Searcharxiv⌕ Search

arXiv subjects

Yi Wang

Publications and source records attributed to Yi Wang.

At least 91 records · Page 5Linked to original sources
arXiv2026

Why Is Cubic-Phase Airy Beamforming Sufficient for Blockage Recovery?

Blockage is a critical challenge for near-field communications, where reliable transmission depends heavily on the line-of-sight (LoS) path and can suffer severe power degradation when that path is obstructed. Near-field Airy beams offer a promising solution for blockage mitigation by forming curved trajectories that guide energy around obstacles, and can be practically generated with phased arrays by imposing a cubic source phase. However, trajectory-based interpretations explain how Airy beams propagate, but not why cubic-phase Airy beamforming is sufficient for blockage recovery or how much received-power gain the cubic term itself contributes. To answer these questions, we identify the blockage-induced phase mismatch relative to conventional near-field focusing and quantify how successive phase orders compensate it. The resulting analysis reveals that the linear and quadratic degrees of freedom, originally used to compensate the free-space geometric phase, can be reoptimized under blockage to provide resteering and refocusing, respectively. The quadratic term can compensate the dominant quadratic component of the additional mismatch, while the Airy cubic provides the first independent correction to the remaining non-quadratic mismatch. Simulations show that linear and quadratic compensation recover most of the available gain. The Airy cubic adds only $0.1433$ dB on average, yet enables the cubic-phase family to attain $99.77\%$ of the phase-only upper bound. Residual-phase analysis further determines when the remaining higher-order components are negligible within a prescribed received-power tolerance. These results explain why cubic-phase Airy beamforming is sufficient: lower-order phase terms provide most of the recovery, while the cubic term closes nearly all of the remaining gap.

eess.SP↗
arXiv2026

Cosmological Collider Signals at Strong Mixing

We study cosmological collider signatures in a two-field inflationary system with constant-turn derivative mixing between the canonically normalized curvature fluctuation and a massive isocurvature field. Building on recently derived exact hypergeometric solutions that treat the quadratic mixing nonperturbatively, we construct the mixed propagators constituting the cosmological correlators. A Mellin-Barnes approach helps isolate the pair of nonanalytic late-time branches carried by a heavy field. For a cubic isocurvature self-interaction, these branches contribute towards a squeezed bispectrum with a power-law envelope and logarithmic oscillations. The oscillation frequency encodes the heavy mass, while the amplitude and phase retain the full mixing dependence rather than a perturbative expansion. We perform the squeezed limit integral analytically and compare it with numerical results across representative masses and mixing strengths. At fixed mixing we derive the large-mass JWKB expansion of the squeezed correlator and its phase. We also give integral representations for the isocurvature Wightman function. The research idea was suggested by ARC, the calculation was performed by GPT, and the results were checked by the authors.

hep-th↗
arXiv2026

SeqAlign3DVG: A Sequence-Aligned Benchmark and Voxel Reasoning Framework for 3D Visual Grounding

Image-based 3D visual grounding is critical for embodied agents, yet existing benchmarks suffer from loose text-observation alignment and neglect temporal ordering. We introduce SeqAlign3DVG, a novel benchmark dedicated to temporally ordered and strictly observation-aligned image-based 3D visual grounding. Unlike prior works using order-agnostic views or global point clouds, SeqAlign3DVG ensures all expressions are human-verified and strictly grounded in the provided RGB observations (single frames or ordered observation sequences). It comprises 9,622 single-view and 14,493 sequence samples featuring rich descriptions, complex relations, and multi-instance ambiguities. To tackle this benchmark, we propose a unified voxel-based pipeline featuring Relevance-Ordered Voxel Memory (ROVM) and Progressive Language-Voxel Fusion (PLVF). ROVM dynamically ranks and aggregates multi-view evidence via a conservative memory to mitigate noisy observations, while PLVF performs coarse-to-fine spatial-linguistic reasoning for precise disambiguation. Our approach achieves state-of-the-art performance under the depth-free protocol, significantly improving localization for targets defined by complex relations and appearance cues.

cs.CV↗
arXiv2026

GUIDE: Generative Unsupervised Chinese Query Correction via Phonetic and Visual Shared-ID Encoding

