arXiv subjects

Yi Yang

Publications and source records attributed to Yi Yang.

At least 271 records · Page 15Linked to original sources

arXiv2025

Merging without Forgetting: Continual Fusion of Task-Specific Models via Optimal Transport

Merging models fine-tuned for different tasks into a single unified model has become an increasingly important direction for building versatile, efficient multi-task systems. Existing approaches predominantly rely on parameter interpolation in weight space, which we show introduces significant distribution shift in the feature space and undermines task-specific knowledge. In this paper, we propose OTMF (Optimal Transport-based Masked Fusion), a novel model merging framework rooted in optimal transport theory to address the distribution shift that arises from naive parameter interpolation. Instead of directly aggregating features or weights, OTMF aligns the semantic geometry of task-specific models by discovering common masks applied to task vectors through optimal transport plans. These masks selectively extract transferable and task-agnostic components while preserving the unique structural identities of each task. To ensure scalability in real-world settings, OTMF further supports a continual fusion paradigm that incrementally integrates each new task vector without revisiting previous ones, maintaining a bounded memory footprint and enabling efficient fusion across a growing number of tasks. We conduct comprehensive experiments on multiple vision and language benchmarks, and results show that OTMF achieves state-of-the-art performance in terms of both accuracy and efficiency. These findings highlight the practical and theoretical value of our approach to model merging.

cs.LG↗

arXiv2025

Multi-Text Guided Few-Shot Semantic Segmentation

Recent CLIP-based few-shot semantic segmentation methods introduce class-level textual priors to assist segmentation by typically using a single prompt (e.g., a photo of class). However, these approaches often result in incomplete activation of target regions, as a single textual description cannot fully capture the semantic diversity of complex categories. Moreover, they lack explicit cross-modal interaction and are vulnerable to noisy support features, further degrading visual prior quality. To address these issues, we propose the Multi-Text Guided Few-Shot Semantic Segmentation Network (MTGNet), a dual-branch framework that enhances segmentation performance by fusing diverse textual prompts to refine textual priors and guide the cross-modal optimization of visual priors. Specifically, we design a Multi-Textual Prior Refinement (MTPR) module that suppresses interference and aggregates complementary semantic cues to enhance foreground activation and expand semantic coverage for structurally complex objects. We introduce a Text Anchor Feature Fusion (TAFF) module, which leverages multi-text embeddings as semantic anchors to facilitate the transfer of discriminative local prototypes from support images to query images, thereby improving semantic consistency and alleviating intra-class variations. Furthermore, a Foreground Confidence-Weighted Attention (FCWA) module is presented to enhance visual prior robustness by leveraging internal self-similarity within support foreground features. It adaptively down-weights inconsistent regions and effectively suppresses interference in the query segmentation process. Extensive experiments on standard FSS benchmarks validate the effectiveness of MTGNet. In the 1-shot setting, it achieves 76.8% mIoU on PASCAL-5i and 57.4% on COCO-20i, with notable improvements in folds exhibiting high intra-class variations.

cs.CV↗

arXiv2025

Melodia: Training-Free Music Editing Guided by Attention Probing in Diffusion Models

Text-to-music generation technology is progressing rapidly, creating new opportunities for musical composition and editing. However, existing music editing methods often fail to preserve the source music's temporal structure, including melody and rhythm, when altering particular attributes like instrument, genre, and mood. To address this challenge, this paper conducts an in-depth probing analysis on attention maps within AudioLDM 2, a diffusion-based model commonly used as the backbone for existing music editing methods. We reveal a key finding: cross-attention maps encompass details regarding distinct musical characteristics, and interventions on these maps frequently result in ineffective modifications. In contrast, self-attention maps are essential for preserving the temporal structure of the source music during its conversion into the target music. Building upon this understanding, we present Melodia, a training-free technique that selectively manipulates self-attention maps in particular layers during the denoising process and leverages an attention repository to store source music information, achieving accurate modification of musical characteristics while preserving the original structure without requiring textual descriptions of the source music. Additionally, we propose two novel metrics to better evaluate music editing methods. Both objective and subjective experiments demonstrate that our approach achieves superior results in terms of textual adherence and structural integrity across various datasets. This research enhances comprehension of internal mechanisms within music generation models and provides improved control for music creation.

