Searcharxiv⌕ Search

arXiv subjects

Tianyi Wang

Publications and source records attributed to Tianyi Wang.

At least 55 records · Page 3Linked to original sources
arXiv2026

When to Lock Attention: Training-Free KV Control in Video Diffusion

Maintaining background consistency while enhancing foreground quality remains a core challenge in video editing. Injecting full-image information often leads to background artifacts, whereas rigid background locking severely constrains the model's capacity for foreground generation. To address this issue, we propose KV-Lock, a training-free framework tailored for DiT-based video diffusion models. Our core insight is that the hallucination metric (variance of denoising prediction) directly quantifies generation diversity, which is inherently linked to the classifier-free guidance (CFG) scale. Building upon this, KV-Lock leverages diffusion hallucination detection to dynamically schedule two key components: the fusion ratio between cached background key-values (KVs) and newly generated KVs, and the CFG scale. When hallucination risk is detected, KV-Lock strengthens background KV locking and simultaneously amplifies conditional guidance for foreground generation, thereby mitigating artifacts and improving generation fidelity. As a training-free, plug-and-play module, KV-Lock can be easily integrated into any pre-trained DiT-based models. Extensive experiments validate that our method outperforms existing approaches in improved foreground quality with high background fidelity across various video editing tasks.

cs.CV↗
arXiv2026

XR-DT: Extended Reality-Enhanced Digital Twin for Safe Motion Planning via Human-Aware Model Predictive Path Integral Control

As mobile robots increasingly operate alongside humans in shared workspaces, ensuring safe, efficient, and interpretable Human-Robot Interaction (HRI) has become a pressing challenge. While substantial progress has been devoted to human behavior prediction, limited attention has been paid to how humans perceive, interpret, and trust robots' inferences and how robots plan safe and efficient trajectories based on predicted human behaviors. To address these challenges, this paper presents XR-DT, an eXtended Reality-enhanced Digital Twin framework for mobile robots, which bridges physical and virtual spaces to enable bi-directional understanding between humans and robots. Our hierarchical XR-DT architecture integrates augmented-, virtual-, and mixed-reality layers, fusing real-time sensor data, simulated environments in the Unity game engine, and human feedback captured through wearable XR devices. Within this framework, we design a novel Human-Aware Model Predictive Path Integral (HA-MPPI) control model, an MPPI-based motion planner that incorporates ATLAS (Attention-based Trajectory Learning with Anticipatory Sensing), a multi-modal Transformer model designed for egocentric human trajectory prediction via XR headsets. Extensive real-world experimental results demonstrate accurate human trajectory prediction, and safe and efficient robot navigation, validating the HA-MPPI's effectiveness within the XR-DT framework. By embedding human behavior, environmental dynamics, and robot navigation into the XR-DT framework, our system enables interpretable, trustworthy, and adaptive HRI.

cs.RO↗
arXiv2026

Word-Anchored Temporal Forgery Localization

Current temporal forgery localization (TFL) approaches typically rely on temporal boundary regression or continuous frame-level anomaly detection paradigms to derive candidate forgery proposals. However, they suffer not only from feature granularity misalignment but also from costly computation. To address these issues, we propose word-anchored temporal forgery localization (WAFL), a novel paradigm that shifts the TFL task from temporal regression and continuous localization to discrete word-level binary classification. Specifically, we first analyze the essence of temporal forgeries and identify the minimum meaningful forgery units, word tokens, and then align data preprocessing with the natural linguistic boundaries of speech. To adapt powerful pre-trained foundation backbones for feature extraction, we introduce the forensic feature realignment (FFR) module, mapping representations from the pre-trained semantic space to a discriminative forensic manifold. This allows subsequent lightweight linear classifiers to efficiently perform binary classification and accomplish the TFL task. Furthermore, to overcome the extreme class imbalance inherent to forgery detection, we design the artifact-centric asymmetric (ACA) loss, which breaks the standard precision-recall trade-off by dynamically suppressing overwhelming authentic gradients while asymmetrically prioritizing subtle forensic artifacts. Extensive experiments demonstrate that WAFL significantly outperforms state-of-the-art approaches in localization performance under both in- and cross-dataset settings, while requiring substantially fewer learnable parameters and operating at high computational efficiency.

