlights that flash misses. With a 2DGS reconstruction framing, our key contribution is a GS-anchored diffuse field: a hash-encoded MLP is queried at the rasterized 2DGS depth. As it depends only on world position, it is view consistent in 3D and allows the flash residual to drive material decomposition instead of being absorbed by alpha-blending drift across views. At the same time, we render static lighting with deferred shading such that it can also supervise material decomposition. On five synthetic and three real indoor scenes, our method outperforms six recent baselines on diffuse color, albedo and roughness material parameters, and in relighting where PSNR improves by 4.17 dB over the next-best baseline.

We consider the fusion of lung ultrasound images with routinely-collected clinical and demographic data for the purpose of automated tuberculosis (TB) screening using deep-learning. Such deep-learning based screening tools for TB could meaningfully support the health care system in Africa, where the burden of disease is severe and resources are constrained. Beginning with an established ResNet baseline for classification of lung ultrasound images, which achieves an area under the receiver operating characteristic (AUROC) curve of 0.91 [0.86,0.96] (95% CI), we consider the incorporation of the clinical and demographic data using three fusion approaches. We find that a simple average-based fusion of the output scores of separately-trained image and clinical data classifiers consistently matches or outperforms a more complex approach where the data is fused earlier and a combined classifier is trained. Fusing the image and the clinical classifiers in this way leads to a classifier with an overall AUROC of 0.95 [0.91,0.99] (specificity of 0.76 at sensitivity 0.93) which is an improvement of 4% absolute over the image-only baseline. We also find that greedy feature selection can be used to reduce the number of clinical and demographic inputs without sacrificing classification performance. Finally, when we differentiate between clinical and demographic data that are self-reported, that require some basic measurement or calculation, and that require a point-of-care (POC) test, we find the inclusion of the POC tests included in this study to be of minimal benefit to classification performance. We conclude that the incorporation of routinely-collected clinical and demographic data is a promising way to improve the performance of lung ultrasound based automatic classification.

Image editing should change only the attributes specified by an instruction while preserving everything else, yet current methods often make unintended changes. We treat this minimal-change principle as an optimization objective for instruction-based editing. Latent L1 regularization is a poor proxy for output locality in modern nonlinear generators and often requires supervision unavailable at scale. We instead optimize edit outcomes with reinforcement learning. An agentic vision-language reward model audits each source image, instruction, and edited image for two failure types: unimplemented requested changes and unintended changes. A group-level rubric merges and verifies these issues to provide consistent rewards across candidate edits without per-instruction human annotations. On FLUX.1 Kontext-dev, ARRO raises average EditScore from 5.21 to 5.88 across MinEval, MagicBrush, AnyBench, and Emu-Edit. On 600 evaluation examples, it reduces off-target pixel change by 8.4% relative to the base editor. Reward and SFT controls, blinded human evaluations, and transfer to OmniGen2 provide complementary evidence. Code: https://github.com/Showwwwwwwww/ARRO

Volume electron microscopy (vEM) has emerged as an essential sensing technique in biomedical research, allowing the three-dimensional imaging of biological cells and tissues at nanometer-scale resolution. The ability to generate extensive datasets has reached the limitations of downstream analysis processes, which depend significantly on the intervention of human experts for preprocessing and annotation. We propose an efficient and reliable patch-based retrieval framework based on self-supervised learning of local image descriptors to locate self-similar structures in vEM datasets. Given a few manual annotations of a given cellular structure, our method can retrieve similar structures across the EM volume. Our framework is interactive, allowing the human expert to refine the search queries and retrieve relevant image patches quickly and using little labeled data. Experiments on real-world vEM images of biological tissues demonstrate that our framework can reliably identify relevant cellular structures, generalize across different organelles and acquisition modalities, and substantially reduce the search space for downstream analysis.