cs.SD↗

arXiv2025

Bayes-Optimal Fair Classification with Multiple Sensitive Features

Existing theoretical work on Bayes-optimal fair classifiers usually considers a single (binary) sensitive feature. In practice, individuals are often defined by multiple sensitive features. In this paper, we characterize the Bayes-optimal fair classifier for multiple sensitive features under general approximate fairness measures, including mean difference and mean ratio. We show that these approximate measures for existing group fairness notions, including Demographic Parity, Equal Opportunity, Predictive Equality, and Accuracy Parity, are linear transformations of selection rates for specific groups defined by both labels and sensitive features. We then characterize that Bayes-optimal fair classifiers for multiple sensitive features become instance-dependent thresholding rules that rely on a weighted sum of these group membership probabilities. Our framework applies to both attribute-aware and attribute-blind settings and can accommodate composite fairness notions like Equalized Odds. Building on this, we propose two practical algorithms for Bayes-optimal fair classification via in-processing and post-processing. We show empirically that our methods compare favorably to existing methods.

stat.ML↗

arXiv2025

PCA++: How Uniformity Induces Robustness to Background Noise in Contrastive Learning

High-dimensional data often contain low-dimensional signals obscured by structured background noise, which limits the effectiveness of standard PCA. Motivated by contrastive learning, we address the problem of recovering shared signal subspaces from positive pairs, paired observations sharing the same signal but differing in background. Our baseline, PCA+, uses alignment-only contrastive learning and succeeds when background variation is mild, but fails under strong noise or high-dimensional regimes. To address this, we introduce PCA++, a hard uniformity-constrained contrastive PCA that enforces identity covariance on projected features. PCA++ has a closed-form solution via a generalized eigenproblem, remains stable in high dimensions, and provably regularizes against background interference. We provide exact high-dimensional asymptotics in both fixed-aspect-ratio and growing-spike regimes, showing uniformity's role in robust signal recovery. Empirically, PCA++ outperforms standard PCA and alignment-only PCA+ on simulations, corrupted-MNIST, and single-cell transcriptomics, reliably recovering condition-invariant structure. More broadly, we clarify uniformity's role in contrastive learning, showing that explicit feature dispersion defends against structured noise and enhances robustness.

stat.ML↗

arXiv2025

An axisymmetric shock breakout indicated by prompt polarized emission from the type II supernova 2024ggi

The death of massive stars is triggered by an infall-induced bounce shock that disrupts the star. How such a shock is launched and propagates through the star is a decade-long puzzle. Some models assume that the shock can be reenergized by absorbing neutrinos, leading to highly aspherical explosions. Other models involve jet-powered shocks that lead to bipolar explosions reflected in the geometry of the shock-breakout emission. We report measurement of the geometry of the shock breakout through unprecedentedly early spectropolarimetry of the nearby type II supernova 2024ggi starting ~1.2 days after the explosion. The measurement indicates a well-defined symmetry axis of the shock breakout, which is also shared by the hydrogen-rich envelope that emerged after the circumstellar matter was engulfed by the ejecta, revealing a persisting and prominent symmetry axis throughout the explosion. These findings suggest that the physical mechanism driving the explosion of massive stars manifests a well-defined axial symmetry and acts on large scales.

astro-ph.SR↗

arXiv2025

Beyond Uniform Deletion: A Data Value-Weighted Framework for Certified Machine Unlearning

As the right to be forgotten becomes legislated worldwide, machine unlearning mechanisms have emerged to efficiently update models for data deletion and enhance user privacy protection. However, existing machine unlearning algorithms frequently neglect the fact that different data points may contribute unequally to model performance (i.e., heterogeneous data values). Treat them equally in machine unlearning procedure can potentially degrading the performance of updated models. To address this limitation, we propose Data Value-Weighted Unlearning (DVWU), a general unlearning framework that accounts for data value heterogeneity into the unlearning process. Specifically, we design a weighting strategy based on data values, which are then integrated into the unlearning procedure to enable differentiated unlearning for data points with varying utility to the model. The DVWU framework can be broadly adapted to various existing machine unlearning methods. We use the one-step Newton update as an example for implementation, developing both output and objective perturbation algorithms to achieve certified unlearning. Experiments on both synthetic and real-world datasets demonstrate that our methods achieve superior predictive performance and robustness compared to conventional unlearning approaches. We further show the extensibility of our framework on gradient ascent method by incorporating the proposed weighting strategy into the gradient terms, highlighting the adaptability of DVWU for broader gradient-based deep unlearning methods.