Chinese query correction (CQC) is important for search and query recommendation on content platforms, but supervised methods rely on large annotated correction pairs that are costly to maintain as query vocabularies evolve. Unsupervised correction with language models is attractive, yet in the short-query setting, unconstrained generation often over-corrects ambiguous inputs toward high-frequency phrases, causing intent drift. We propose \textsc{GUIDE}, a generative unsupervised framework for CQC based on a confuse-then-clarify paradigm. \textsc{GUIDE} encodes phonetically or visually confusable characters with shared-IDs and reconstructs the original query with an encoder--decoder architecture, which constrains correction to plausible confusion neighborhoods while learning from unlabeled query streams. A time-decayed, query-frequency-weighted objective further supports adaptation to rapidly changing query vocabularies. Experiments on \textit{QSpell 250K} and a large-scale real-world dataset (\textit{KwaiSearch}) show that \textsc{GUIDE} consistently outperforms strong baselines, while online A/B testing further confirms gains in correction quality and downstream engagement.

cs.CL↗
arXiv2026

Geometry-guided Emotion Modulation for Controllable and Photorealistic Emotional Talking Face Generation

Audio-driven emotional talking face generation aims to synthesize realistic videos with expressive facial dynamics. However, existing methods struggle to balance controllability and visual fidelity. Although implicit representations capture rich semantics, they lack structural guidance, often resulting in averaged emotional expressions. In contrast, explicit geometric methods offer better control over facial expressions but tend to sacrifice high-frequency texture details. To address it, we propose GemTalk, a diffusion-based framework that combines the semantic richness of implicit representations with the structural precision of explicit geometric priors. We introduce a Vision-guided Audio Emotion Projection (V-AEP) module to extract implicit emotional lip and expression features. At the same time, a Diffusion-based Geometric Priors Generator (D-GPG) generates identity-aware blendshape coefficients as explicit structural priors. Crucially, our Geometry-guided Emotion Modulation (GEM) module leverages these geometric priors to recalibrate the magnitude of implicit features, enabling precise, continuous control over emotional expressions, especially emotion intensity, without sacrificing visual quality. Extensive experiments show GemTalk achieves superior performance in photo-realism, and facial emotional dynamics.

cs.CV↗
arXiv2026

Mixture of Multicenter Experts in Multimodal AI for Debiased Radiotherapy Target Delineation

Clinical decision-making reflects diverse strategies shaped by regional patient populations and institutional protocols. However, most existing medical artificial intelligence (AI) models are trained on highly prevalent data patterns, which reinforces biases and fails to capture the breadth of clinical expertise. Inspired by the recent advances in Mixture of Experts (MoE), we propose a Mixture of Multicenter Experts (MoME) framework to address AI bias in the medical domain without requiring data sharing across institutions. MoME integrates specialized expertise from diverse clinical strategies to enhance model generalizability and adaptability across medical centers. We validate this framework using a multimodal target volume delineation model for prostate cancer radiotherapy. With few-shot training that combines imaging and clinical notes from each center, the model outperformed baselines, particularly in settings with high inter-center variability or limited data availability. Furthermore, MoME enables model customization to local clinical preferences without cross-institutional data exchange, making it especially suitable for resource-constrained settings while promoting broadly generalizable medical AI.

eess.IV↗
arXiv2026

How Far Should Tokenization Go? Predictive Effectiveness and Relational Losslessness

GPT-style models have achieved remarkable success with finite vocabularies of reusable tokens, making the token interface a central component of modern sequence modeling. Symbolic music appears naturally compatible with this paradigm: it consists of discrete note events and recurring structures such as chords, motifs, and phrases. However, when tokenization moves beyond language, the interface must be specified for each domain. Existing work offers many effective designs, but no unified criterion for deciding what tokenization should represent and how far it should go. Using predictive codelength as a common criterion, we formulate the Effectiveness--Losslessness Framework to define where tokenization should begin and where it should end. The Fact--Token Boundary marks where observation-determined structure should enter the token interface, through operations such as coordinate construction. Within this interface, the resulting carrier may be reversibly recoded without changing the represented facts. The Token--State Boundary marks where tokenization should stop: relations that depend on context should remain for model-state computation rather than being fixed in advance by the tokenizer. We validate the framework through controlled multi-seed symbolic-music experiments, with an independent-corpus replication of the temporal intervention. Making musical time explicit consistently reduces predictive code and also improves pitch and duration prediction, while tonal-frame canonicalization and pitch factorization provide further gains. Fixed circle-of-fifths pitch coordinates instead increase predictive code, suggesting that imposing a fixed pitch relation before context can burden prediction. Reversible BPE substantially shortens the carrier but increases predictive codelength in every seed, showing that carrier compaction alone does not guarantee predictive gain.

