arXiv subjects

Haoyu Yang

Publications and source records attributed to Haoyu Yang.

At least 19 recordsLinked to original sources

arXiv2026

LaViT: Aligning Latent Visual Thoughts for Multi-modal Reasoning

Current multimodal latent reasoning often relies on external supervision (e.g., auxiliary images), ignoring intrinsic visual attention dynamics. In this work, we identify a critical Perception Gap in distillation: student models frequently mimic a teacher's textual output while attending to fundamentally divergent visual regions, effectively relying on language priors rather than grounded perception. To bridge this, we propose LaViT, a framework that aligns latent visual thoughts rather than static embeddings. LaViT compels the student to autoregressively reconstruct the teacher's visual semantics and attention trajectories prior to text generation, employing a curriculum sensory gating mechanism to prevent shortcut learning. Extensive experiments show that LaViT significantly enhances visual grounding, achieving up to +16.9% gains on complex reasoning tasks and enabling a compact 3B model to outperform larger open-source variants and proprietary models like GPT-4o.

cs.CV↗

arXiv2026

EliSeg: Verified Target Construction for Report-Grounded Abnormality Segmentation

Radiology reports describe clinical observations but do not specify executable segmentation targets. They may contain present, negated, prior,uncertain, or irrelevant findings, while multiple valid abnormalities may coexist. Existing segmentation methods largely bypass this ambiguity by receiving a target identity or spatial prompt before inference, which acts as a hidden target oracle. We study report-grounded abnormality segmentation, where a model must determine target eligibility, cardinality, and finding-to-mask correspondence directly from an unfiltered report before delineating the corresponding regions. We propose \textbf{EliSeg}, an atcor--verify--revise framework that integrates target construction with mask generation. A grammar-constrained Actor proposes target slots and masks, an independent text-only Verifier reconstructs the eligible finding inventory, and Revision selectively re-executes the shared Actor when their target structures disagree. EliSeg requires no predefined target identity, finding prompt, point, or bounding box. Experiments on MIMIC-CXR-ILS show that EliSeg consistently outperforms direct segmentation methods and extract-then-segment cascades across findings, while effectively suppressing masks for ineligible report mentions. Ablation studies confirm the complementary roles of verification and revision, and evaluation on CheXlocalize demonstrates effective transfer of the EliSeg to an external dataset.Code is available at https://github.com/Maybach-dream/EliSeg.

cs.CV↗

arXiv2026

MedPixel: A Unified Pixel-Language Model for Medical Reasoning and Segmentation

Reliable medical image understanding requires models to connect clinical language and visual reasoning with pixel-level grounding. Yet medical vision-language models often lack precise localization, whereas medical segmenters typically rely on explicit target categories or precise spatial prompts. This divide is reinforced by a supervision mismatch: segmentation datasets provide precise masks but little language supervision, whereas medical vision-language data rarely pair language with dense spatial annotations. To address this gap, we present MedPixel, a unified medical pixel-language model built around a shared language--mask interface. To provide scalable supervision, we introduce MedPLG-440K, comprising approximately 440K pixel-language task samples constructed through a clinically motivated synthesis process without external LLM annotation. MedPixel is trained with joint multi-task supervised fine-tuning followed by Pixel-Level Preference Optimization, which uses ground-truth masks as offline verifiers to derive response preferences from mask quality. MedPixel supports a broad spectrum of tasks spanning explicit grounding, implicit reasoning, spatial interaction, grounded explanation, and medical VQA. Across this task spectrum, MedPixel achieves strong performance in both pixel-level prediction and response generation, together with effective zero-shot transfer to external grounding benchmarks and robustness to imperfect spatial prompts. Code and model checkpoints will be released at https://github.com/yhy-whu/Medpixel.