Spatio-temporal video grounding (STVG) aims to localize objects or events described by natural language queries in both space and time. Existing STVG models are typically trained and evaluated under the assumption that each query is relevant to the input video. In this work, we challenge this assumption by studying the behavior of state-of-the-art STVG models under irrelevant queries and missing textual input. Our experiments show that current models can still produce plausible spatio-temporal predictions even when the query is unrelated to the video or removed entirely. We further analyze HCSTVG-v2 and VidSTG to identify dataset regularities that may encourage such query-insensitive behavior. Our study highlights an underexplored limitation of STVG models and motivates negative-aware evaluation protocols and architectures that explicitly assess query relevance.

Ultrasound is widely used, but acquisition quality is heavily dependent on the operator's knowledge and expertise. With demand for examinations outpacing the supply of trained sonographers, operator-guidance systems aim to close this gap by instructing a less trained user how to move the probe toward a target view. In this paper, we propose a retrieval-induced latent transition model for ultrasound acquisition dynamics, formulating ultrasound operator guidance as multi-step planning and retrieval in a world model. Using a V-JEPA 2.1 backbone, observations are first encoded into a latent space where anatomically related views lie close together. We then retrieve similar views from a reference database containing encoded latent states and corresponding probe positions and orientations. Rather than learning a parametric transition function, we directly use physically executed transitions from the database to establish our nonparametric, retrieval-induced transition model that supports receding-horizon planning. At deployment, guidance is generated from the live ultrasound image feed alone, without any probe tracking hardware. Applied to carotid ultrasound, the proposed planner reaches the target view in 86% of retrospective closed-loop episodes, versus 52% and 43% for representative baselines, outperforming both on every target view, including the challenging longitudinal internal and external carotid artery views. A prospective feasibility study on unseen volunteers, run in real time on a CPU using distillation, reaches 83% target-view reachability. Because planning is driven by proximity to any encodable goal latent, the same world model can navigate back to any previously acquired, patient-specific frame, supporting reproducible longitudinal imaging for e.g. perioperative or follow-up monitoring.

Composed image retrieval (CIR) aims to retrieve a desired target image from a query consisting of a reference image and a modification text. This task exhibits an unusual representational asymmetry: the modification text specifies a transition from the reference state, whereas retrieval candidates depict completed target states. This creates a representation mismatch for zero-shot methods that query pretrained vision-language spaces directly with transformation-oriented language. We study this mismatch and reformulate zero-shot composed image retrieval as target-state reconstruction followed by retrieval. We instantiate this formulation with ASAP-CIR, a training-free framework that reconstructs a static target representation using a frozen multimodal large language model (MLLM). The representation combines multiple holistic descriptions with a variable set of importance-weighted atomic semantics, thereby preserving both overall target identity and fine-grained visual constraints. Retrieval then integrates holistic state alignment, atomic constraint grounding, and calibrated target-state evidence aggregation. A controlled text-only diagnostic shows that target-side static query formulations achieve more reliable retrieval than dynamic composed query formulations, particularly when source-state semantics must be suppressed or transformed. Experiments on FashionIQ, CIRR, and CIRCO further characterize the effectiveness and limitations of this representation principle, with the clearest gains on the multi-target CIRCO benchmark. These results show that how composed intent is represented before retrieval is a consequential design choice, distinct from the choice of retrieval backbone itself.

Computational pathology has the potential to improve clinical outcomes through a demonstrated increase in diagnostic and prognostic accuracy. However, the development and validation of deep learning algorithms still require annotated data, a costly procedure involving expert pathologists who already face critical workforce shortages. Existing coreset selection methods to optimize annotation efforts currently all rely on hyperparameters tuned on natural-image benchmarks that do not transfer to histopathology and are cumbersome to use in clinical practice. In this study, we present GCcore, a novel label-free coreset selection method that embeds every image of a dataset with any pathology foundation model and greedily selects the samples that collectively maximize the global coverage of the embedding space. The proposed method provides a lower-bound guarantee on the global coverage of the returned coreset for any coreset size, while being completely hyperparameter-free and deterministic. We demonstrate GCcore's superior performance over 14 baselines including state-of-the-art methods across 10 tasks and datasets spanning whole slide image classification, tile classification, and tissue segmentation, where it ranks first on six and within the top three on nine, while also demonstrating how existing methods can shift by up to five rank positions depending on their hyperparameter settings. Code is publicly available at https://github.com/OncoAI-ULBHUB/GCcore.