cs.LG↗

arXiv2025

Breaking the Modality Barrier: Generative Modeling for Accurate Molecule Retrieval from Mass Spectra

Retrieving molecular structures from tandem mass spectra is a crucial step in rapid compound identification. Existing retrieval methods, such as traditional mass spectral library matching, suffer from limited spectral library coverage, while recent cross-modal representation learning frameworks often encounter modality misalignment, resulting in suboptimal retrieval accuracy and generalization. To address these limitations, we propose GLMR, a Generative Language Model-based Retrieval framework that mitigates the cross-modal misalignment through a two-stage process. In the pre-retrieval stage, a contrastive learning-based model identifies top candidate molecules as contextual priors for the input mass spectrum. In the generative retrieval stage, these candidate molecules are integrated with the input mass spectrum to guide a generative model in producing refined molecular structures, which are then used to re-rank the candidates based on molecular similarity. Experiments on both MassSpecGym and the proposed MassRET-20k dataset demonstrate that GLMR significantly outperforms existing methods, achieving over 40% improvement in top-1 accuracy and exhibiting strong generalizability.

cs.LG↗

arXiv2025

Multi-Object Tracking Retrieval with LLaVA-Video: A Training-Free Solution to MOT25-StAG Challenge

In this report, we present our solution to the MOT25-Spatiotemporal Action Grounding (MOT25-StAG) Challenge. The aim of this challenge is to accurately localize and track multiple objects that match specific and free-form language queries, using video data of complex real-world scenes as input. We model the underlying task as a video retrieval problem and present a two-stage, zero-shot approach, combining the advantages of the SOTA tracking model FastTracker and Multi-modal Large Language Model LLaVA-Video. On the MOT25-StAG test set, our method achieves m-HIoU and HOTA scores of 20.68 and 10.73 respectively, which won second place in the challenge.

cs.CV↗

arXiv2025

Revisited for existence proof of optimal solution in Bernoulli free boundary problem using an energy-gap cost functional

Bernoulli free boundary problem is numerically solved via shape optimization that minimizes a cost functional subject to state problems constraints. In \cite{1}, an energy-gap cost functional was formulated based on two auxiliary state problems, with existence of optimal solution attempted through continuity of state problems with respect to the domain. Nevertheless, there exists a corrigendum in Eq.(48) in \cite{1}, where the boundedness of solution sequences for state problems with respect to the domain cannot be directly estimated via the Cauchy-Schwarz inequality as \textbf{Claimed}. In this comment, we rectify this proof by Poincaré-Friedrichs inequality.

math.AP↗

arXiv2025

Finding Non-Redundant Simpson's Paradox from Multidimensional Data

Simpson's paradox, a long-standing statistical phenomenon, describes the reversal of an observed association when data are disaggregated into sub-populations. It has critical implications across statistics, epidemiology, economics, and causal inference. Existing methods for detecting Simpson's paradox overlook a key issue: many paradoxes are redundant, arising from equivalent selections of data subsets, identical partitioning of sub-populations, and correlated outcome variables, which obscure essential patterns and inflate computational cost. In this paper, we present the first framework for discovering non-redundant Simpson's paradoxes. We formalize three types of redundancy - sibling child, separator, and statistic equivalence - and show that redundancy forms an equivalence relation. Leveraging this insight, we propose a concise representation framework for systematically organizing redundant paradoxes and design efficient algorithms that integrate depth-first materialization of the base table with redundancy-aware paradox discovery. Experiments on real-world datasets and synthetic benchmarks show that redundant paradoxes are widespread, on some real datasets constituting over 40% of all paradoxes, while our algorithms scale to millions of records, reduce run time by up to 60%, and discover paradoxes that are structurally robust under data perturbation. These results demonstrate that Simpson's paradoxes can be efficiently identified, concisely summarized, and meaningfully interpreted in large multidimensional datasets.

cs.DB↗

arXiv2025

SN 2024gy: Multi-epoch Spectroscopic Features Suggestive of Delayed Detonation in a Type Ia Supernova

