Searcharxiv⌕ Search

arXiv subjects

Lin Chen

Publications and source records attributed to Lin Chen.

At least 73 records · Page 4Linked to original sources
arXiv2026

A First Exploration of Neuromorphic OT-CFM for Multi-Speaker VSR

Visual Speech Recognition (VSR) tasks in complex multi-speaker scenarios are severely hindered by rapid head motions, occlusions, and subtle lip articulations. Traditional RGB-based methods struggle here due to low rates and motion blur of frames. To overcome these, we propose LipsFlow, a neuromorphic-inspired VSR framework that converts RGB videos into high-temporal-resolution event streams. For multi-speaker, we employ ByteTrack tracking and TalkNet active speaker detection to temporally segment scenes into single-speaker clips, enabling focused per-speaker analysis. By explicitly capturing microsecond-level articulatory dynamics via learnable event-based representations, LipsFlow achieves inherent robustness against visual degradation. To efficiently model these dense event-based features and adapt to speaker-specific articulatory patterns, we introduce Optimal Transport Conditional Flow Matching (OT-CFM). It enforces deterministic, straight-line trajectory generation in a semantic latent space, slashing inference latency to just two Ordinary Differential Equation (ODE) steps. Furthermore, we design a Dual-Level Semantic Supervision mechanism combining token-level BERT weight tying and sentence-level priors to resolve homophene ambiguities. Validated on competitive benchmarks, LipsFlow achieves a state-of-the-art WER of 22.3\% at 240 ms latency, establishing a highly robust and efficient paradigm for event-based VSR.

cs.MM↗
arXiv2026

Beyond Sequential Distance: Inter-Modal Distance Invariant Position Encoding

Despite the remarkable capabilities of Multimodal Large Language Models (MLLMs), they still suffer from visual fading in long-context scenarios. Specifically, the attention to visual tokens diminishes as the text sequence lengthens, leading to text generation detached from visual constraints. We attribute this degradation to the inherent inductive bias of Multimodal RoPE, which penalizes inter-modal attention as the distance between visual and text tokens increases. To address this, we propose inter-modal Distance Invariant Position Encoding (DIPE), a simple but effective mechanism that disentangles position encoding based on modality interactions. DIPE retains the natural relative positioning for intra-modal interactions to preserve local structure, while enforcing an anchored perceptual proximity for inter-modal interactions. This strategy effectively mitigates the inter-modal distance-based penalty, ensuring that visual signals remain perceptually consistent regardless of the context length. Experimental results demonstrate that by integrating DIPE with Multimodal RoPE, the model maintains stable visual grounding in long-context scenarios, significantly alleviating visual fading while preserving performance on standard short-context benchmarks. Code is available at https://github.com/lchen1019/DIPE.

cs.CV↗
arXiv2026

Tri-Efficient Transfer Learning for Point Cloud Videos

While point cloud foundation models have significantly advanced point cloud video understanding, existing parameter-efficient fine-tuning (PEFT) methods still suffer from two critical limitations: prohibitive annotation costs for large-scale point cloud datasets and severe memory bottlenecks. In this paper, we aim to mine richer supervision signals from existing data rather than blindly scaling datasets. A further key principle is that the memory footprint of fine-tuning must be drastically reduced compared to full fine-tuning, which remains elusive for current PEFT techniques. Driven by these challenges, we identify three core desiderata: data-, parameter-, and memory efficiency, and present PoinTriE, a unified framework that excels along all three dimensions. For pre-training, pseudo-motion trajectories are synthesized via rigid transformations, paired with text corpora and 2D projections derived from raw point clouds. We then propose a Geometric-Motion Duality Network optimized via multimodal contrastive learning, rigid rotation prediction, and motion distribution divergence to produce dense self-supervision. During fine-tuning, we freeze the pretrained backbone and only update a lightweight Spatio-temporal Side Network built with LoRA units. Equipped with a gradient flow masking strategy, PoinTriE simultaneously reduces memory consumption and parameter overhead. Extensive experiments confirm that PoinTriE establishes new state-of-the-art results on action recognition and semantic segmentation tasks.

