arXiv subjects

Ruitong Sun

Publications and source records attributed to Ruitong Sun.

11 recordsLinked to original sources

arXiv2026

Self-Evolving Code-with-Image Reasoning

Multimodal models increasingly reach for tools when solving visual tasks (crop, zoom, rotate, brighten), a paradigm known as thinking-with-images. The central challenge is one of perception: tools mostly serve to expose visual evidence, reasoning over that evidence stays in language, and most targets are ones a human could in principle determine by inspection. Some visual questions, however, are not bottlenecked by perception: recovering their answers requires executing a multi-step visual algorithm over the pixels. On such questions a model often names the correct algorithm at once yet still answers wrong, because language can describe an algorithm without being able to run one. Code-with-Image crosses that line: given nothing but a Python interpreter, the model must implement a genuine visual algorithm in code to solve the task; the program itself becomes the reasoning. The bottleneck then shifts from executing code to deciding which algorithm to implement. So we let the model teach itself: a training-free reflection loop studies its own failed programs, tests repairs against constructive ground truth, and keeps what survives as portable skills. On our Code-with-Image Bench (CwI-Bench), thirty task families induced by hidden visual computations with disjoint learning and evaluation splits, even GPT-5.6-luna stays below 30% with tool-free chain of thought; given a bare interpreter it reaches 43%, and with skills evolved through its own executable reflection, 67%. The open 27B model climbs the same ladder (9% $\rightarrow$ 33% $\rightarrow$ 56%), and the skills are plain text, transferable across scales and families. When code carries the reasoning, debugging code becomes debugging reasoning.

cs.CV↗

arXiv2026

TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL

Reinforcement learning (RL) for visual reasoning needs scalable, verifiable, and controllable training signals. Existing visual RL post-training trains on static curated datasets, with fixed image-question-answer samples bounded by their collection budget. In this work, we introduce TRON (Targeted, Rule-verifiable Online eNvironments), an online environment substrate: a training rollout is generated on demand by a controllable generator-verifier program that samples a fresh latent visual state, renders an image, asks a question, and exactly verifies the answer. A single run can therefore draw an unbounded stream of fresh instances at the difficulty level required by the current curriculum. The current TRON suite contains 520 environments organized into five ability buckets (spatial, mathematical, diagram, pattern/logic, and counting); the same substrate supports both a single full model trained on all buckets and per-bucket ability-specialist models, with no additional data collection. We also introduce a substrate analysis covering generation reliability, instance and level diversity, cross-environment near-duplicates, and base-model pass rate by difficulty level. RL post-training with METHOD consistently improves performance on ten external multimodal reasoning benchmarks across Qwen3-VL-4B, Qwen2.5-VL-7B, and MiMo-VL-7B-SFT.

cs.AI↗

arXiv2026

Self-Improving Small Object Grounding in LVLMs

Can internal attention patterns in Large Vision Language Models (LVLMs) identify reliable small-object boxes without fine-tuning? In this work, we provide an affirmative answer. Attention structure in LVLMs encodes grounding quality-a lightweight IoU regressor trained solely on attention maps achieves strong IoU prediction (Pearson r > 0.67). This regressor powers the regressor-based variant of our Attention-based Candidate Selection (ACS) framework, called ACS-Learned, which selects the best box from multiple sampled candidates to improve object grounding. By analyzing what the regressor learns, we reveal which transformer layers and heads are most critical and derive ACS-Free: a training-free selector that ranks candidates by attention entropy on these discriminative heads, with no learned component at inference. Experiments on COCO and Objects365 demonstrate up to 19% self-improvement on small object localization, with ACS-Free ranking best among all training-free methods, demonstrating that useful attention structure improves both localization reliability and interpretability in LVLMs.