cs.CV↗
arXiv2026

LLM-MLFFN: Multi-Level Autonomous Driving Behavior Feature Fusion via Large Language Model

Accurate classification of autonomous vehicle (AV) driving behaviors is critical for safety validation, performance diagnosis, and traffic integration analysis. However, existing approaches primarily rely on numerical time-series modeling and often lack semantic abstraction, limiting interpretability and robustness in complex traffic environments. This paper presents LLM-MLFFN, a novel large language model (LLM)-enhanced multi-level feature fusion network designed to address the complexities of multi-dimensional driving data. The proposed LLM-MLFFN framework integrates priors from largescale pre-trained models and employs a multi-level approach to enhance classification accuracy. LLM-MLFFN comprises three core components: (1) a multi-level feature extraction module that extracts statistical, behavioral, and dynamic features to capture the quantitative aspects of driving behaviors; (2) a semantic description module that leverages LLMs to transform raw data into high-level semantic features; and (3) a dual-channel multi-level feature fusion network that combines numerical and semantic features using weighted attention mechanisms to improve robustness and prediction accuracy. Evaluation on the Waymo open trajectory dataset demonstrates the superior performance of the proposed LLM-MLFFN, achieving a classification accuracy of over 94%, surpassing existing machine learning models. Ablation studies further validate the critical contributions of multi-level fusion, feature extraction strategies, and LLM-derived semantic reasoning. These results suggest that integrating structured feature modeling with language-driven semantic abstraction provides a principled and interpretable pathway for robust autonomous driving behavior classification.

cs.AI↗
arXiv2026

WAVE: Learning Unified & Versatile Audio-Visual Embeddings with Multimodal LLM

While embeddings from multimodal large language models (LLMs) excel as general-purpose representations, their application to dynamic modalities like audio and video remains underexplored. We introduce WAVE (\textbf{u}nified \& \textbf{v}ersatile \textbf{a}udio-\textbf{v}isual \textbf{e}mbeddings), the first LLM-based embedding that creates a unified representation space for text, audio, and video modalities. WAVE employs a novel hierarchical feature fusion strategy and a joint multi-modal, multi-task training approach to enable two key capabilities: any-to-any cross-modal retrieval and the generation of prompt-aware embeddings tailored to user instructions. Experimentally, WAVE sets a new state-of-the-art on the MMEB-v2 video benchmark and achieves superior results in audio and video-to-audio retrieval. Its prompt-aware nature also yields remarkable performance in multimodal question answering, significantly outperforming existing embedding models. Ablation studies validate our joint training strategy, demonstrating improved performance across all modalities. With a newly introduced benchmark for versatile audio-visual learning, WAVE opens up broad possibilities for cross-modal, any-to-any applications. Our code and checkpoints are released at \href{https://github.com/TCL606/WAVE}{https://github.com/TCL606/WAVE}.

cs.CV↗
arXiv2026

Found-RL: foundation model-enhanced reinforcement learning for autonomous driving

Reinforcement Learning (RL) has emerged as a dominant paradigm for end-to-end autonomous driving (AD). However, RL suffers from sample inefficiency and a lack of semantic interpretability in complex scenarios. Foundation Models, particularly Vision-Language Models (VLMs), can mitigate this by offering rich, context-aware knowledge, yet their high inference latency hinders deployment in high-frequency RL training loops. To bridge this gap, we present Found-RL, a platform tailored to efficiently enhance RL for AD using foundation models. A core innovation is the asynchronous batch inference framework, which decouples heavy VLM reasoning from the simulation loop, effectively resolving latency bottlenecks to support real-time learning. We introduce diverse supervision mechanisms: Value-Margin Regularization (VMR) and Advantage-Weighted Action Guidance (AWAG) to effectively distill expert-like VLM action suggestions into the RL policy. Additionally, we adopt high-throughput CLIP for dense reward shaping. We address CLIP's dynamic blindness via Conditional Contrastive Action Alignment, which conditions prompts on discretized speed/command and yields a normalized, margin-based bonus from context-specific action-anchor scoring. Found-RL provides an end-to-end pipeline for fine-tuned VLM integration and shows that a lightweight RL model can achieve near-VLM performance compared with billion-parameter VLMs while sustaining real-time inference (approx. 500 FPS). Code, data, and models will be publicly available at https://github.com/ys-qu/found-rl.