cs.CV↗
arXiv2026

Quantifying Entanglement via Quantum Wasserstein Distances

We propose a bipartite entanglement measure defined as the minimal order-1 quantum Wasserstein distance from a state to the set of separable states. Owing to the universal data-processing inequality of the Wasserstein metric, the measure satisfies all fundamental axioms within a single geometric framework. A Lipschitz dual formulation yields explicit lower bounds for pure and mixed states, a sharp constant for two-qubit systems, and an expected value for Haar-random pure states. We further establish a quantitative connection to entanglement witnesses: any negative witness expectation value certifies a lower bound, and the dual variational bound is exactly the maximal violation achievable by a Lipschitz-1 witness. The approach naturally provides subadditivity, trace-distance estimates, and bounds on local observables, while pointing toward large-deviation conjectures. This work introduces a framework at the interface of entanglement theory, optimal transport, and experimental entanglement detection.

quant-ph↗
arXiv2026

Can We Hear from Events? Generating Speech from Event Camera

Traditional RGB-based speech generation faces Temporal Granularity Mismatch since fixed camera exposure times inevitably blur the high-frequency articulatory transients essential for rendering emotional speech. To break this ceiling, we propose EventSpeech as a novel text-conditioned framework pioneering the use of neuromorphic events for expressive speech generation, since these microsecond-precise events naturally align with acoustic waveform dynamics. Our architecture integrates a dedicated Event Encoder to model sparse neuromorphic events alongside a multi-scale Audio Encoder featuring a Hierarchical Wavelet Contextualizer (HWC). A bidirectional alignment mechanism seamlessly synchronizes linguistic content and visual dynamics with dense acoustic features. Furthermore, we construct EVT-SPK as the first benchmark comprising large-scale synthetic data and real-world recordings from specialized neuromorphic hardware. Extensive evaluations demonstrate that EventSpeech significantly outperforms current baselines by preserving fine-grained emotions and resisting motion blur to establish a new paradigm for multimodal speech generation. Code and demo are available at https://xrfang-0102.github.io/EventSpeechWeb/.

cs.MM↗
arXiv2026

Scalable Graph State Generation with O(1) Local Feedforward in Quantum Networks

The development of quantum networks faces a key challenge: the contradiction between probabilistic long-range entanglement generation and finite coherence time. Existing routing protocols typically focus on global state computation or path optimization. As the network scales up, classical delays accumulate and exacerbate decoherence, leading to a decrease in entanglement fidelity. To reduce routing decision delays to levels far below the coherence time of qubits, we propose a protocol based on local measurement and classical feedforward. This protocol reduces the local decision complexity to amortized O(1) level, ensuring that the decision delay is always much smaller than the coherence time of qubits. We map this protocol onto a dual-species trapped-ion platform and perform hybrid simulations. The results show that the proposed protocol performs well in terms of both resource efficiency and time feasibility. Noise analysis indicates that readout fidelity is the main bottleneck of this protocol, but noise suppression can be achieved by employing an erasure transformation in the dual-species architecture, combined with spatial multiplexing and branch independence, thereby ensuring the generation of high-fidelity star subgraphs. This protocol provides a clear path to achieving high-fidelity star subgraphs. These subgraphs can serve as general modules, merging to construct arbitrary subgraphs, providing a feasible solution for future fault-tolerant distributed quantum computing.

quant-ph↗
arXiv2026

ACC: Compiling Agent Trajectories for Long-Context Training

Recent development of agents has renewed demand for long-context reasoning capacity of LLMs. However, training LLMs for this capacity requires costly long-document curation or heuristic context synthesis. We observe that agents produce massive trajectories when solving problems, invoking tools and receiving environment observations across many turns. The evidence needed to answer the original question is thus scattered throughout these turns, requiring integration of distant context segments. Nevertheless, standard agent SFT masks tool responses and only trains turn-level tool selection, creating a supervision blind spot where these scattered signals go unused. We propose Agent Context Compilation (ACC), which converts trajectories from search, software engineering, and database querying agents into long-context QA pairs that combine the original question with tool responses and environment observations gathered across multiple turns, training the model to answer directly without tool use. This makes the dependencies between the question and the evidence explicit, enabling direct supervision of long-context reasoning over distant segments without additional annotation. ACC is a simple but effective approach that can be combined with any existing long-context extension or training method, providing scalable supervised fine-tuning data. We validate ACC on long-range dependency modeling tasks through MRCR and GraphWalks, challenging benchmarks requiring cross-turn coreference resolution and graph traversal over extended contexts. Training Qwen3-30B-A3B with ACC achieves 68.3 on MRCR (+18.1) and 77.5 on GraphWalks (+7.6), results comparable to Qwen3-235B-A22B, while preserving general capabilities on GPQA, MMLU-Pro, AIME, and IFEval. Further mechanism analysis reveals that the ACC-trained model exhibits task-adaptive attention restructuring and expert specialization.

