arXiv subjects

Yuxuan Xia

Publications and source records attributed to Yuxuan Xia.

At least 19 recordsLinked to original sources

arXiv2026

An Extended Object Poisson Multi-Bernoulli Filter with Zero-Inflated Poisson Measurement Model Using Belief Propagation

This paper presents an efficient implementation of the extended object Poisson multi-Bernoulli (PMB) filter under the zero-inflated Poisson (ZIP) object measurement model using particle belief propagation (BP). The ZIP measurement model separates a Bernoulli object detection event from the conditional Poisson generation of object measurements, enabling principled handling of empty measurement sets. Building upon the PMB mixture posterior, we present a factorized joint posterior over set of objects with object detection variables and a dual representation of data association using both object-oriented and measurement-oriented association variables. Notably, this representation replaces the implicit high-order global hypothesis constraint by local consistency factors, yielding a factor graph amenable to BP. In addition, we present a particle-based implementation, where the single object densities of Bernoulli components are represented using particles. Simulation results show that the proposed method achieves filtering performance comparable to a sampling-based PMBM implementation, while having lower runtime. We also validate the efficacy of the proposed method using real-world lidar data for pedestrian tracking.

eess.SP↗

arXiv2026

Smoothing Dark Areas in Molecular Latent Diffusion

Latent diffusion is a promising framework for scalable 3D molecular generation, but it requires a latent space that remains smooth, valid, and navigable beyond posterior samples. Existing molecular VAEs, however, are typically learned through reconstruction-based objectives, which do not guarantee such a latent space. We show that this leads to dark areas: regions of latent space that are reachable during diffusion sampling but decode to disconnected or chemically invalid molecules. Unlike in image generation, molecular decoding requires strict structural and chemical precision, so even small latent perturbations can produce catastrophic failures. We therefore propose TopVAE, a topology-optimized VAE that reduces dark areas by making the decoder internalize structural and chemical constraints during training, eliminating the need for test-time chemical correction. TopVAE greatly improves off-posterior robustness, and when paired with a standard DiT, achieves $77\%$ lower FCD-3D on QM9, the highest V&C, $52\%$ lower FCD-3D on GEOM-Drugs, and $1.29{\times}$ more stable and connected molecules on zero-shot scaffold inpainting.

cs.LG↗

arXiv2026

MeanVC 2: Robust Low-Latency Streaming Zero-Shot Voice Conversion

Streaming zero-shot voice conversion (VC) has become increasingly popular due to its potential for real-time applications. The recently proposed MeanVC achieves lightweight streaming zero-shot VC, but it has several limitations: its chunk-wise autoregressive denoising doubles the effective training sequence length, conversion quality degrades under small-chunk settings, and its timbre encoder directly relies on reference mel-spectrograms, making it sensitive to reference audio quality. To address these limitations we propose MeanVC 2. We introduce future-receptive chunking (FRC), which explicitly schedules past and future receptive fields across diffusion transformer decoder layers and removes clean-chunk teacher forcing. By incorporating bounded future context, FRC enables stable conversion with a 40 ms chunk size. We further introduce a universal timbre token encoder, which constructs a timbre representation from a global speaker embedding and retrieves fine-grained timbre cues via cross-attention, improving robustness to low-quality references and enhancing zero-shot speaker similarity. Experimental results show that MeanVC 2 significantly outperforms MeanVC, while reducing latency from 211 ms to 110 ms. Audio samples are publicly available. The source code will be publicly released.

eess.AS↗

arXiv2026

S2Accompanist: A Semantic-Aware and Structure-Guided Diffusion Model for Music Accompaniment Generation