cs.CV↗

arXiv2025

RSTR: Reducing SpatioTemporal Redundancy in Diffusion Transformers

Diffusion Transformers (DiTs) have achieved remarkable success in image generation, yet their deployment is hindered by high computational costs. We identify two sources of redundancy. First, temporal redundancy: Classifier-Free Guidance (CFG) applies costly dual forward passes at every timestep, yet guidance matters only at specific steps, and variable scales at critical steps can compensate for skipping others. Second, spatial redundancy: under variable guidance, different transformer blocks exhibit heterogeneous sensitivity, yet uniform calibration across all blocks wastes computation while failing to address their varying requirements. We present RSTR, the first framework to jointly reduce spatiotemporal redundancy in diffusion transformers. Stage-1 addresses temporal redundancy through evolutionary search, discovering sparse guidance schedules with variable scales. Stage-2 addresses spatial redundancy through adaptive rank allocation, assigning calibration capacities to transformer regions based on their sensitivity. Experiments on DiT-XL/2, PixArt-$\alpha$, FLUX, and state-of-the-art Qwen-Image demonstrate 50%-70% compute savings while maintaining or improving quality. On DiT-XL/2, RSTR achieves 57% savings with 15% FID improvement; on Qwen-Image, 3.43$\times$ speedup with preserved quality.

cs.CV↗

arXiv2025

Common Inpainted Objects In-N-Out of Context

We present Common Inpainted Objects In-N-Out of Context (COinCO), a novel dataset addressing the scarcity of out-of-context examples in existing vision datasets. By systematically replacing objects in COCO images through diffusion-based inpainting, we create 97,722 unique images featuring both contextually coherent and inconsistent scenes, enabling effective context learning. Each inpainted object is meticulously verified and categorized as in- or out-of-context through Large Vision Language Model assessments. We demonstrate three key tasks enabled by COinCO: (1) a fine-grained context reasoning approach that classifies objects as in- or out-of-context based on three criteria; (2) a novel Objects-from-Context prediction task that determines which new objects naturally belong in given scenes at both instance and clique level semantics, and (3) context-enhanced fake detection on state-of-the-art methods without fine-tuning. COinCO provides a controlled testbed with contextual variations, establishing a foundation for advancing context-aware visual understanding in computer vision, including image forensics. Code and dataset are available at https://co-in-co.github.io/.

cs.CV↗

arXiv2025

Magnetostrictive Phononic Frequency Combs

Magnetostriction, mechanical-to-magnetic or magnetic-to-mechanical response, plays a pivotal role in magneto-mechanical systems. Here, we propose and experimentally demonstrate a magneto-mechanical frequency comb via the three-wave mixing mechanism, which solely requires the involvement of the fundamental mode f0 of a magnetostrictive macroresonator. Two types of combs, i.e., the integer-harmonic combs and the half-integer-harmonic combs, are observed in kHz regime with Hz resolution by magnetically pumping the mm-scale resonator with near-resonant f0. The integer-harmonic combs are centered at lfp, while the half-integer-harmonic combs are centered at (2n - 1) fp/2 resulting from the period-doubling bifurcation of fp. The tooth spacing of both types of combs is determined and can be continuously tuned by changing fs from Hz to kHz. Moreover, the half-integer-harmonic combs can be purposely switched with frequency shifting half a tooth spacing via suppressing period-doubling bifurcation. The experimentally observed formation, evolution, and switching of combs can be well understood by introducing the bias magnetical force and modulated linear stiffness into the Duffing equation. Our findings on magnetically manipulated phononic frequency comb could provide a magneto-mechanical platform for potential non-invasive and contactless sensing and even antenna for wireless operation.

cond-mat.str-el↗

arXiv2025

Magnonic chaotic comb

Optical chaotic comb, possessing the key metrics of intrinsic random amplitude, phase, and frequency modulation of comb lines, emerges as a novel chaotic source in information systems for coherence tomography, parallel ranging, and secure communications. Considering the analogies between magnons and photons, the magnonic analog of optical chaotic combs is expected but not yet explored. Here, we propose a scenario of generating magnonic chaotic combs based on mode coupling mechanism in magnonic systems. Especially, we theoretically demonstrate the realization of magnonic frequency combs through three-wave mixing between ultra-strongly coupled magnons in silicon based synthetic antiferromagnet platform. It is found that the realized magnonic frequency combs can transition to chaos via various routes, i.e., subcritical Hopf bifurcation, torus-doubling bifurcation, and torus breakdown. The robustness of magnonic chaotic combs is verified by characterizing the Poincare map, the bifurcation diagrams, and the largest Lyapunov exponents. Furthermore, the unique characters of chaotic combs, perturbation hypersensitivity and noise immunity, are conceptually validated by identifying latent magnetic signal contaminated by inherent noise. Our findings provide a magnonic paradigm of chaotic dynamics in complex systems for potential applications in CMOS-integrated metrology, sensing, and communication.