cs.CL↗
arXiv2026

Construction of three-qubit positive-partial-transpose entangled states of rank four

Multiqubit positive-partial-transpose (PPT) entangled states play an important role in quantum information theory. We characterize such states of minimum rank in three-qubit system, namely rank four. Depending on whether the Lorentz invariant is zero, we classify such states into two types. The PPT entangled states constructed by unextendible product bases (UPB) have nonzero invariants, which belong to type I. We provide a method to effectively determine whether a state can be constructed from UPB. For states with zero invariant, which belong to type II, we provide an explicit expression up to equivalence of stochastic local operations and classical communications (SLOCC). It turns out that we can represent them with only one complex parameter. We further study SLOCC-equivalence relation within the expression. We also investigate the Lorentz invariants of multiqubit states with rank less than three and analyze their range.

quant-ph↗
arXiv2026

Mixed States Uniquely Determined by Marginals and Additivity

Identifying whether a mixed quantum state is uniquely determined among all states (UDA) by its local marginals is a basic problem in quantum information theory. We establish necessary and sufficient conditions under which several classes of multipartite mixed states are UDA by their $k$-partite marginals. We also prove structural properties based on ranges and marginals, and formulate a recursive procedure for the determination of UDA states. We show that sufficiently high rank rules out unique determination from fixed-order marginals, implying that almost all multipartite mixed states are not UDA from such marginals. Finally, we completely characterize the additivity of bipartite UDA states, three-qubit and several families of $n$-qubit product UDA states. These results clarify the boundary between UDA and non-UDA mixed states and provide a framework for local-marginal reconstruction and related certification tasks.

quant-ph↗
arXiv2026

Weak order one convergence of structure-preserving stochastic theta methods for stochastic differential algebraic equations with time-dependent singular matrices

This paper studies the weak convergence order of structure-preserving stochastic theta methods for a class of index-$1$ stochastic differential algebraic equations with time-dependent singular matrices. The singular matrix is allowed to vary in time but preserves a fixed differential-algebraic splitting, thereby extending the constant singular-matrix setting while retaining the projector structure required for constraint preservation. By exploiting the index-$1$ algebraic-differential decomposition of the exact solution, we establish an abstract weak convergence theorem for constraint-preserving one-step approximations and apply it to the stochastic theta method with $θ\in (0,1]$. Under global Lipschitz, linear growth, and suitable smoothness assumptions, the considered method is proved to be well posed, to preserve the algebraic constraints at all time levels, and to converge with weak order one. Numerical experiments are finally presented to confirm the structure-preserving property and the theoretical convergence order.

math.NA↗
arXiv2026

Non-Perturbative Geometric Framework for Single-Qubit Gates under Always-On Couplings

In qubit arrays with always-on couplings, single-qubit gates pose a control challenge often as demanding as entangling operations. The same interactions that enable two-qubit entanglement induce crosstalk that significantly degrades single-qubit fidelity. We present a non-perturbative analytical framework for constructing high-fidelity single-qubit gates in the presence of such couplings. From the geometric structure of SU(2) dynamics, we derive a crosstalk-suppression criterion. The dynamics must trace closed loops on a 2-sphere, with a net-zero enclosed-area condition arising when zero-detuning subspaces are present, and the pulse waveform corresponds to the geodesic curvature of the loop. Unlike previous Euclidean-geometric and perturbative dynamically-corrected-gate constructions, the framework operates on a 2-sphere whose intrinsic curvature is set by the detuning, enabling crosstalk suppression even when couplings are comparable to the drive amplitude. Noise robustness enters as an additional constraint along the same closed loop via the Magnus expansion. In two- and three-qubit Heisenberg chains, the resulting pulses are robust against fluctuations in both coupling strength and qubit frequency. Our pulses outperform a representative perturbative robust-control pulse by more than an order of magnitude in fidelity when the always-on coupling approaches the drive amplitude, where perturbative methods break down.