cs.LG↗
arXiv2026

Node-wise Feature Encoding for Neural Performance Prediction

As neural networks are increasingly deployed on resource constrained edge devices, accurate prediction of latency and energy is critical for efficient neural architecture search. Existing GNN and transformer based predictors achieve strong results but largely ignore node-level computational cost, limiting their ability to model performance critical operations. To address this, we introduce FeatureFormer, a neural performance predictor that incorporates explicit node-wise encodings of FLOPs, parameter counts, and memory proxies within a gated graph attention architecture. We also present NNEQ, a new large-scale energy consumption dataset that enables unified evaluation of latency and energy prediction. Extensive experiments demonstrate that FeatureFormer achieves state-of-the-art performance across both metrics, including challenging out-of-domain settings. Finally, we show that the proposed encoding is broadly applicable and consistently improves existing predictors with negligible overhead.

cs.LG↗
arXiv2026

Real ideal points, Conway spheres, and left-orderable Dehn fillings

We study real ideal points of $SL_2(\mathbb C)$-character varieties of knot exteriors containing essential Conway spheres. Under explicit deformation-theoretic hypotheses on the two complementary tangles, we show that essential Conway spheres are detected via Culler-Shalen theory with real ideal points; such ideal points are then deformed into arcs of $SL_2(\mathbb R)$ representations on both sides. We then compute the asymptotic behavior of the representations on these arcs and compute the translation numbers of their lifts to $\widetilde{PSL}_2(\mathbb R)$ representations. Combining this with a result of Gao, we conclude that for certain knots with essential Conway spheres, all sufficiently large positive and negative rational fillings have left-orderable fundamental group. This verifies the $L$-space conjecture for large-slope Dehn fillings of knots which were previously unknown in the literature. We verify the criteria for three infinite families of knots assembled from torus-trivial and twist-trivial tangles, then count the resulting real ideal points, determine the longitudinal translation numbers of all constructed branches, and identify exactly the zero-translation arcs. Computations for the $28$ distinct verified census exteriors listed in Table \ref{tab:instances} agree with the certified branches.

math.GT↗
arXiv2026

A proof of the Arnold-Givental conjecture

We prove the Arnold-Givental conjecture in full generality: given a closed symplectic manifold $(X, ω)$, an anti-symplectic involution $τ_X: X \to X$ with fixed point set $L={\rm Fix}(τ_X)$, and a Hamiltonian diffeomorphism $ϕ: X \to X$ such that $ϕ(L)$ intersects transversely with $L$, the following inequality holds: \[ \# \big( ϕ(L) \cap L \big) \geq {\mathrm dim}_{{\mathbb F}_2} H_*(L; {\mathbb F}_2).\] The proof combines the methods of integral Floer theory of the first and fourth authors, a reduction to Hamiltonian Floer cohomology due to Lu, and a new idea related to localization in a $\mathbb Z/2$-equivariant Floer theory tailored to the problem.

math.SG↗
arXiv2026

A General-Purpose Molecular Foundation Model Transfers Across Diverse Olfactory Tasks

Foundation models have transformed molecular property prediction, yet it remains unclear whether a molecular foundation model, fine-tuned on a single canonical olfactory prediction task, can learn representations that transfer across diverse machine olfaction problems. We investigate this question by fine-tuning Uni-Mol2 on the GS-LF benchmark for multi-label odor descriptor prediction and evaluating the resulting model, without additional deep-learning training, on four complementary downstream settings: cross-dataset odor descriptor prediction, odorous-versus-odorless classification, enantiomer evaluation, and odor mixture discriminability. The fine-tuned model matches or exceeds the performance of the state-of-the-art olfaction-specific baseline on the primary GS-LF benchmark and consistently transfers across these downstream evaluations. The enantiomer analysis further shows that three-dimensional molecular representations distinguish mirror-image molecules in a way that two-dimensional graph models fundamentally cannot, although accurately predicting the perceptual consequences of stereochemistry remains an open challenge. Together, these results support a train-once, transfer-across-tasks paradigm for machine olfaction and suggest that chemically pretrained molecular representations provide a strong foundation for transferable olfactory prediction.

cs.LG↗
arXiv2026

EgoErrorVQA: Assess Egocentric Comprehension Capabilities through Procedural Errors for Ego-Agentic AI