cs.AI↗
arXiv2026

D-ORCA: Dialogue-Centric Optimization for Robust Audio-Visual Captioning

Spoken dialogue is a primary source of information in videos; therefore, accurately identifying who spoke what and when is essential for deep video understanding. We introduce D-ORCA, a \textbf{d}ialogue-centric \textbf{o}mni-modal large language model optimized for \textbf{r}obust audio-visual \textbf{ca}ptioning. We further curate DVD, a large-scale, high-quality bilingual dataset comprising nearly 40,000 multi-party dialogue videos for training and 2000 videos for evaluation in English and Mandarin, addressing a critical gap in the open-source ecosystem. To ensure fine-grained captioning accuracy, we adopt group relative policy optimization with three novel reward functions that assess speaker attribution accuracy, global speech content accuracy, and sentence-level temporal boundary alignment. These rewards are derived from evaluation metrics widely used in speech processing and, to our knowledge, are applied for the first time as reinforcement learning objectives for audio-visual captioning. Extensive experiments demonstrate that D-ORCA substantially outperforms existing open-source models in speaker identification, speech recognition, and temporal grounding. Notably, despite having only 8 billion parameters, D-ORCA achieves performance competitive with Qwen3-Omni across several general-purpose audio-visual understanding benchmarks. Demos are available at \href{https://d-orca-llm.github.io/}{https://d-orca-llm.github.io/}. Our code, data, and checkpoints will be available at \href{https://github.com/WeChatCV/D-ORCA/}{https://github.com/WeChatCV/D-ORCA/}.

cs.CV↗
arXiv2026

Anchored Policy Optimization: Mitigating Exploration Collapse Via Support-Constrained Rectification

Reinforcement Learning with Verifiable Rewards (RLVR) is increasingly viewed as a tree pruning mechanism. However, we identify a systemic pathology termed Recursive Space Contraction (RSC), an irreversible collapse driven by the combined dynamics of positive sharpening and negative squeezing, where the sampling probability of valid alternatives vanishes. While Kullback-Leibler (KL) regularization aims to mitigate this, it imposes a rigid Shape Matching constraint that forces the policy to mimic the reference model's full density, creating a gradient conflict with the sharpening required for correctness. We propose Anchored Policy Optimization (APO), shifting the paradigm from global Shape Matching to Support Coverage. By defining a Safe Manifold based on the reference model's high-confidence support, APO permits aggressive sharpening for efficiency while selectively invoking a restorative force during error correction to prevent collapse. We theoretically derive that APO serves as a gradient-aligned mechanism to maximize support coverage, enabling an Elastic Recovery that re-inflates valid branches. Empirical evaluations on mathematical benchmarks demonstrate that APO breaks the accuracy-diversity trade-off, significantly improving Pass@1 while restoring the Pass@K diversity typically lost by standard policy gradient methods.

cs.AI↗
arXiv2026

ARCAS: An Augmented Reality Collision Avoidance System with SLAM-Based Tracking for Enhancing VRU Safety

Vulnerable road users (VRUs) face high collision risks in mixed traffic, yet most existing safety systems prioritize driver or vehicle assistance over direct VRU support. This paper presents ARCAS, a real-time augmented reality (AR) collision avoidance system that provides personalized spatial alerts to VRUs via wearable AR headsets. By fusing roadside 360° 3D LiDAR with SLAM-based headset tracking and an automatic 3D calibration procedure, ARCAS accurately overlays world-locked 3D bounding boxes and directional arrows onto approaching hazards in the user's passthrough view. The system also enables multi-headset coordination through shared world anchoring. Evaluated in real-world pedestrian interactions with e-scooters and vehicles (180 trials), ARCAS nearly doubles pedestrians' time to collision and increases counterparts' reaction margins by up to 4x compared to unaided eye conditions. Results validate the feasibility and effectiveness of LiDAR-driven AR guidance and highlight the potential of wearable AR as a promising next generation safety tool for urban mobility.