cs.CV↗

arXiv2026

Component-Level Inverse Design of Transmon Qubits Using Neural Networks

Designing a superconducting qubit to realize specific Hamiltonian parameters typically requires iterating through a time and compute-intensive forward loop in which the designer chooses a layout geometry, simulates it, extracts circuit parameters such as capacitances, and refines the geometry. We study the inverse version of this task using a neural-network workflow that maps target Hamiltonian parameters directly to component-level layout parameters, which we subsequently demonstrate on a planar transmon layout. During training, we pair the inverse model with a frozen forward surrogate model and evaluate the loss in Hamiltonian space rather than in layout-parameter space. In validation against a conventional EM solver, 97% of generated designs produce usable geometries, and the inverse-plus-surrogate pipeline reaches mean percent errors of 0.73% for qubit frequency and 1.58% for anharmonicity, comparable to or below the fabrication and simulation-to-measurement uncertainty expected for academic-process transmon devices of this type. A single pipeline query takes ~60 ms on CPU, versus ~2 min for a conventional EM capacitance extraction on the same hardware, a speedup of approximately 2,000x. Batching minimizes the AI model inference overhead, reducing the runtime to 3.1 microseconds per sample on CPU and 2.6 microseconds per sample on GPU at a batch size of 2048, resulting in speedups of 3.9 x 10^7 and 4.6 x 10^7, respectively, relative to a single conventional CPU EM extraction. Our results indicate that component-level inverse design usefully extends and complements conventional EM simulation, including for small datasets on the order of 1,000 samples.

quant-ph↗

arXiv2026

SCALE: Self-Supervised Constraint-Aware Layout GEneration for Local P&R DRV Fixing at Advanced Nodes

As semiconductor manufacturing advances toward sub-2nm nodes, local place-and-route (P&R) design-rule violation (DRV) fixing is increasingly limited by complex rule interactions, dense multi-layer routing geometries, and foundry-specific constraints. While Large Language Models (LLMs) have recently demonstrated strong capabilities in EDA scripting and documentation, their application to visual layout understanding remains largely unexplored: diagnosing DRC violations from layout imagery demands precise geometric reasoning and foundry-specific rule knowledge absent from general-purpose VLM training. We propose SCALE, a framework with a self-supervised layout-generation stage for local DRV fixing at advanced nodes. Multi-layer layout geometry is serialized into structured text, and a fine-tuned language model learns to reconstruct randomly masked polygons from surrounding BEOL context alone without violation labels. At inference, natural-language rule constraints and high-temperature sampling steer generation toward diverse, violation-prone layout variants validated by an industrial signoff DRC checker, producing DRC-annotated layout--violation pairs used to fine-tune a domain-adapted DRC-VLM. This VLM provides rule-aware geometric guidance for local DRV repair, boosting state-of-the-art agents' solve rates by +12--25% (up to 97%) on 100 real sub-2nm cases spanning enclosure, spacing, width, and color-spacing violations.

cs.CV↗

arXiv2026

EvoDRC: A Self-Evolving Agentic Framework for Automated DRC Violation Repair

Design rule check (DRC) closure remains a major bottleneck in advanced-node physical design. Although detailed routers are rule-aware, residual design rule violations (DRVs) often require manual engineering change order iterations. Automating this process is challenging because repairs must account for complex geometric interactions, preserve circuit connectivity, and avoid introducing new violations. We present EvoDRC, a skill-evolution framework for agentic block-level DRC repair. EvoDRC initializes layer-specific repair skills using knowledge distilled from an unrelated reference design and continuously evolves these skills using traceable repair experience collected from the target design. EvoDRC decomposes the layout into bounded repair regions and assigns an LLM repair agent to each region. Local DRC analysis, connectivity-checking, and impact-preview tools provide feedback on proposed modifications. Repair operations and their resulting DRV changes are stored in a knowledge database and used to evolve the repair skills. Experiments on seven block-level designs from the DAC26 DRC Benchmark show that EvoDRC achieves a 73.5\% overall reduction compared to the reported baseline.

cs.AI↗

arXiv2026

OptiClear: Differentiable Curvilinear Design Rule Legalization for Inverse-Designed Photonic Devices