The majority of our everyday activities are procedural and consist of sequences of interdependent steps. However, existing benchmarks for Visual Agents and Visual Language Models (VLMs) overlook the evaluation of their procedural comprehension ability from an egocentric visual perspective, particularly for detecting procedural errors, a critical capability for everyday assistance. To bridge this gap, the EgoErrorVQA task is firstly proposed for egocentric procedural comprehension with explicit procedural errors modeling. Besides, we develop a user-friendly evaluator agent based on the Agent2Agent (A2A) protocol, enabling rigorous and standardized evaluation of visual agents through VQA-based interaction. A range of models are evaluated using both open-ended and multiple-choice questions, revealing persistent weaknesses in handling procedural errors and error types. Moreover, we introduce Ego-ADR, an Adaptive Decoupled Reasoning framework that decouples complex procedural reasoning to enhance models' understanding of procedural errors. It achieves consistent performance gains over the selected baselines and attains state-of-the-art results on several metrics under comparable settings. Code: https://github.com/z1oong/EgoErrorVQA

cs.CV↗
arXiv2026

Asymptotic behavior toward viscous shock for the outflow problem of barotropic Navier-Stokes equations

We study the time-asymptotic stability of viscous shock profile for the outflow problem of the barotropic Navier-Stokes equations on the half-line. We consider the case where the far-field state, as the right end state of the 2-Hugoniot curve, belongs to the subsonic region or the transonic curve. We employ the method of $a$-contraction with shifts to prove that, if both the shock strength and the initial perturbation are suitably small, and the viscous shock is far from the outflow boundary, then the solution asymptotically converges to the viscous shock up to a dynamical shift. We also prove that the speed of the time-dependent shift decays to zero as time goes to infinity, so that the shifted viscous shock still retains its original profile asymptotically. Since the outflow problem in the Lagrangian mass coordinate leads to a free-boundary problem due to the absence of a boundary condition for the fluid density, we consider the outflow problem in the original Eulerian coordinate instead. Although the method of $a$-contraction with shifts is technically more complicated in the Eulerian coordinate than in the Lagrangian one, this provides a more favorable framework by avoiding the difficulties arising from a free boundary. Note that this is the first result on the time-asymptotic stability of viscous shock for the outflow problem of the Navier-Stokes equations.

math.AP↗
arXiv2026

VideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding

Recent advances in video understanding have spanned motion, long video, and streaming interaction, driving this field toward real-world applications. Despite this progress, current open-source models remain limited in several ways. They often struggle to generalize across diverse video types, making them effective only in specific domains. High computational demands further restrict their efficiency and scalability. Moreover, most models are only partially open, with key components such as training code, strategy, or datasets unavailable, which hinders reproducibility and slows community-driven development. To address these issues, we introduce VideoChat3, a fully open, efficient, and generalist video-centric MLLM. VideoChat3 advances video understanding through two complementary designs. For efficiency, we introduce Inflated 3D Vision Transformer (I3D-ViT) and Adaptive Frame Resolution for Streaming Video Perception, which enables efficient spatiotemporal representation and reduces the cost of processing video inputs during training and inference. For effectiveness, we develop a scalable video data synthesis pipeline that curates three diverse, high-quality training datasets: VideoChat3-Academic2M, VideoChat3-LV116K, and VideoChat3-OL617K, covering general, long-form, and streaming video scenarios, improving the model's generalization across domains. By integrating these designs, VideoChat3 achieves a rare balance of broad generalization and computational efficiency. Experiments across general, long-form, and streaming benchmarks demonstrate that VideoChat3 surpasses prior open-source models with equal or larger parameter counts with only 4B parameters and higher efficiency.

cs.CV↗
arXiv2026

Existence of Large Boundary Layer Solutions for the Outflow Problem of Full Compressible Navier-Stokes Equations

We investigate the existence and non-existence of large-amplitude boundary layer solutions to the outflow problem for the one-dimensional full compressible Navier-Stokes equations on the half-line $\mathbb{R}_+$. Through a delicate global phase-plane analysis, we substantially extend the small-amplitude boundary layer solutions obtained via the center-manifold approach to the large-amplitude regime. Based on the sign of $M_+-1$ (with $M_+$ denoting the Mach number at the right end state) and the Prandtl-number-related parameter $ζ$, we provide a complete characterization of the existence and non-existence of large-amplitude boundary layer solutions for this half-space outflow problem. In sharp contrast to the corresponding inflow problem, boundary layer solutions to the outflow problem are permitted to be non-monotone, and their existence region in the phase plane can be an open set.