High-fidelity text-to-music generation typically relies on massive proprietary datasets and immense computational resources. Existing models often struggle to generate coherent pure musical accompaniments and lack precise, localized semantic control due to their reliance on coarse, track-level annotations. To address these limitations under constrained data and computing resources, we propose S2Accompanist, a Semantic-Aware and Structure-Guided Diffusion Model developed for the ICME2026 ATTM Grand Challenge. Specifically, we design an automated data pipeline comprising structural segmentation, Large Audio-Language Model driven segment-level captioning, and dual-metric quality grading to overcome the absence of localized metadata in raw datasets. Furthermore, we propose a semantic-aware Variational Autoencoder fine-tuning strategy that explicitly distills foundational LeadSheet structures into the acoustic latent space, effectively improving the overall audio fidelity. Extensive experiments demonstrate that S2Accompanist achieves state-of-the-art objective performance on the ATTM Grand Challenge benchmark across both the Efficiency and Performance Tracks. With only 402M parameters, our model remains competitive compared to larger-scale unconstrained models and secured first place in the Efficiency Track.

eess.AS↗

arXiv2026

Variational PMB filter via coordinate descent Kullback-Leibler divergence minimisation

This paper presents a new derivation of the variational Poisson multi-Bernoulli (V-PMB) filter for multi-target estimation proposed in [#Williams15]. The proposed derivation is based on considering an augmented space that includes the set of target states with their track indices and the global hypothesis variable. Then, we show that the V-PMB projection performs a coordinate descent Kullback-Leibler divergence (KLD) minimisation on this augmented space to fit the best possible PMB density to the Poisson multi-Bernoulli mixture (PMBM) posterior. We also show that this V-PMB projection keeps the probability hypothesis density of the posterior. The paper also includes a comparison with the PMBM filter and other PMB filter variants, including a track-oriented Murty-based implementation, a track-oriented loopy belief propagation implementation and a global nearest neighbour implementation, showing the benefits of the V-PMB filter compared to the other PMB filters when targets get in close proximity and then separate.

eess.SY↗

arXiv2026

Parallel Prefix Verification for Speculative Generation

We introduce PARSE (PArallel pRefix Speculative Engine), a speculative generation framework that accelerates large language model (LLM) inference by parallelizing prefix verification on a semantic level. Existing speculative decoding methods are fundamentally limited by token-level equivalence: the target model must verify each token, leading to short acceptance lengths and modest speedups. Moving to semantic or segment-level verification can substantially increase acceptance granularity, but prior approaches rely on sequential verification, introducing significant overhead and limiting practical gains. PARSE introduces parallel prefix verification, enabling semantic-level verification without sequential checks. Given a full draft from a draft model, the target model evaluates correctness across multiple prefixes in a single forward pass using a custom attention mask, directly identifying the maximal valid prefix. This eliminates sequential segment verification, and makes verification compute-efficient. PARSE is orthogonal to token-level speculative decoding and can be composed with it for additional gains. Across models and benchmarks, PARSE delivers $1.25\times$ to $4.3\times$ throughput gain over the target model, and $1.6\times$ to $4.5\times$ when composed with EAGLE-3, all with negligible accuracy degradation. This demonstrates parallel prefix verification as an effective, general approach to accelerating LLM inference.

cs.AI↗

arXiv2026

Efficient Implementations of Extended Object PMBM Filters with Blocked Gibbs Sampling

This paper considers multiple extended object tracking based on Poisson multi-Bernoulli mixture (PMBM) filtering, which gives the closed-form Bayesian solution for standard multiple extended object models with Poisson birth. To efficiently address the challenging extended object data association problem in PMBM filtering, we develop implementations of the extended object PMBM filter using blocked Gibbs sampling. By formulating the PMBM density on an augmented state space with auxiliary variables and leveraging the Poisson object measurement model, we first derive a joint posterior over potential objects, previous global hypotheses, and current measurement association variables, together with its corresponding factorization. This factorized representation leads to blocked Gibbs samplers that efficiently generate high-weight global hypotheses and thereby provide an efficient implementation of the PMBM update step. We further introduce a collapsed Gibbs sampling variant, in which the Bernoulli object existence variables are marginalized out, yielding higher sampling efficiency, especially for the initiation of newly detected objects. The proposed methods, implemented under the gamma Gaussian inverse-Wishart model, are compared with an extended object Poisson multi-Bernoulli filter based on particle belief propagation. Simulation results demonstrate that the proposed approaches achieve comparable tracking performance while requiring substantially less runtime.