Photonic inverse design enables ultra-compact, high-performance devices with highly curvilinear and non-intuitive geometries, but the resulting layouts often violate fabrication design rules and hinder foundry manufacturing. Legalization methods designed for rectilinear Manhattan electrical layouts are not directly applicable to curvilinear inverse-designed photonic devices. Meanwhile, existing fabrication-aware inverse-design methods apply soft penalties on small features and sharp curvatures, but still cannot guarantee design-rule-compliant final layouts. In this work, we present OptiClear, a curvilinear design rule legalization framework for inverse-designed photonic devices. OptiClear provides two complementary legalization engines: OptiClear-R, a rule-based morphological legalizer that efficiently resolves violation regions through iterative morphology-guided mask processing, and OptiClear-D, a differentiable legalizer that formulates legalization as a minimum-distortion mask optimization problem under morphological stationary-point constraints, explicitly seeking a rule-compliant layout with minimal geometric deviation from the original design. We further develop customized differentiable morphological GPU operators that significantly improve the scalability of high-resolution mask legalization. Comprehensive evaluation across diverse inverse-designed photonic devices and a wide range of design-rule settings shows that OptiClear reduces design-rule violations from thousands to zero. The rule-based legalizer offers high runtime efficiency, while the differentiable legalizer more faithfully preserves the original optical functionality. This work establishes curvilinear design rule legalization as a practical post-design electronic-photonic design automation (EPDA) stage for translating high-performance inverse-designed photonic layouts into manufacturable tape-out-ready devices.

physics.optics↗

arXiv2026

Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction

Video event prediction (VEP) requires models to infer unobserved future states from partial video evidence. Existing video MLLMs usually verbalize intermediate future reasoning in text space: once visual evidence is verbalized, fine-grained motion, geometry, and interaction cues can be lost, leading to plausible but visually ungrounded hallucinations. We introduce Future-L1, an interleaved latent visual reasoning framework that lets an MLLM alternate between language tokens and continuous latent visual spans during autoregressive decoding. To train this capability, we construct Future-L1-50K by selecting examples where future visual hints help prediction and align latent states to future-frame embeddings, then further optimize sampled latent trajectories with LA-DAPO, a latent-aware RL objective with outcome-contrastive and temporal-diversity rewards. Future-L1 achieves new state-of-the-art results on both benchmarks: on FutureBench, it improves Qwen3-VL-8B from 61.0 to 85.4 and exceeds the previous best Video-CoE by 10.4 points; on TwiFF-Bench, it improves the average score from 2.44 to 3.04. These results suggest that future-oriented video reasoning benefits from preserving intermediate visual semantics in latent space rather than translating every reasoning step into text.

cs.CV↗

arXiv2026

PRISM: Photonics-Informed Inverse Lithography for Manufacturable Inverse-Designed Photonic Integrated Circuits

Recent advances in photonic inverse design have demonstrated the ability to automatically synthesize compact, high-performance photonic components that surpass conventional, hand-designed structures, offering a promising path toward scalable and functionality-rich photonic hardware. However, the practical deployment of inverse-designed PICs is bottlenecked by manufacturability: their irregular, subwavelength geometries are highly sensitive to fabrication variations, leading to large performance degradation, low yield, and a persistent gap between simulated optimality and fabricated performance. Unlike electronics, photonics lacks a systematic, flexible mask optimization flow. Fabrication deviations in photonic components cause large optical response drift and compounding error in cascaded circuits, while calibrating fabrication models remains costly and expertise-heavy, often requiring repeated fabrication cycles that are inaccessible to most designers. To bridge this gap, we introduce PRISM, a photonics-informed inverse lithography workflow that makes photonic mask optimization data-efficient, reliable, and optics-informed. PRISM (i) synthesizes compact, informative calibration patterns to minimize required fabrication data, (ii) trains a physics-grounded differentiable fabrication model, enabling gradient-based optimization, and (iii) performs photonics-informed inverse mask optimization that prioritizes performance-critical features beyond geometry matching. Across multiple inverse-designed components with both electron-beam lithography and deep ultra-violet photolithography processes, PRISM significantly boosts post-fabrication performance and yield while reducing calibration area and turnaround time, enabling and democratizing manufacturable and high-yield inverse-designed photonic hardware at scale.