Image compression, restoration, and generation all require a way to measure how different two images look to a person. Pixel error ignores how people see, while the most accurate perceptual distances are typically fitted to human judgments, tying them to a fixed data and resolution. For example, when image resolution is doubled, the correlation of DISTS with human scores on TID2013 drops from 0.815 to 0.717. We introduce the Jacobian Lens Distance (JLD), which derives its perceptual geometry from a frozen vision encoder rather than from human labels. JLD combines the locality of early patch features with the perceptual sensitivity captured by later encoder representations. Specifically, we use the encoder Jacobian to identify directions in the early feature space that most strongly affect the encoder output, producing a fixed metric tensor, $E[J^\top J]$, which we call the Jacobian lens. The lens is fitted only once from 100 unlabeled images, taking about 35 seconds. Locally, this construction defines a pullback metric in pixel space, giving JLD a clear geometric interpretation that can be directly analyzed on real images. Across four standard perceptual databases, JLD achieves state-of-the-art performance and consistently outperforms LPIPS, DISTS, PieAPP, and DreamSim. JLD is also robust to changes in image resolution, on TID2013, its lens-term correlation remains nearly unchanged when the resolution is doubled, decreasing only from 0.850 to 0.845. We further introduce JLD-fast, which is $4\times$ faster than LPIPS-VGG while achieving a mean correlation of 0.911. Finally, JLD naturally extends to video, reaching a correlation of 0.786 on Waterloo IVC 4K compared with 0.611 for VMAF.

Reward post-training of flow models either reweights the model's own samples under a KL penalty or a frozen reference, often for thousands of updates, or backpropagates the reward and moves every sample without checking that the move is worth its size. We introduce MEND, a reinforcement learning method built on proximal velocity matching. MEND caps rewards within each prompt group, so samples that already score well receive no move. Below the cap, it proposes moves along the reward gradient and accepts one only when its capped reward gain exceeds a quadratic displacement price. The model then regresses onto the resulting velocity targets, with no KL term, frozen reference model, or advantage weights. In 100 updates, MEND outperforms Flow-GRPO (about 4k updates) on five of six evaluators at the same distance to base-model images. Under an equal-budget protocol, it surpasses ReFL and DiffusionNFT at every evaluated update across four training rewards, reaching PickScore 24.03 versus 23.92 and 23.43, respectively. A 300-update three-reward run also surpasses the five-reward DiffusionNFT model on all three rewards it trains on. MEND is general and easy to adopt: it applies to any flow backbone with a differentiable reward.

Despite their impressive performance on a wide range of video understanding tasks, current Vision Language Models (VLMs) are predominantly designed for offline scenarios and struggle to handle online streaming videos that demand low latency response. Several studies have explored memory and token compression strategies in an attempt to adapt offline VLMs for streaming video understanding tasks. However, through our probing experiment, we identify that most existing works tend to progressively lose long context information as length of input stream increases. To address this, we propose ReMem, a novel training-free adaptation technique that enables VLMs to process streaming videos of arbitrary lengths while improving their long context information retention capability. ReMem exploits memory from two perspectives, implemented as two core components. The Streaming Context Memory (SCM) continuously compresses historical context with query-independent attention. The Retrieved Vision Memory (RVM) then retrieves the most salient, query-relevant context from memory to augment the VLM's input. Comprehensive experiments demonstrate that the proposed ReMem achieves state-of-the-art (SOTA) performance across a variety of widely used benchmarks, spanning both streaming video and general long video understanding tasks.

This paper establishes the first general structural solution to the problem of state estimation for nonlinear systems driven by unknown inputs. Building upon nonlinear unknown-input observability theory, we show that every such system admits a structurally equivalent representation, referred to as the UID-induced normal form. The proposed representation decomposes the information carried by the unknown inputs into two complementary components: unknown-input directions that are structurally decoupled from the observable dynamics and observable quantities that completely represent the unknown-input information affecting the observable dynamics. As a consequence, the UID-induced normal form provides a unified structural solution to unknown-input decoupling and unknown-input reconstruction, without requiring any model or stochastic assumption on the unknown inputs. The practical significance of the proposed framework is demonstrated through a previously unexplored minimal Structure-from-Motion configuration. The proposed representation enables recursive state estimation from only three point features and a single-axis gyroscope, allowing the recovery of the three-dimensional structure and camera motion up to an unknown global scale factor. Experiments on real-world data validate the proposed framework and demonstrate the feasibility of this minimal sensing configuration.