stat.ME↗

arXiv2026

Event-Triggered Distributed Target Tracking via PRIMEX

PRIMEX (prime-based graph encoding and extraction) is a recently proposed framework for scalable distributed fusion. In PRIMEX, the information pedigree of state estimates or probability density functions is encoded using the information codes, enabling lightweight arithmetic for redundancy removal and data integration. Building on PRIMEX and its memoryless fusion strategy based on a least-squares approximation, in this paper we present two efficient distributed tracking algorithms: a consensus-based PRIMEX method that fuses information from all neighbors, and a greedy gossip-based PRIMEX method that fuses with the most informative neighbor. To further increase communication efficiency, we incorporate an event-triggered mechanism, in which transmission decisions are driven by information novelty measured using differences between the information codes. The proposed methods are evaluated and compared with covariance intersection and centralized fusion in a distributed single target tracking scenario. Simulation results show that PRIMEX-based methods remain competitive in tracking accuracy while improving communication efficiency.

eess.SP↗

arXiv2026

GOSPA and T-GOSPA quasi-metrics for evaluation of multi-object tracking algorithms

This paper introduces two quasi-metrics for performance assessment of multi-object tracking (MOT) algorithms. One quasi-metric is an extension of the generalised optimal subpattern assignment (GOSPA) metric and measures the discrepancy between sets of objects. The other quasi-metric is an extension of the trajectory GOSPA (T-GOSPA) metric and measures the discrepancy between sets of trajectories. Similar to the GOSPA-based metrics, these quasi-metrics include costs for localisation error for properly detected objects, the number of false objects and the number of missed objects. The T-GOSPA quasi-metric also includes a track switching cost. Differently from the GOSPA and T-GOSPA metrics, the proposed quasi-metrics have the flexibility of penalising missed and false objects with different costs, and the localisation costs are not required to be symmetric. We also explain how to obtain similarity score functions based on these quasi-metrics. The performance of several Bayesian MOT algorithms is assessed with the T-GOSPA quasi-metric via simulations.

cs.CV↗

arXiv2026

Occlusion-Aware Multi-Object Tracking via Expected Probability of Detection

This paper addresses multi-object systems, where objects may occlude one another relative to the sensor. The standard point-object model for detection-based sensors is enhanced so that the probability of detection considers the presence of all objects. A principled tracking method is derived, assigning each object an expected probability of detection, where the expectation is taken over the reduced Palm density, which means conditionally on the object's existence. The assigned probability thus considers the object's visibility relative to the sensor, under the presence of other objects. Unlike existing methods, the proposed method systematically accounts for uncertainties related to all objects in a clear and manageable way. The method is demonstrated through a visual tracking application using the multi-Bernoulli mixture (MBM) filter with marks.

eess.SY↗

arXiv2026

B$^2$F-Map: Crowd-sourced Mapping with Bayesian B-spline Fusion

Crowd-sourced mapping offers a scalable alternative to creating maps using traditional survey vehicles. Yet, existing methods either rely on prior high-definition (HD) maps or neglect uncertainties in the map fusion. In this work, we present a complete pipeline for HD map generation using production vehicles equipped only with a monocular camera, consumer-grade GNSS, and IMU. Our approach includes on-cloud localization using lightweight standard-definition maps, on-vehicle mapping via an extended object trajectory (EOT) Poisson multi-Bernoulli (PMB) filter with Gibbs sampling, and on-cloud multi-drive optimization and Bayesian map fusion. We represent the lane lines using B-splines, where each B-spline is parameterized by a sequence of Gaussian distributed control points, and propose a novel Bayesian fusion framework for B-spline trajectories with differing density representation, enabling principled handling of uncertainties. We evaluate our proposed approach, B$^2$F-Map, on large-scale real-world datasets collected across diverse driving conditions and demonstrate that our method is able to produce geometrically consistent lane-level maps.