We present photometric and spectroscopic observations of SN 2024gy, a Type Ia supernova (SN Ia) exhibiting high-velocity features (HVFs) in its early-time spectra. This SN reaches a peak $B$-band magnitude of $-19.25 \pm 0.29$ mag and subsequently declines by $Δm_{15}(B) \approx 1.12$ mag, consistent with the luminosity-width relation characteristic of normal SNe Ia. Based on the peak thermal luminosity of $(1.2 \pm 0.3) \times 10^{43}$ erg s$^{-1}$, we estimate that $0.57 \pm 0.14~\rm M_{\odot}$ of $^{56}$Ni was synthesized during the explosion. Our dense early spectral monitoring revealed significant velocity disparities within the ejecta. Notably, absorption features from the Ca II near-infrared triplet were observed at velocities exceeding 25,000 km s$^{-1}$, while the Si II $λ$6355 line velocity at the same epoch was significantly lower at $\sim$ 16,000 km s$^{-1}$. This velocity disparity likely reflects distinct ionization states of intermediate-mass elements in the outermost layers. The prominent Ca II HVFs may originate from ionization suppression within the highest-velocity ejecta, potentially indicative of minimal hydrogen mixing in a delayed-detonation explosion scenario. Additionally, the Ni/Fe ratio derived from the nebular spectrum of SN 2024gy provides further support for this model.

astro-ph.HE↗

arXiv2025

DeltaPhi: Physical States Residual Learning for Neural Operators in Data-Limited PDE Solving

The limited availability of high-quality training data poses a major obstacle in data-driven PDE solving, where expensive data collection and resolution constraints severely impact the ability of neural operator networks to learn and generalize the underlying physical system. To address this challenge, we propose DeltaPhi, a novel learning framework that transforms the PDE solving task from learning direct input-output mappings to learning the residuals between similar physical states, a fundamentally different approach to neural operator learning. This reformulation provides implicit data augmentation by exploiting the inherent stability of physical systems where closer initial states lead to closer evolution trajectories. DeltaPhi is architecture-agnostic and can be seamlessly integrated with existing neural operators to enhance their performance. Extensive experiments demonstrate consistent and significant improvements across diverse physical systems including regular and irregular domains, different neural architectures, multiple training data amount, and cross-resolution scenarios, confirming its effectiveness as a general enhancement for neural operators in data-limited PDE solving.

cs.LG↗

arXiv2025

Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation

Graph-based RAG constructs a knowledge graph (KG) from text chunks to enhance retrieval in Large Language Model (LLM)-based question answering. It is especially beneficial in domains such as biomedicine, law, and political science, where effective retrieval often involves multi-hop reasoning over proprietary documents. However, these methods demand numerous LLM calls to extract entities and relations from text chunks, incurring prohibitive costs at scale. Through a carefully designed ablation study, we observe that certain words (termed concepts) and their associated documents are more important. Based on this insight, we propose Graph-Guided Concept Selection (G2ConS). Its core comprises a chunk selection method and an LLM-independent concept graph. The former selects salient document chunks to reduce KG construction costs; the latter closes knowledge gaps introduced by chunk selection at zero cost. Evaluations on multiple real-world datasets show that G2ConS outperforms all baselines in construction cost, retrieval effectiveness, and answering quality.

cs.LG↗

arXiv2025

Endowing GPT-4 with a Humanoid Body: Building the Bridge Between Off-the-Shelf VLMs and the Physical World

Humanoid agents often struggle to handle flexible and diverse interactions in open environments. A common solution is to collect massive datasets to train a highly capable model, but this approach can be prohibitively expensive. In this paper, we explore an alternative solution: empowering off-the-shelf Vision-Language Models (VLMs, such as GPT-4) to control humanoid agents, thereby leveraging their strong open-world generalization to mitigate the need for extensive data collection. To this end, we present \textbf{BiBo} (\textbf{B}uilding humano\textbf{I}d agent \textbf{B}y \textbf{O}ff-the-shelf VLMs). It consists of two key components: (1) an \textbf{embodied instruction compiler}, which enables the VLM to perceive the environment and precisely translate high-level user instructions (e.g., {\small\itshape ``have a rest''}) into low-level primitive commands with control parameters (e.g., {\small\itshape ``sit casually, location: (1, 2), facing: 90$^\circ$''}); and (2) a diffusion-based \textbf{motion executor}, which generates human-like motions from these commands, while dynamically adapting to physical feedback from the environment. In this way, BiBo is capable of handling not only basic interactions but also diverse and complex motions. Experiments demonstrate that BiBo achieves an interaction task success rate of 90.2\% in open environments, and improves the precision of text-guided motion execution by 16.3\% over prior methods. The code will be made publicly available.

