arXiv subjects

Qi Kang

Publications and source records attributed to Qi Kang.

9 recordsLinked to original sources

arXiv2026

VidUEU-Agent: A Data-Curation Agent for Multimodal Understanding, Editing, and Unified Tasks

The advancement of multimodal learning is heavily constrained by the scarcity and high annotation cost of high-quality instruction data. While raw videos offer an abundant and dynamic source of visual knowledge, transforming them into precise instruction data remains highly challenging. To bridge this gap, we present VidUEU-Agent, an automated pipeline that synthesizes instruction data from raw videos for multimodal understanding, editing, and unified tasks. VidUEU-Agent follows a four-stage pipeline: it performs global perception to filter low-quality shots, mines keyframes and multimodal signals, constructs metadata and routes the metadata to synthesize instruction samples. More importantly, our agent framework can autonomously iterate the data distribution based on downstream training results, thereby optimizing data quality, balance, and training effectiveness. Scaling this pipeline, we construct VidUEU-100K, a large-scale dataset of 100K training samples spanning multiple domains and tasks, providing a robust foundation for developing comprehensive multimodal capabilities. Extensive experiments demonstrate that fine-tuning baseline models on VidUEU-100K yields significant and consistent performance gains across diverse representative benchmarks, validating the effectiveness of our proposed agent framework and dataset.

cs.CV↗

arXiv2026

Hi-CoLA: High Confidence Lower Bound Approximation Based Reinforcement Learning for Flex-Route Transit Operation Control

This paper studies safe policy improvement for real-time operation control of flex-route transit (FRT) systems, where robustness is critical and any performance decrease is undesirable. While reinforcement learning (RL) has shown strong empirical success, its adoption remains limited in safety and cost critical applications, such as intelligent transportation, since learned policies lack statistically certified performance guarantees before deployment. In this paper, we propose the High Confidence Lower Bound Approximation (Hi-CoLA) RL framework, which combines importance sampling and concentration inequalities with a differentiable surrogate network to efficiently estimate the high-confidence performance lower bound and optimize it via policy gradient for robust performance improvement. On the FRT operation control problem, Hi-CoLA outperforms state-of-the-art offline and safe RL methods in both average and tail-risk performance. A case study on an actual route in Chicago, implemented in the SUMO simulator, serves as a proof of concept. Hi-CoLA enhances real-time FRT control with guaranteed performance and is broadly applicable to other safety critical decision-making scenarios.

cs.RO↗

arXiv2026

OSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific Software

Scientific software presents a demanding test for computer-using agents based on visual language models (VLMs): completing a research workflow requires interpreting specialized interfaces, manipulating scientific objects, and producing verifiable results. We thus introduce OSWorld-Science, a benchmark and evaluation environment that combines scientifically meaningful tasks, artifact-based evaluation, and an efficient agent harness for studying computer use in the scientific domain. The benchmark contains 12 VLMs and 146 high-quality tasks across several scientific domains and software configurations, covering workflows such as molecular drawing and retrosynthesis, pathology image analysis, statistical computing, and physical simulation. Tasks are developed through expert proposals and iterative human--AI co-design, with selection guided by scientific value and difficulty. Task-specific execution-based evaluators inspect application states and generated artifacts, including molecular structures, segmentation masks, plots, and numerical results, and award partial credit for incomplete outcomes. Our special harness integrates model adapters, interaction-loop control, and trajectory logging to support comparisons of models and interaction strategies. Our results show that current state-of-the-art VLMs with a strong harness still face challenges in addressing key questions in the scientific domains. We also analyze the benchmarking results across multi-linguistics, reasoning efforts, context length and other factors and derive several important conclusions and directions to assist future development. Overall, we provide an integrated framework connecting expert-defined scientific goals to verifiable software outcomes, enabling systematic evaluation of both agent capabilities and harness design in scientific workflows.

cs.AI↗

arXiv2026

VeriFuse: Bounded Vision-Language Arbitration and Reason-Guided Refinement for Cooperative 3D Perception

Vision-language models (VLMs) have demonstrated strong scene understanding and semantic judgment across diverse tasks, but their appropriate role in cooperative perception remains unclear. Directly asking a VLM to regress 3D detections is unreliable and computationally expensive, whereas using it to select the output of a single source discards useful information from other agents. We introduce VeriFuse, a bounded arbitration framework for vehicle-infrastructure cooperative 3D detection. Each agent first produces detections independently. Around each vehicle and roadside proposal, VeriFuse generates source-conditioned geometric candidates and combines the original detections, their perturbations, and cross-source hypotheses into a unified candidate pool. A frozen VLM then chooses among three admissible actions: SELECT an adequate candidate; REFINE an existing anchor when an object is supported but all candidates are geometrically inadequate; or REJECT an unsupported infrastructure-only proposal. Experiments on the DAIR-V2X dataset show that VeriFuse achieves 0.494/0.357 cooperative 3D AP50/AP70 and limits the relative vehicle-side BEV AP50 drop under a 300 ms delay to 1.7%. Overall, VeriFuse assigns the VLM a clear and constrained role in cooperative perception: semantic reasoning resolves ambiguity among cross-agent hypotheses, while deterministic constraints determine the final 3D geometry.

cs.CV↗

arXiv2026

IMUG-Bench: Benchmarking Unified Multimodal Models on Interleaved Understanding and Generation