cs.RO↗

arXiv2026

OptiPMB: Enhancing 3D Multi-Object Tracking with Optimized Poisson Multi-Bernoulli Filtering

Accurate 3D multi-object tracking (MOT) is crucial for autonomous driving, as it enables robust perception, navigation, and planning in complex environments. While deep learning-based solutions have demonstrated impressive 3D MOT performance, model-based approaches remain appealing for their simplicity, interpretability, and data efficiency. Conventional model-based trackers typically rely on random vector-based Bayesian filters within the tracking-by-detection (TBD) framework but face limitations due to heuristic data association and track management schemes. In contrast, random finite set (RFS)-based Bayesian filtering handles object birth, survival, and death in a theoretically sound manner, facilitating interpretability and parameter tuning. In this paper, we present OptiPMB, a novel RFS-based 3D MOT method that employs an optimized Poisson multi-Bernoulli (PMB) filter while incorporating several key innovative designs within the TBD framework. Specifically, we propose a measurement-driven hybrid adaptive birth model for improved track initialization, employ adaptive detection probability parameters to effectively maintain tracks for occluded objects, and optimize density pruning and track extraction modules to further enhance overall tracking performance. Extensive evaluations on nuScenes and KITTI datasets show that OptiPMB achieves superior tracking accuracy compared with state-of-the-art methods, thereby establishing a new benchmark for model-based 3D MOT and offering valuable insights for future research on RFS-based trackers in autonomous driving.

cs.CV↗

arXiv2026

The ICASSP 2026 Automatic Song Aesthetics Evaluation Challenge

This paper summarizes the ICASSP 2026 Automatic Song Aesthetics Evaluation (ASAE) Challenge, which focuses on predicting the subjective aesthetic scores of AI-generated songs. The challenge consists of two tracks: Track 1 targets the prediction of the overall musicality score, while Track 2 focuses on predicting five fine-grained aesthetic scores. The challenge attracted strong interest from the research community and received numerous submissions from both academia and industry. Top-performing systems significantly surpassed the official baseline, demonstrating substantial progress in aligning objective metrics with human aesthetic preferences. The outcomes establish a standardized benchmark and advance human-aligned evaluation methodologies for modern music generation systems.

eess.AS↗

arXiv2026

SDCD: Structure-Disrupted Contrastive Decoding for Mitigating Hallucinations in Large Vision-Language Models

Large Vision-Language Models (LVLMs) demonstrate significant progress in multimodal understanding and reasoning, yet object hallucination remains a critical challenge. While existing research focuses on mitigating language priors or high-level statistical biases, they often overlook the internal complexities of the visual encoding process. We identify that visual statistical bias, arising from the inherent Bag-of-Patches behavior of Vision Encoders under weak structural supervision, acts as a contributing factor of object hallucinations. Under this bias, models prioritize local texture features within individual patches over holistic geometric structures. This tendency may induce spurious visual confidence and result in hallucinations. To address this, we introduce a training-free algorithm called Structure-Disrupted Contrastive Decoding (SDCD), which performs contrastive calibration of the output distribution by introducing a shuffled structure-disrupted view. By penalizing tokens that maintain high confidence under this structure-less view, SDCD effectively suppresses the texture-driven bias. Experimental results demonstrate that SDCD significantly mitigates hallucinations across multiple benchmarks and enhances the overall multimodal capabilities of LVLMs.

cs.CV↗

arXiv2025

VEGAS: Mitigating Hallucinations in Large Vision-Language Models via Vision-Encoder Attention Guided Adaptive Steering