cs.RO↗

arXiv2025

CoMo: Compositional Motion Customization for Text-to-Video Generation

While recent text-to-video models excel at generating diverse scenes, they struggle with precise motion control, particularly for complex, multi-subject motions. Although methods for single-motion customization have been developed to address this gap, they fail in compositional scenarios due to two primary challenges: motion-appearance entanglement and ineffective multi-motion blending. This paper introduces CoMo, a novel framework for $\textbf{compositional motion customization}$ in text-to-video generation, enabling the synthesis of multiple, distinct motions within a single video. CoMo addresses these issues through a two-phase approach. First, in the single-motion learning phase, a static-dynamic decoupled tuning paradigm disentangles motion from appearance to learn a motion-specific module. Second, in the multi-motion composition phase, a plug-and-play divide-and-merge strategy composes these learned motions without additional training by spatially isolating their influence during the denoising process. To facilitate research in this new domain, we also introduce a new benchmark and a novel evaluation metric designed to assess multi-motion fidelity and blending. Extensive experiments demonstrate that CoMo achieves state-of-the-art performance, significantly advancing the capabilities of controllable video generation. Our project page is at https://como6.github.io/.

cs.CV↗

arXiv2025

Developments in the applications of density functional theory to fractional quantum Hall systems

The fractional quantum Hall effect remains a captivating area in condensed matter physics, characterized by strongly correlated topological order, which manifests as fractionalized excitations and anyonic statistics. Numerical simulations, such as exact diagonalization, density matrix renormalization group, matrix product states, and Monte Carlo methods, are essential to examine the properties of strongly correlated systems. Recently, density functional theory has been employed in this field within the framework of composite fermion theory. This paper systematically evaluates how density functional theory approaches have addressed fundamental challenges in fractional quantum Hall systems, including ground state and low-energy excitations. Special attention is given to the insights provided by density functional theory regarding composite fermion behavior, edge effects, and the nature of fractional charge and magnetoroton excitations. The discussion critically examines both the advantages and limitations of these approaches, while highlighting the productive interplay between numerical simulations and theoretical models. Future directions are explored, particularly the promising potential of time-dependent density functional theory for modeling non-equilibrium dynamics in quantum Hall systems.

cond-mat.str-el↗

arXiv2025

JWST and Ground-based Observations of the Type Iax Supernovae SN 2024pxl and SN 2024vjm: Evidence for Weak Deflagration Explosions

We present panchromatic optical $+$ near-infrared (NIR) $+$ mid-infrared (MIR) observations of the intermediate-luminosity Type Iax supernova (SN Iax) 2024pxl and the extremely low-luminosity SN Iax 2024vjm. JWST observations provide unprecedented MIR spectroscopy of SN Iax, spanning from $+$11 to $+$42 days past maximum light. We detect forbidden emission lines in the MIR at these early times while the optical and NIR are dominated by permitted lines with an absorption component. Panchromatic spectra at early times can thus simultaneously show nebular and photospheric lines, probing both inner and outer layers of the ejecta. We identify spectral lines not seen before in SN Iax, including [Mg II] 4.76 $μ$m, [Mg II] 9.71 $μ$m, [Ne II] 12.81 $μ$m, and isolated O I 2.76 $μ$m that traces unburned material. Forbidden emission lines of all species are centrally peaked with similar kinematic distributions, indicating that the ejecta are well mixed in both SN 2024pxl and SN 2024vjm, a hallmark of pure deflagration explosion models. Radiative transfer modeling of SN 2024pxl shows good agreement with a weak deflagration of a near-Chandrasekhar-mass white dwarf, but additional IR flux is needed to match the observations, potentially attributable to a surviving remnant. Similarly, we find SN 2024vjm is also best explained by a weak deflagration model, despite the large difference in luminosity between the two supernovae. Future modeling should push to even weaker explosions and include the contribution of a bound remnant. Our observations demonstrate the diagnostic power of panchromatic spectroscopy for unveiling explosion physics in thermonuclear supernovae.

astro-ph.HE↗