Visual voice cloning requires intelligible, speaker-consistent speech synchronized with visible articulation. However, sequential multimodal conditioning can disrupt previously established temporal and speaker cues, while imbalanced inference guidance can improve linguistic accuracy at the expense of lip synchronization. In this paper, we propose UltraDub, a Unifying Visually-Steered Flow learning and trajectory Guidance Dubbing framework that leverages vision in two ways: as continuous motion for multimodal context aggregation, and as structural rhythm for trajectory rectification. Specifically, we introduce the Motion-guided Dual-context Retrieving (MDR) module, which continually recalibrates linguistic and speaker-style retrieval through shared lip-motion query residuals, utilizing independent time-conditioned gates to regulate their contributions. Furthermore, we propose Rhythm-anchored Trajectory Guidance (RTG), a training-free mechanism that evaluates hierarchical multimodal corrections at a visual-only predictive midpoint, safely strengthening semantic conditioning while better preserving temporal alignment. Finally, we construct DiverseDub, a multi-scenario benchmark to evaluate video dubbing in the wild. Extensive experiments demonstrate that UltraDub achieves state-of-the-art performance across four datasets.

In generative models, Optimal Transport (OT) is used to improve Flow Matching (FM) by reducing noise-data coupling cost. However, different noise-to-output assignments can yield nearly equal costs, raising a key question. Is cost alone sufficient to guide coupling design? We address this question by separating transport cost from routing, i.e., the destination reached by each noise sample. We show numerically how FM and OT can differ in routing while remaining close in cost. We examine its consequences in learned neural FM. Using the exact FM routing as an oracle, we further construct a routing-aware training coupling and find that it yields a directionally consistent improvement in generation over a cost-matched, cost-only counterpart. Our findings highlight what cost minimization can overlook and motivate using both cost and routing to evaluate the design of OT-based FM couplings. Code will be released upon acceptance.

Existing data-driven infrared small target detection (ISTD) methods typically require large-scale datasets with accurate pixel-level annotations for model training. However, such labor-intensive requirements are difficult to satisfy in real-world applications due to the heavy reliance on expert knowledge and the inherently weak distinctiveness of infrared small targets. Consequently, the presence of noisy labels during model training is inevitable, which can severely mislead the learning of target perception toward spurious patterns. To address this challenge, we propose Prompt and Refinement (PAR), a label-noise-robust asymmetric mutual learning paradigm for ISTD. Specifically, PAR comprises a pretrained Segment Anything Model (SAM) and an ISTD-specific detector trained from scratch, which learn collaboratively through a peer-teaching scheme. Coupled with local contrast regularity, the predictions of the two asymmetric peer models are mutually exploited as rectification cues for the supervisory masks of their counterparts. The interaction between complementary inductive biases effectively prevents the label correction process from degenerating into the self-confirmation loop of a single model, enabling progressive refinement of the annotations toward intrinsic target characteristics. In addition, the detector predictions are utilized as corrective mask prompts to facilitate task-specific adaptation of the vision foundation model. Moreover, an evidential uncertainty estimation strategy is introduced into the optimization process to further alleviate the adverse effects of noisy labels. Extensive experiments under diverse noisy label scenarios on three ISTD datasets demonstrate that PAR consistently achieves state-of-the-art performance.