In recent years, unified multimodal models (UMMs) have emerged to support both understanding and generation within a single framework. Mastering dynamic, multi-turn interleaved image-text dialogues is a crucial task for UMMs in real-world applications. However, existing benchmarks fail to evaluate this important task, as they are often limited to single-turn or static settings, and typically overlook exposure bias in multi-turn interactions. To bridge this gap, we propose IMUG-Bench, a comprehensive benchmark for multi-turn interleaved image-text dialogue of UMMs that jointly evaluates their understanding and generation capabilities. Our IMUG-Bench comprises three classes: Static Spatial, Temporal Causal, and Hybrid, covering 3,113 samples and 12,034 interaction turns. It also includes dynamic understanding questions, thereby supporting evaluation that better reflects real-world multi-turn interaction scenarios. Large-scale experiments on IMUG-Bench systematically evaluate mainstream open-source and closed-source UMMs, revealing their capability boundaries and failure modes, and uncovering pronounced exposure bias on the generation side in multi-turn interactions. We further explore several test-time scaling strategies, including Chain-of-Thought, Self-Verification, and Best-of-N Sampling, which effectively improve generation accuracy and mitigate exposure bias in generation tasks. These findings provide insights into enhancing the robustness and multi-turn interaction capability of future UMMs.

cs.AI↗

arXiv2026

Drive-KD: Multi-Teacher Distillation for VLMs in Autonomous Driving

Autonomous driving is an important and safety-critical task, and recent advances in LLMs/VLMs have opened new possibilities for reasoning and planning in this domain. However, large models demand substantial GPU memory and exhibit high inference latency, while conventional supervised fine-tuning (SFT) often struggles to bridge the capability gaps of small models. To address these limitations, we propose Drive-KD, a framework that decomposes autonomous driving into a "perception-reasoning-planning" triad and transfers these capabilities via knowledge distillation. We identify layer-specific attention as the distillation signal to construct capability-specific single-teacher models that outperform baselines. Moreover, we unify these single-teacher settings into a multi-teacher distillation framework and introduce asymmetric gradient projection to mitigate cross-capability gradient conflicts. Extensive evaluations validate the generalization of our method across diverse model families and scales. Experiments show that our distilled InternVL3-1B model, with ~42 times less GPU memory and ~11.4 times higher throughput, achieves better overall performance than the pretrained 78B model from the same family on DriveBench, and surpasses GPT-5.1 on the planning dimension, providing insights toward efficient autonomous driving VLMs.

cs.AI↗

arXiv2026

Drive-P2D: A Progressive Perception-to-Decision Benchmark for VLMs in Autonomous Driving

Autonomous driving requires reliable perception and safe decision-making in complex scenarios. Recent vision-language models (VLMs) demonstrate reasoning and generalization abilities, opening new possibilities for autonomous driving; however, existing benchmarks often evaluate perception and decision-making separately, limit failure analysis with choice-only formats, or introduce evaluation bias through LLM-scored long-form outputs. To address these issues, we present Drive-P2D, a progressive perception-to-decision benchmark with 6,650 questions across Object, Scene, and Decision levels. Drive-P2D adopts a separated reasoning-and-answer protocol: final answers are scored objectively, while reasoning is analyzed to identify error modes exposed along the progressive perception-to-decision chain. We evaluate mainstream VLMs across all and high-risk scenarios, and further characterize the perception-to-decision capability boundary through correlation analysis and similar-scene robustness testing. Reasoning further exposes failure modes such as logical reasoning errors and semantic feature omissions, and we train a lightweight analyzer model to automate large-scale error-mode annotation of reasoning. Together, these designs provide practical insights for building safer and more reliable VLMs for real-world autonomous driving.

cs.AI↗

arXiv2024

Solving the Food-Energy-Water Nexus Problem via Intelligent Optimization Algorithms

The application of evolutionary algorithms (EAs) to multi-objective optimization problems has been widespread. However, the EA research community has not paid much attention to large-scale multi-objective optimization problems arising from real-world applications. Especially, Food-Energy-Water systems are intricately linked among food, energy and water that impact each other. They usually involve a huge number of decision variables and many conflicting objectives to be optimized. Solving their related optimization problems is essentially important to sustain the high-quality life of human beings. Their solution space size expands exponentially with the number of decision variables. Searching in such a vast space is challenging because of such large numbers of decision variables and objective functions. In recent years, a number of large-scale many-objectives optimization evolutionary algorithms have been proposed. In this paper, we solve a Food-Energy-Water optimization problem by using the state-of-art intelligent optimization methods and compare their performance. Our results conclude that the algorithm based on an inverse model outperforms the others. This work should be highly useful for practitioners to select the most suitable method for their particular large-scale engineering optimization problems.

cs.NE↗

arXiv2022

MOGAN: Morphologic-structure-aware Generative Learning from a Single Image

In most interactive image generation tasks, given regions of interest (ROI) by users, the generated results are expected to have adequate diversities in appearance while maintaining correct and reasonable structures in original images. Such tasks become more challenging if only limited data is available. Recently proposed generative models complete training based on only one image. They pay much attention to the monolithic feature of the sample while ignoring the actual semantic information of different objects inside the sample. As a result, for ROI-based generation tasks, they may produce inappropriate samples with excessive randomicity and without maintaining the related objects' correct structures. To address this issue, this work introduces a MOrphologic-structure-aware Generative Adversarial Network named MOGAN that produces random samples with diverse appearances and reliable structures based on only one image. For training for ROI, we propose to utilize the data coming from the original image being augmented and bring in a novel module to transform such augmented data into knowledge containing both structures and appearances, thus enhancing the model's comprehension of the sample. To learn the rest areas other than ROI, we employ binary masks to ensure the generation isolated from ROI. Finally, we set parallel and hierarchical branches of the mentioned learning process. Compared with other single image GAN schemes, our approach focuses on internal features including the maintenance of rational structures and variation on appearance. Experiments confirm a better capacity of our model on ROI-based image generation tasks than its competitive peers.

cs.CV↗