physics.optics↗

arXiv2026

TAGRPO: Boosting GRPO on Image-to-Video Generation with Direct Trajectory Alignment

Recent studies have demonstrated the efficacy of integrating Group Relative Policy Optimization (GRPO) into flow matching models, particularly for text-to-image and text-to-video generation. However, we find that directly applying these techniques to image-to-video (I2V) models often fails to yield consistent reward improvements. To address this limitation, we present TAGRPO, a robust post-training framework for I2V models inspired by contrastive learning. Our approach is grounded in the observation that rollout videos generated from identical initial noise provide superior guidance for optimization. Leveraging this insight, we propose a novel GRPO loss applied to intermediate latents, encouraging direct alignment with high-reward trajectories while maximizing distance from low-reward counterparts. Furthermore, we introduce a memory bank for rollout videos to enhance diversity and reduce computational overhead. Despite its simplicity, TAGRPO achieves significant improvements over DanceGRPO in I2V generation. The deliverables will be updated at https://tagrpo.github.io/ .

cs.CV↗

arXiv2026

End-to-End Physical Design Automation Flow for Yield-Optimized Inverse-Designed Large-Scale Electronic-Photonic Integrated Circuits

As AI systems scale to multi-chiplet and wafer-level architectures, the demand for ultra-high bandwidth and system scalability has outpaced the capabilities of electrical interconnects and computing units. Large-scale heterogeneous electronic-photonic integrated chiplets (EPICs) provide a promising solution, but their practical adoption is limited by the lack of a unified, fabrication-aware physical design automation stack. At the same time, inverse-designed ultra-compact photonic devices offer orders-of-magnitude improvements in spatial and spectral density, yet remain constrained by insufficient design-for-manufacturing support and yield optimization. In this work, we present OptoSynthesizer, an end-to-end physical design automation flow for yield-optimized, inverse-designed EPICs. It integrates three key components across the physical design pipeline: (1) OptoSynthesizer-InvDes, a physical-AI-augmented, digital-twin-assisted photonic inverse design and photonics-aware inverse lithography framework; (2) OptoSynthesizer-Place, a GPU-accelerated routing-informed EPIC placer for large-scale routability-optimized layout; and (3) OptoSynthesizer-Route, a hierarchical curvy-aware waveguide router with global-planning-assisted electrical-optical co-routing. Together, these toolkits form a seamless flow from EPIC netlists to fabrication-ready, yield-robust GDS layouts. We demonstrate how this framework enables compact large-scale photonic tensor cores and high-bandwidth interconnect fabrics for heterogeneous EPIC platforms, providing a practical foundation for manufacturable large-scale EPICs in next-generation AI systems.

physics.optics↗

arXiv2026

Pushing the Limits of Inverse Lithography with Generative Reinforcement Learning

Inverse lithography (ILT) is critical for modern semiconductor manufacturing but suffers from highly non-convex objectives that often trap optimization in poor local minima. Generative AI has been explored to warm-start ILT, yet most approaches train deterministic image-to-image translators to mimic sub-optimal datasets, providing limited guidance for escaping non-convex traps during refinement. We reformulate mask synthesis as conditional sampling: a generator learns a distribution over masks conditioned on the design and proposes multiple candidates. The generator is first pretrained with WGAN plus a reconstruction loss, then fine-tuned using Group Relative Policy Optimization (GRPO) with an ILT-guided imitation loss. At inference, we sample a small batch of masks, run fast batched ILT refinement, evaluate lithography metrics (e.g., EPE, process window), and select the best candidate. On \texttt{LithoBench} dataset, the proposed hybrid framework reduces EPE violations under a 3\,nm tolerance and roughly doubles throughput versus a strong numerical ILT baseline, while improving final mask quality. We also present over 20\% EPE improvement on \texttt{ICCAD13} contest cases with 3$\times$ speedup over the SOTA numerical ILT solver. By learning to propose ILT-friendly initializations, our approach mitigates non-convexity and advances beyond what traditional solvers or GenAI can achieve.