quant-ph↗
arXiv2026

Generating 2-Gray codes for grand Motzkin paths and grand Dyck paths with air pockets in constant amortized time

A grand Motzkin path with air pockets is a non-empty lattice path in the first and fourth quadrant of $\mathbb{Z}^2$, starting at the origin $(0,0)$, ending on the $x$-axis, and consisting of up-steps $(1, 1)$, horizontal steps $(1, 0)$, down-steps $(1, -k)$ where $k \geq 1$, and with no consecutive down-steps. A {grand Dyck path with air pockets} is a grand Motzkin path with air pockets that uses no horizontal steps. We present the first known 2-Gray codes for grand Motzkin paths with air pockets. Setting the number of horizontal steps to zero in our algorithm yields the first known 2-Gray codes for grand Dyck paths with air pockets. Our three-stage algorithm generates each path in constant amortized time per string, using $O(n^2)$ memory. We also provide enumeration formulae for grand Motzkin paths and grand Dyck paths with air pockets.

math.CO↗
arXiv2026

On the saturated cases of the distillability conjecture

The distillability conjecture for two-copy four-by-four Werner states has been an open problem in quantum information for years. We investigate the conditions under which the conjectured inequality becomes an equality. For all known cases where the conjecture has been verified, we characterize the saturation conditions and show that equality forces the matrices $A$ and $B$ to be two-by-two block-diagonal. In particular, several previously obtained partial results, including the cases of one normal matrix, unitary similarity between $B$ and $-A$ or $-A^T$, and anti-diagonal block structures, are reduced to this common block-diagonal structure. We also employ a manifold optimization method, which provides numerical evidence that the two-by-two block-diagonal structure is essential for saturating the inequality. Furthermore, we prove that the identified saturation points are critical points of the objective function on the constraint manifold.

quant-ph↗
arXiv2026

Near-Field Position and Orientation Tracking With Hybrid ELAA Architecture

This paper investigates near-field (NF) position and orientation tracking of a multi-antenna mobile station (MS) using an extremely large antenna array (ELAA)-equipped base station (BS) with a limited number of radio frequency (RF) chains. Under this hybrid array architecture, the received uplink pilot signal at the BS is first combined by analog phase shifters, producing a low-dimensional observation before digital processing. Such analog compression provides only partial access to the ELAA measurement, making it essential to design an analog combiner that can preserve pose-relevant signal components despite channel uncertainty and unit-modulus hardware constraints. To address this, we propose a predictive analog combining-assisted extended Kalman filter (PAC-EKF) framework, where the analog combiner can leverage the temporal correlation in the MS pose variation to capture the most informative signal components predictively. We then analyze fundamental performance limits via Bayesian Cramér-Rao bound and Fisher information matrix, explicitly quantifying how the analog combiner, array size, signal-to-noise ratio, and MS pose influence the pose information contained in the uplink observation. Building on these insights, we develop two methods for designing a low-complexity analog combiner. Numerical results show that the proposed predictive analog combining approach significantly improves tracking accuracy, even with fewer RF chains and lower transmit power.

eess.SP↗
arXiv2026

Near-Field Multi-User Communications via Polar-Domain Beamfocusing: Analytical Framework and Performance Analysis

As wireless systems evolve toward higher frequencies and extremely large antenna arrays, near-field (NF) propagation becomes increasingly dominant. Unlike far-field (FF) communication, which relies on a planar-wavefront model and is limited to angular-domain beamsteering, NF propagation exhibits spherical wavefronts that enable beamfocusing in both angle and distance, i.e., the polar domain, offering new opportunities for spatial multiple access. This paper develops an analytical stochastic geometry (SG) framework for a multi-user system assisted by polar-domain beamfocusing, which jointly captures NF propagation characteristics and the spatial randomness of user locations. The intrinsic coupling between angle and distance in the NF antenna pattern renders inter-user interference analysis intractable. To address this challenge, we propose a tractable near-field multi-level antenna pattern (NF-MLAP) approximation, which enables computationally efficient expressions and tight upper bounds for key performance metrics, including coverage probability, spectrum efficiency, and area spectrum efficiency. Analytical and simulation results demonstrate that the proposed framework accurately captures performance trends and reveals fundamental trade-offs between hardware configuration (including the number of antennas and radio frequency chains) and system performance (in terms of spatial resource reuse and interference mitigation).