eess.SY↗
arXiv2026

Damper-B-PINN: Damper Characteristics-Based Bayesian Physics-Informed Neural Network for Vehicle State Estimation

Accurate state estimation is fundamental to intelligent vehicles. Wheel load, one of the most important chassis states, serves as an essential input for advanced driver assistance systems (ADAS) and exerts a direct influence on vehicle stability and safety. However, wheel load estimation remains challenging due to the complexity of chassis modeling and the susceptibility of nonlinear systems to noise. To address these issues, this paper first introduces a refined suspension linkage-level modeling approach that constructs a nonlinear instantaneous dynamic model by explicitly considering the complex geometric structure of the suspension. Building upon this, we propose a damper characteristics-based Bayesian physics-informed neural network (Damper-B-PINN) framework to estimate dynamic wheel load, which leverages the suspension dynamics as physical guidance of PINN while employing Bayesian inference to mitigate the effects of system noise and uncertainty. Moreover, a damper-characteristic physics conditioning (DPC) module is designed for embedding physical prior. The proposed Damper-B-PINN is evaluated using both high-fidelity simulation datasets generated by CarSim software and real-world datasets collected from a Formula Student race car. Experimental results demonstrate that our Damper-B-PINN consistently outperforms existing methods across various test conditions, particularly extreme ones. These findings highlight the potential of the proposed Damper-B-PINN framework to enhance the accuracy and robustness of dynamic wheel load estimation, thereby improving the reliability and safety of ADAS applications.

cs.AI↗
arXiv2026

Gazeify Then Voiceify: Physical Object Referencing Through Gaze and Voice Interaction with Displayless Smart Glasses

Smart glasses enhance interactions with the environment by using head-mounted cameras to observe the user's viewpoint, but lack the visual feedback used for common interactions. We introduce Gazeify then Voiceify, a multimodal approach allowing object selection via gaze and voice using displayless smart glasses. Users can select a physical object with their gaze, and the system generates a digital mask and a voice description of the object's semantics. Users can further correct errors through free-form conversation. To demonstrate our approach, we develop an interactive system by integrating advanced object segmentation and detection with a vision-language model. User studies reveal that participants achieve correct gaze selection in 53% of the task trials and use voice disambiguation to correct 58% of the remaining errors. Participants also rated the system as likable, useful, and easy to use.

cs.HC↗
arXiv2026

Do VLMs Have a Moral Backbone? A Study on the Fragile Morality of Vision-Language Models

Despite substantial efforts toward improving the moral alignment of Vision-Language Models (VLMs), it remains unclear whether their ethical judgments are stable in realistic settings. This work studies moral robustness in VLMs, defined as the ability to preserve moral judgments under textual and visual perturbations that do not alter the underlying moral context. We systematically probe VLMs with a diverse set of model-agnostic multimodal perturbations and find that their moral stances are highly fragile, frequently flipping under simple manipulations. Our analysis reveals systematic vulnerabilities across perturbation types, moral domains, and model scales, including a sycophancy trade-off where stronger instruction-following models are more susceptible to persuasion. We further show that lightweight inference-time interventions can partially restore moral stability. These results demonstrate that moral alignment alone is insufficient and that moral robustness is a necessary criterion for the responsible deployment of VLMs.

cs.CY↗
arXiv2026

Retrieval-Augmented Self-Taught Reasoning Model with Adaptive Chain-of-Thought for ASR Named Entity Correction

End-to-end automatic speech recognition (ASR) systems frequently misrecognize domain-specific phrases like named entities, which can cause catastrophic failures in downstream tasks. A new family of named entity correction methods based on large language models (LLMs) has recently emerged. However, these approaches have yet to fully exploit the sophisticated reasoning capabilities inherent to LLMs. To bridge this gap, we propose a novel retrieval-augmented generation framework for correcting named entity errors in ASR. Our approach consists of two key components: (1) a rephrasing language model (RLM) for named entity recognition, followed by candidate retrieval using a phonetic-level edit distance; and (2) a novel self-taught reasoning model with adaptive chain-of-thought (A-STAR) that dynamically adjusts the depth of its reasoning based on task difficulty. Experiments on the AISHELL-1 and Homophone datasets demonstrate the effectiveness of our method, which achieves relative reductions in the named entity character error rate of 17.96\% and 34.42\%, respectively, compared to a strong baseline.