math.AP↗
arXiv2026

An Erdős--Ko--Rado theorem for cross-intersecting families in the Euclidean inner product

Let $\binom{[n]}{k}$ be the set of all $k$-element subsets of the set $\{1,\ldots,n\}$ and let $\mathcal A,\mathcal B \subseteq \binom{[n]}{k}$ be two cross-intersecting families, that is, $A\cap B\neq \emptyset$ for any $A\in \mathcal A$ and $B\in \mathcal B$. The classical cross-intersecting version of the Erdős--Ko--Rado theorem, due to Pyber and Matsumoto--Tokushige, states that if $n\geq 2k$, then $|\mathcal A||\mathcal B|\leq \binom{n-1}{k-1}^2,$ where the equality holds for $n>2k$ if and only if $\mathcal A=\mathcal B$ is a star. In the present paper, we first give a stability result of this theorem by using Filmus's FKN theorem on the slice and linear algebra method as follows: There exists a constant $C>1$ such that if $n\geq 2.07k$ and $|\mathcal A||\mathcal B|\geq (1-ε)\binom{n-1}{k-1}^2$, where $ε\leq \frac{k^2}{C^2 n ^2 }$, then there is a star $\mathcal{S}$ such that $|\mathcal{S} Δ\mathcal A|\leq C ε\binom{n}{k}$ and $|\mathcal{S} Δ\mathcal B|\leq C ε\binom{n}{k}.$ Moreover, based on this stability result and the eigenvalues of the matrices of the Johnson scheme, we present an Erdős--Ko--Rado theorem for cross-intersecting families in the Euclidean inner product showing that if $n\geq 2k$ and $k\geq d \geq 0$, then $$\big\langle\mathbf{v}_d(\mathcal A),\mathbf{v}_d(\mathcal B)\big\rangle \leq \frac{\binom{k}{d}\binom{k-1}{d}}{\binom{n-1}{d}}\binom{n-1}{k-1}^2 +\binom{k-1}{d-1} \binom{n-d-1}{k-d}\binom{n-1}{k-1},$$ together with uniqueness and a corresponding stability result, where $\mathbf{v}_d(\mathcal A) \in \mathbb R^{\binom{[n]}{d}}$ is the $d$-degree vector of $\mathcal A$ whose $U$-entry is the number of members in $\mathcal A$ containing $U$.

math.CO↗
arXiv2026

Extremal graphs for a conjecture on the square energy of graphs

For a graph $G$, let $s^+(G)$ and $s^-(G)$ denote the sums of the squares of its positive and negative adjacency eigenvalues. We determine all equality cases in the conjecture of Elphick, Farber, Goldberg, and Wocjan that every connected graph $G$ on $n$ vertices satisfies \[ \min \{s^+(G),s^-(G)\}\ge n-1. \] Namely, equality for $s^+$ holds exactly for trees, whereas equality for $s^-$ holds exactly for trees and complete graphs. The proof combines the $P_3$-removal lemma in the no-cut-vertex case with a detailed equality analysis of the underlying doubly nonnegative matrix inequality. Every block is forced to be complete, and a minimal-counterexample argument gives an exact rank-one decomposition of the folded matrix $M^c$. The resulting non-edge vanishings, together with $AX=XA$, rule out an interface between a bridge and a nontrivial block.

math.CO↗
arXiv2026

DR.GAP: Mitigating Bias in Large Language Models using Gender-Aware Prompting with Decoupled Reasoning

Large Language Models (LLMs) exhibit strong natural language understanding capabilities but also inherit and amplify societal biases, particularly gender bias, raising fairness concerns. Existing prompt-based debiasing strategies share a key limitation: they fail to disentangle gender information from task semantics. Bias steering compels models to overemphasize gender cues, while reasoning-based prompting induces gender-biased reasoning chains. To address these challenges, we propose DR.GAP (Decoupled Reasoning for Gender-Aware Prompting), an automated and model-agnostic pipeline that mitigates gender bias while preserving model performance. DR.GAP generates gender-neutral reasoning traces and applies them as in-context demonstrations during inference, effectively decoupling gender attributes from task semantics without modifying model parameters. Extensive experiments on coreference resolution and question-answering tasks across six LLMs demonstrate DR.GAP's effectiveness, generalizability, and robustness, supported by detailed mechanism analyses. Moreover, DR.GAP can be extended to vision-language models (VLMs), achieving substantial bias reduction.

cs.CL↗