cond-mat.str-el↗

arXiv2024

Cross-Domain Distribution Alignment for Segmentation of Private Unannotated 3D Medical Images

Manual annotation of 3D medical images for segmentation tasks is tedious and time-consuming. Moreover, data privacy limits the applicability of crowd sourcing to perform data annotation in medical domains. As a result, training deep neural networks for medical image segmentation can be challenging. We introduce a new source-free Unsupervised Domain Adaptation (UDA) method to address this problem. Our idea is based on estimating the internally learned distribution of a relevant source domain by a base model and then generating pseudo-labels that are used for enhancing the model refinement through self-training. We demonstrate that our approach leads to SOTA performance on a real-world 3D medical dataset.

cs.CV↗

arXiv2024

Revolutionizing Finance with LLMs: An Overview of Applications and Insights

In recent years, Large Language Models (LLMs) like ChatGPT have seen considerable advancements and have been applied in diverse fields. Built on the Transformer architecture, these models are trained on extensive datasets, enabling them to understand and generate human language effectively. In the financial domain, the deployment of LLMs is gaining momentum. These models are being utilized for automating financial report generation, forecasting market trends, analyzing investor sentiment, and offering personalized financial advice. Leveraging their natural language processing capabilities, LLMs can distill key insights from vast financial data, aiding institutions in making informed investment choices and enhancing both operational efficiency and customer satisfaction. In this study, we provide a comprehensive overview of the emerging integration of LLMs into various financial tasks. Additionally, we conducted holistic tests on multiple financial tasks through the combination of natural language instructions. Our findings show that GPT-4 effectively follow prompt instructions across various financial tasks. This survey and evaluation of LLMs in the financial domain aim to deepen the understanding of LLMs' current role in finance for both financial practitioners and LLM researchers, identify new research and application prospects, and highlight how these technologies can be leveraged to solve practical challenges in the finance industry.

cs.CL↗

arXiv2024

Relating Events and Frames Based on Self-Supervised Learning and Uncorrelated Conditioning for Unsupervised Domain Adaptation

Event-based cameras provide accurate and high temporal resolution measurements for performing computer vision tasks in challenging scenarios, such as high-dynamic range environments and fast-motion maneuvers. Despite their advantages, utilizing deep learning for event-based vision encounters a significant obstacle due to the scarcity of annotated data caused by the relatively recent emergence of event-based cameras. To overcome this limitation, leveraging the knowledge available from annotated data obtained with conventional frame-based cameras presents an effective solution based on unsupervised domain adaptation. We propose a new algorithm tailored for adapting a deep neural network trained on annotated frame-based data to generalize well on event-based unannotated data. Our approach incorporates uncorrelated conditioning and self-supervised learning in an adversarial learning scheme to close the gap between the two source and target domains. By applying self-supervised learning, the algorithm learns to align the representations of event-based data with those from frame-based camera data, thereby facilitating knowledge transfer.Furthermore, the inclusion of uncorrelated conditioning ensures that the adapted model effectively distinguishes between event-based and conventional data, enhancing its ability to classify event-based images accurately.Through empirical experimentation and evaluation, we demonstrate that our algorithm surpasses existing approaches designed for the same purpose using two benchmarks. The superior performance of our solution is attributed to its ability to effectively utilize annotated data from frame-based cameras and transfer the acquired knowledge to the event-based vision domain.

cs.CV↗

arXiv2023

Explainable Artificial Intelligence Architecture for Melanoma Diagnosis Using Indicator Localization and Self-Supervised Learning

Melanoma is a prevalent lethal type of cancer that is treatable if diagnosed at early stages of development. Skin lesions are a typical indicator for diagnosing melanoma but they often led to delayed diagnosis due to high similarities of cancerous and benign lesions at early stages of melanoma. Deep learning (DL) can be used as a solution to classify skin lesion pictures with a high accuracy, but clinical adoption of deep learning faces a significant challenge. The reason is that the decision processes of deep learning models are often uninterpretable which makes them black boxes that are challenging to trust. We develop an explainable deep learning architecture for melanoma diagnosis which generates clinically interpretable visual explanations for its decisions. Our experiments demonstrate that our proposed architectures matches clinical explanations significantly better than existing architectures.

cs.LG↗