eess.SP↗
arXiv2026

MigrationBench: Repository-Level Code Migration Benchmark from Java 8

With the rapid advancement of powerful large language models (LLMs) in recent years, a wide range of software engineering tasks can now be addressed using LLMs, significantly enhancing productivity and scalability. Numerous benchmark datasets have been developed to evaluate the coding capabilities of these models, while they primarily focus on code generation and issue-resolution tasks. In contrast, we introduce a new coding benchmark MigrationBench with a distinct focus: code migration. MigrationBench aims to serve as a comprehensive benchmark for migration from Java 8 to the latest long-term support (LTS) versions (Java 17, 21), including a full dataset and its subset selected with 5,102 and 300 repositories respectively. selected is a representative subset curated for complexity and difficulty, offering a versatile resource to support research in the field of code migration. Additionally, we provide a comprehensive evaluation framework to facilitate rigorous and standardized assessment of LLMs on this challenging task. We further propose an agentic framework and demonstrate that LLMs can effectively tackle repository-level code migration to Java 17. For the selected subset with Claude-4.5-Sonnet, our agentic framework achieves 71.67% and 53.33% success rate (pass@1) for minimal and maximal migration respectively. The dataset and evaluation source code are available at: https://huggingface.co/collections/AmazonScience/migrationbench and https://github.com/amazon-science/MigrationBench respectively.

cs.SE↗
arXiv2026

Toward Robust Semi-supervised Regression via Dual-stream Knowledge Distillation

Semi-supervised regression (SSR), which aims to predict continuous scores for samples while reducing the reliance on large-scale labeled data, has recently attracted considerable attention across various applications, including computer vision, natural language processing, audio analysis, and medical analysis. Existing SSR methods typically train models with scarce labeled data by introducing constraint-based regularization or ordinal ranking to mitigate overfitting. However, these approaches often fail to fully exploit the abundance of unlabeled samples. Although consistency-driven pseudo-labeling methods attempt to incorporate unlabeled data, their performance is highly sensitive to pseudo-label quality and noisy predictions. To address these challenges, we propose a Dual-stream Knowledge Distillation framework (DKD), which is specifically designed for SSR to distill both continuous-valued knowledge and distributional information. This design better preserves regression magnitude information and improves sample efficiency. Specifically, in DKD, the teacher is optimized solely with ground-truth labels for label distribution estimation, while the student learns from a mixture of real labels and teacher-generated pseudo targets on unlabeled data. The distillation process enables effective supervision transfer, allowing the student to leverage pseudo labels more robustly. Furthermore, we introduce a Decoupled Distribution Alignment (DDA) module, which separately aligns the target and non-target distributions between the teacher and student. To improve the reliability of non-target knowledge transfer, DDA incorporates a variance-guided non-target distribution alignment strategy that adaptively downweights uncertain teacher predictions, thereby enhancing the student's ability to mitigate noise in pseudo-label supervision and learn a better-calibrated regression predictor.

cs.LG↗
arXiv2026

Joint Training of Multi-Token Prediction in Reinforcement Learning via Optimal Coefficient Calibration

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as the standard paradigm for improving reasoning capability of large language models, while Multi-Token Prediction (MTP) has been a widely adopted module in pretraining. Combining them is a natural approach, yet current RL practices detach MTP gradients because joint training degrades the performance. We revisit this failure from an optimization perspective. We show that the per-step effect of MTP on the RL objective can be decomposed into two terms: a first-order correlation and a second-order perturbation penalty. This decomposition unifies three MTP training regimes: Detach, Cross-Entropy loss, and Policy loss, and explains why each succeeds or fails. Further analysis of policy loss reveals that, although it aligns with intuition, performance still degrades: the correlation term decays while the quadratic penalty persists. Guided by the analysis, we propose Optimal Coefficient Calibration (OCC), an adaptive scheme that tracks the optimal coefficient online via a log-probability proxy at negligible cost. Across six competition-level mathematical reasoning benchmarks, OCC consistently matches or exceeds the detach baseline, delivering improved joint MTP-RL training performance.

cs.LG↗