cs.LG↗

arXiv2026

Hunyuan-GameCraft-2: Instruction-following Interactive Game World Model

Recent advances in generative world models have enabled remarkable progress in creating open-ended game environments, evolving from static scene synthesis toward dynamic, interactive simulation. However, current approaches remain limited by rigid action schemas and high annotation costs, restricting their ability to model diverse in-game interactions and player-driven dynamics. To address these challenges, we introduce Hunyuan-GameCraft-2, a new paradigm of instruction-driven interaction for generative game world modeling. Instead of relying on fixed keyboard inputs, our model allows users to control game video contents through natural language prompts, keyboard, or mouse signals, enabling flexible and semantically rich interaction within generated worlds. We formally defined the concept of interactive video data and developed an automated process to transform large-scale, unstructured text-video pairs into causally aligned interactive datasets. Built upon a 14B image-to-video Mixture-of-Experts(MoE) foundation model, our model incorporates a text-driven interaction injection mechanism for fine-grained control over camera motion, character behavior, and environment dynamics. We introduce an interaction-focused benchmark, InterBench, to evaluate interaction performance comprehensively. Extensive experiments demonstrate that our model generates temporally coherent and causally grounded interactive game videos that faithfully respond to diverse and free-form user instructions such as "open the door", "draw a torch", or "trigger an explosion".

cs.CV↗

arXiv2026

OMG-Agent: Toward Robust Missing Modality Generation with Decoupled Coarse-to-Fine Agentic Workflows

Data incompleteness severely impedes the reliability of multimodal systems. Existing reconstruction methods face distinct bottlenecks: conventional parametric/generative models are prone to hallucinations due to over-reliance on internal memory, while retrieval-augmented frameworks struggle with retrieval rigidity. Critically, these end-to-end architectures are fundamentally constrained by Semantic-Detail Entanglement -- a structural conflict between logical reasoning and signal synthesis that compromises fidelity. In this paper, we present \textbf{\underline{O}}mni-\textbf{\underline{M}}odality \textbf{\underline{G}}eneration Agent (\textbf{OMG-Agent}), a novel framework that shifts the paradigm from static mapping to a dynamic coarse-to-fine Agentic Workflow. By mimicking a \textit{deliberate-then-act} cognitive process, OMG-Agent explicitly decouples the task into three synergistic stages: (1) an MLLM-driven Semantic Planner that resolves input ambiguity via Progressive Contextual Reasoning, creating a deterministic structured semantic plan; (2) a non-parametric Evidence Retriever that grounds abstract semantics in external knowledge; and (3) a Retrieval-Injected Executor that utilizes retrieved evidence as flexible feature prompts to overcome rigidity and synthesize high-fidelity details. Extensive experiments on multiple benchmarks demonstrate that OMG-Agent consistently surpasses state-of-the-art methods, maintaining robustness under extreme missingness, e.g., a $2.6$-point gain on CMU-MOSI at $70$\% missing rates.

cs.AI↗

arXiv2026

Rotate Your Character: Revisiting Video Diffusion Models for High-Quality 3D Character Generation

Generating high-quality 3D characters from single images remains a significant challenge in digital content creation, particularly due to complex body poses and self-occlusion. In this paper, we present RCM (Rotate your Character Model), an advanced image-to-video diffusion framework tailored for high-quality novel view synthesis (NVS) and 3D character generation. Compared to existing diffusion-based approaches, RCM offers several key advantages: (1) transferring characters with any complex poses into a canonical pose, enabling consistent novel view synthesis across the entire viewing orbit, (2) high-resolution orbital video generation at 1024x1024 resolution, (3) controllable observation positions given different initial camera poses, and (4) multi-view conditioning supporting up to 4 input images, accommodating diverse user scenarios. Extensive experiments demonstrate that RCM outperforms state-of-the-art methods in both novel view synthesis and 3D generation quality.