Multi-view panoptic segmentation assigns a semantic class and a scene-level instance ID to every pixel of an unordered set of images, and recent feed-forward 3D models predict these labels for the input views in a single forward pass. Their predictions, however, have been evaluated with the scene-level PQ (PQ^scene) borrowed from per-scene optimization methods, typically on rendered held-out views. PQ^scene tiles all views of a scene into a single image, so that a missed appearance or a change of ID lowers the score of the matched pair only in proportion to its area. We propose View-Consistent Panoptic Quality (VC-PQ), which extends PQ from a single image to a set of input views, counts equally every view in which an instance is visible, and penalizes a prediction that is not visible in the same views as its ground truth. A decomposition of VC-PQ attributes the score a method loses to mask accuracy, view consistency, and the matching threshold. A single additional parameter recovers the area weighting of tiling for comparison. Under a fixed evaluation protocol on ScanNet++ and ScanNetv2, recent feed-forward methods are evaluated with VC-PQ and PQ^scene, and the decomposition shows where each of them loses its score. Controlled perturbations of the ground truth show that VC-PQ responds to the number of views in which an instance is missed or changes ID, whereas PQ^scene responds to their area. The aim of this work is to make view consistency part of the evaluation of multi-view panoptic segmentation, with VC-PQ reported alongside PQ^scene. The evaluation code and the input sets of the protocol are available at https://github.com/Renem24/vcpq.

Real-world thermal super-resolution (SR) is constrained by limited sensor resolution and the difficulty of obtaining corresponding high-resolution (HR) observations for direct model supervision. Conventional SR methods typically construct training pairs by treating captured thermal images with real-world degradations as HR references and applying predefined degradation to generate synthetic low-resolution (LR) inputs. Such a construction not only introduces a domain gap between synthetic and captured LR observations but also retains acquisition degradations in the supervision. To address this issue, we propose AstraSR, a real-world thermal SR method guided by GPT-6 Astra, a frontier multimodal generative model endowed with emergent and transformative visual capabilities. Specifically, we construct a dataset of image pairs by using captured LR thermal images to condition GPT-based HR reference. We develop a direct generative supervision strategy that learns from captured thermal inputs paired with GPT-generated HR references. Pixel, gradient, and perceptual losses jointly supervise the transfer of intensity patterns, structural boundaries, and visual details from the generated references. Qualitative comparisons with seven existing state-of-the-art real-world SR methods show continuous object contours, distinct structural boundaries, and smooth intensity transitions in the thermal scenes. These results demonstrate that AstraSR outperforms existing real-world SR methods in both thermal clarity and structural coherence.

Recent Text-to-Image (T2I) models achieve remarkable visual image generation performance, but they can still generate NSFW (Not-Safe-For-Work) contents, including violent or explicit images. Existing safety checker mechanisms are largely confined to pre-generation filtering (e.g. prompt-level text classifiers) or post-hoc moderation applied after an image is completely synthesized. However, adversarial attack methods operate over a much broader space. This imbalance highlights the need for a safety mechanism that intervenes during the generation process. We propose an in-generation safety framework that monitors the denoising trajectory and detects emerging NSFW signals from intermediate representations. Rather than merely detecting NSFW generations, our method applies reinforcement learning to generate safe images from NSFW prompts. By coupling in-generation detection with controllable steering, our approach mitigates unsafe trajectories even when NSFW signals emerge after generation has already begun. Experiments results show that our method consistently outperforms existing safe image generation methods across both standard and adversarial evaluation sets, while preserving perceptual quality and prompt fidelity. Code will be released upon acceptance.

"Don't tune hyperparameters on the test set" is often stated in machine learning textbooks. Violating it is considered a cardinal sin that produces misleadingly optimistic results, corrupts benchmark integrity, and thus can even be interpreted as scientific fraud. Yet evidence suggests that test set hyperparameter tuning does occur in practice, making it all the more important to understand its actual consequences. So how bad is it, really? In this work we question this dogma and put it to an empirical test. We systematically study the magnitude of the performance inflation caused by tuning the hyperparameters on the test set for MNIST-1D, CIFAR-10, and three tasks from the GLUE benchmark. Our experiments show that while the effect is real and significant, it is frequently small relative to other sources of noise. In many cases, we find that tuning on the test set recovers exactly the same model as when tuning on the validation set. Most importantly, we find that the rankings of models remain essentially preserved after tuning on the test set and therefore that consistent test-set tuning may not invalidate benchmarks or model selection. Our results call for a more nuanced view of tuning hyperparameters on the test set, stimulating researchers to openly report test tuning.