cs.CL↗
arXiv2026

Heterogeneous gene network estimation for single-cell transcriptomic data via a joint regularized deep neural network

Estimation of intracellular gene networks has been a critical component of single-cell transcriptomic data analysis, which can provide crucial insights into the complex interplay between genes, facilitating the discovery of the biological basis of human life at single-cell resolution. Despite notable achievements, existing methodologies often falter in their practicality, primarily due to their narrow focus on simplistic linear relationships and inadequate handling of cellular heterogeneity. To bridge these gaps, we propose a joint regularized deep neural network method incorporating Mahalanobis distance-based K-means clustering (JRDNN-KM) to estimate multiple networks for various cell subgroups simultaneously, accounting for both unknown cellular heterogeneity and zero inflation, and, more importantly, complex nonlinear relationships among genes. We introduce an innovative selection layer for network construction, along with hidden layers that include both shared and subgroup-specific neurons, to capture common patterns and subgroup-specific variations across networks. Applied to real single-cell transcriptomic data from multiple tissues and species, JRDNN-KM demonstrates higher accuracy and biological interpretability in network estimation, and more accurately identifies cell subgroups compared to current state-of-the-art methods.Building on network construction, we further find hub genes with important biological implications and modules with statistical enrichment of biological processes.

stat.AP↗
arXiv2025

Hybrid digital-analog protocols for simulating quantum multi-body interactions

While quantum simulators promise to explore quantum many-body physics beyond classical computation, their capabilities are limited by the available native interactions in the hardware. On many platforms, accessible Hamiltonians are largely restricted to one- and two-body interactions, limiting access to multi-body Hamiltonians and to systems governed by simultaneous, non-commuting interaction terms that are central to condensed matter, quantum chemistry, and high-energy physics. We introduce and experimentally demonstrate a hybrid digital-analog protocol that overcomes these limitations by embedding analog evolution between shallow entangling-gate layers. This method produces effective Hamiltonians with simultaneous non-commuting three- and four-body interactions that are generated non-perturbatively and without Trotter error -- capabilities not practically attainable on near-term hardware using purely digital or purely analog schemes. We implement our scheme on a trapped-ion quantum processor and use it to realize a topological spin chain exhibiting prethermal strong zero modes persisting at high temperature, as well as models featuring three- and four-body interactions. Our hardware-agnostic and scalable method opens new routes to realizing complex many-body physics across quantum platforms.

quant-ph↗
arXiv2025

KEPT: Knowledge-Enhanced Prediction of Trajectories from Consecutive Driving Frames with Vision-Language Models

Accurate short-horizon trajectory prediction is crucial for safe and reliable autonomous driving. However, existing vision-language models (VLMs) often fail to accurately understand driving scenes and generate trustworthy trajectories. To address this challenge, this paper introduces KEPT, a knowledge-enhanced VLM framework that predicts ego trajectories directly from consecutive front-view driving frames. KEPT integrates a temporal frequency-spatial fusion (TFSF) video encoder, which is trained via self-supervised learning with hard-negative mining, with a k-means & HNSW retrieval-augmented generation (RAG) pipeline. Retrieved prior knowledge is added into chain-of-thought (CoT) prompts with explicit planning constraints, while a triple-stage fine-tuning paradigm aligns the VLM backbone to enhance spatial perception and trajectory prediction capabilities. Evaluated on nuScenes dataset, KEPT achieves the best open-loop performance compared with baseline methods. Ablation studies on fine-tuning stages, Top-K value of RAG, different retrieval strategies, vision encoders, and VLM backbones are conducted to demonstrate the effectiveness of KEPT. These results indicate that KEPT offers a promising, data-efficient way toward trustworthy trajectory prediction in autonomous driving.

cs.CV↗