Large vision-language models (LVLMs) exhibit impressive ability to jointly reason over visual and textual inputs. However, they often produce outputs that are linguistically fluent but factually inconsistent with the visual evidence, i.e., they hallucinate. Despite growing efforts to mitigate such hallucinations, a key question remains: what form of visual attention can effectively suppress hallucinations during decoding? In this work, we provide a simple answer: the vision encoder's own attention map. We show that LVLMs tend to hallucinate when their final visual-attention maps fail to concentrate on key image objects, whereas the vision encoder's more concentrated attention maps substantially reduce hallucinations. To further investigate the cause, we analyze vision-text conflicts during decoding and find that these conflicts peak in the language model's middle layers. Injecting the vision encoder's attention maps into these layers effectively suppresses hallucinations. Building on these insights, we introduce VEGAS, a simple yet effective inference-time method that integrates the vision encoder's attention maps into the language model's mid-layers and adaptively steers tokens which fail to concentrate on key image objects. Extensive experiments across multiple benchmarks demonstrate that VEGAS consistently achieves state-of-the-art performance in reducing hallucinations.

cs.CV↗

arXiv2025

ToMA: Token Merge with Attention for Diffusion Models

Diffusion models excel in high-fidelity image generation but face scalability limits due to transformers' quadratic attention complexity. Plug-and-play token reduction methods like ToMeSD and ToFu reduce FLOPs by merging redundant tokens in generated images but rely on GPU-inefficient operations (e.g., sorting, scattered writes), introducing overheads that negate theoretical speedups when paired with optimized attention implementations (e.g., FlashAttention). To bridge this gap, we propose Token Merge with Attention (ToMA), an off-the-shelf method that redesigns token reduction for GPU-aligned efficiency, with three key contributions: 1) a reformulation of token merge as a submodular optimization problem to select diverse tokens; 2) merge/unmerge as an attention-like linear transformation via GPU-friendly matrix operations; and 3) exploiting latent locality and sequential redundancy (pattern reuse) to minimize overhead. ToMA reduces SDXL/Flux generation latency by 24%/23%, respectively (with DINO $Δ< 0.07$), outperforming prior methods. This work bridges the gap between theoretical and practical efficiency for transformers in diffusion. Code available at https://github.com/WenboLuu/ToMA.

cs.LG↗

arXiv2025

The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimization

As the adoption of Generative AI in real-world services grow explosively, energy has emerged as a critical bottleneck resource. However, energy remains a metric that is often overlooked, under-explored, or poorly understood in the context of building ML systems. We present the ML$.$ENERGY Benchmark, a benchmark suite and tool for measuring inference energy consumption under realistic service environments, and the corresponding ML$.$ENERGY Leaderboard, which have served as a valuable resource for those hoping to understand and optimize the energy consumption of their generative AI services. In this paper, we explain four key design principles for benchmarking ML energy we have acquired over time, and then describe how they are implemented in the ML$.$ENERGY Benchmark. We then highlight results from the early 2025 iteration of the benchmark, including energy measurements of 40 widely used model architectures across 6 different tasks, case studies of how ML design choices impact energy consumption, and how automated optimization recommendations can lead to significant (sometimes more than 40%) energy savings without changing what is being computed by the model. The ML$.$ENERGY Benchmark is open-source and can be easily extended to various customized models and application scenarios.

cs.LG↗

arXiv2025

HilbertA: Hilbert Attention for Image Generation with Diffusion Models

Designing sparse attention for diffusion transformers requires reconciling two-dimensional spatial locality with GPU efficiency, a trade-off that current methods struggle to achieve. Existing approaches enforce two-dimensional spatial locality but often incur uncoalesced memory access. We present HilbertA, a 2D-aware and GPU-efficient sparse attention mechanism. HilbertA reorders image tokens along Hilbert curves to achieve a contiguous memory layout while preserving spatial neighborhoods, and employs a sliding schedule across layers to enable long-range information propagation without repeated or uncoalesced memory access. To further enhance cross-tile communication and positional awareness, HilbertA introduces a small central shared region. Implemented in Triton, HilbertA delivers comparable image quality with significant acceleration over prior methods on Flux.1-dev, demonstrating the feasibility of hardware-aligned two-dimensional sparse attention for high-resolution image generation. HilbertA delivers attention speedups of $2.3\times$ when generating $1024\times 1024$ images, and up to $4.17\times$ at $2048\times 2048$, while achieving image quality comparable to or surpassing baselines.

cs.AI↗