cs.CV↗

arXiv2025

TK-Mamba: Marrying KAN With Mamba for Text-Driven 3D Medical Image Segmentation

3D medical image segmentation is important for clinical diagnosis and treatment but faces challenges from high-dimensional data and complex spatial dependencies. Traditional single-modality networks, such as CNNs and Transformers, are often limited by computational inefficiency and constrained contextual modeling in 3D settings. To alleviate these limitations, we propose TK-Mamba, a multimodal framework that fuses the linear-time Mamba with Kolmogorov-Arnold Networks (KAN) to form an efficient hybrid backbone. Our approach is characterized by two primary technical contributions. Firstly, we introduce the novel 3D-Group-Rational KAN (3D-GR-KAN), which marks the first application of KAN in 3D medical imaging, providing a superior and computationally efficient nonlinear feature transformation crucial for complex volumetric structures. Secondly, we devise a dual-branch text-driven strategy using Pubmedclip's embeddings. This strategy significantly enhances segmentation robustness and accuracy by simultaneously capturing inter-organ semantic relationships to mitigate label inconsistencies and aligning image features with anatomical texts. By combining this advanced backbone and vision-language knowledge, TK-Mamba offers a unified and scalable solution for both multi-organ and tumor segmentation. Experiments on multiple datasets demonstrate that our framework achieves state-of-the-art performance in both organ and tumor segmentation tasks, surpassing existing methods in both accuracy and efficiency. Our code is publicly available at https://github.com/yhy-whu/TK-Mamba

cs.CV↗

arXiv2025

An LLM-based Quantitative Framework for Evaluating High-Stealthy Backdoor Risks in OSS Supply Chains

In modern software development workflows, the open-source software supply chain contributes significantly to efficient and convenient engineering practices. With increasing system complexity, using open-source software as third-party dependencies has become a common practice. However, the lack of maintenance for underlying dependencies and insufficient community auditing create challenges in ensuring source code security and the legitimacy of repository maintainers, especially under high-stealthy backdoor attacks exemplified by the XZ-Util incident. To address these problems, we propose a fine-grained project evaluation framework for backdoor risk assessment in open-source software. The framework models stealthy backdoor attacks from the viewpoint of the attacker and defines targeted metrics for each attack stage. In addition, to overcome the limitations of static analysis in assessing the reliability of repository maintenance activities such as irregular committer privilege escalation and limited participation in reviews, the framework uses large language models (LLMs) to conduct semantic evaluation of code repositories without relying on manually crafted patterns. The framework is evaluated on sixty six high-priority packages in the Debian ecosystem. The experimental results indicate that the current open-source software supply chain is exposed to various security risks.

cs.SE↗

arXiv2025

FaceSleuth-R: Adaptive Orientation-Aware Attention for Robust Micro-Expression Recognition

Micro-expression recognition (MER) has achieved impressive accuracy in controlled laboratory settings. However, its real-world applicability faces a significant generalization cliff, severely hindering practical deployment due to poor performance on unseen data and susceptibility to domain shifts. Existing attention mechanisms often overfit to dataset-specific appearance cues or rely on fixed spatial priors, making them fragile in diverse environments. We posit that robust MER requires focusing on quasi-invariant motion orientations inherent to micro-expressions, rather than superficial pixel-level features. To this end, we introduce \textbf{FaceSleuth-R}, a framework centered on our novel \textbf{Single-Orientation Attention (SOA)} module. SOA is a lightweight, differentiable operator that enables the network to learn layer-specific optimal orientations, effectively guiding attention towards these robust motion cues. Through extensive experiments, we demonstrate that SOA consistently discovers a universal near-vertical motion prior across diverse datasets. More critically, FaceSleuth-R showcases superior generalization in rigorous Leave-One-Dataset-Out (LODO) protocols, significantly outperforming baselines and state-of-the-art methods when confronted with domain shifts. Furthermore, our approach establishes \textbf{state-of-the-art results} across several benchmarks. This work highlights adaptive orientation-aware attention as a key paradigm for developing truly generalized and high-performing MER systems.

cs.CV↗