arXiv subjects

Yixuan Wang

Publications and source records attributed to Yixuan Wang.

At least 73 records · Page 4Linked to original sources

arXiv2026

RA-SSU: Towards Fine-Grained Audio-Visual Learning with Region-Aware Sound Source Understanding

Audio-Visual Learning (AVL) is one fundamental task of multi-modality learning and embodied intelligence, displaying the vital role in scene understanding and interaction. However, previous researchers mostly focus on exploring downstream tasks from a coarse-grained perspective (e.g., audio-visual correspondence, sound source localization, and audio-visual event localization). Considering providing more specific scene perception details, we newly define a fine-grained Audio-Visual Learning task, termed Region-Aware Sound Source Understanding (RA-SSU), which aims to achieve region-aware, frame-level, and high-quality sound source understanding. To support this goal, we innovatively construct two corresponding datasets, i.e. fine-grained Music (f-Music) and fine-grained Lifescene (f-Lifescene), each containing annotated sound source masks and frame-by-frame textual descriptions. The f-Music dataset includes 3,976 samples across 22 scene types related to specific application scenarios, focusing on music scenes with complex instrument mixing. The f-Lifescene dataset contains 6,156 samples across 61 types representing diverse sounding objects in life scenarios. Moreover, we propose SSUFormer, a Sound-Source Understanding TransFormer benchmark that facilitates both the sound source segmentation and sound region description with a multi-modal input and multi-modal output architecture. Specifically, we design two modules for this framework, Mask Collaboration Module (MCM) and Mixture of Hierarchical-prompted Experts (MoHE), to respectively enhance the accuracy and enrich the elaboration of the sound source description. Extensive experiments are conducted on our two datasets to verify the feasibility of the task, evaluate the availability of the datasets, and demonstrate the superiority of the SSUFormer, which achieves SOTA performance on the Sound Source Understanding benchmark.

cs.CV↗

arXiv2026

CuriousBot: Interactive Mobile Exploration via Actionable 3D Relational Object Graph

Mobile exploration is a longstanding challenge in robotics, yet current methods primarily focus on active perception instead of active interaction, limiting the robot's ability to interact with and fully explore its environment. Existing robotic exploration approaches via active interaction are often restricted to tabletop scenes, neglecting the unique challenges posed by mobile exploration, such as large exploration spaces, complex action spaces, and diverse object relations. In this work, we introduce a 3D relational object graph that encodes diverse object relations and enables exploration through active interaction. We develop a system based on this representation and evaluate it across diverse scenes. Our qualitative and quantitative results demonstrate the system's effectiveness and generalization across object instances, relations, and scenes, outperforming methods solely relying on vision-language models (VLMs).

cs.RO↗

arXiv2026

Interactive World Simulator for Robot Policy Training and Evaluation

Action-conditioned video prediction models (often referred to as world models) have shown strong potential for robotics applications, but existing approaches are often slow and struggle to capture physically consistent interactions over long horizons, limiting their usefulness for scalable robot policy training and evaluation. We present Interactive World Simulator, a framework for building interactive world models from a moderate-sized robot interaction dataset. Our approach leverages consistency models for both image decoding and latent-space dynamics prediction, enabling fast and stable simulation of physical interactions. In our experiments, the learned world models produce interaction-consistent pixel-level predictions and support stable long-horizon interactions for more than 10 minutes at 15 FPS on a single RTX 4090 GPU. Our framework enables scalable demonstration collection solely within the world models to train state-of-the-art imitation policies. Through extensive real-world evaluation across diverse tasks involving rigid objects, deformable objects, object piles, and their interactions, we find that policies trained on world-model-generated data perform comparably to those trained on the same amount of real-world data. Additionally, we evaluate policies both within the world models and in the real world across diverse tasks, and observe a strong correlation between simulated and real-world performance. Together, these results establish the Interactive World Simulator as a stable and physically consistent surrogate for scalable robotic data generation and faithful, reproducible policy evaluation.

cs.RO↗

arXiv2026

PonderLM-2: Pretraining LLM with Latent Thoughts in Continuous Space

The remarkable success of Chain-of-Thought (CoT), which enhances performance by scaling generation steps at test-time, inspires us to ask: can we leverage a similar scaling of computational steps during pretraining to improve the generation of each individual token? To address this, we propose a novel pre-training methodology: Pretraining Language Models with Latent Thoughts (PonderLM-2). Our approach pretrains a language model (LM) to first generate an intermediate latent thought-the last hidden state of the current position-which is then used as input to predict the actual subsequent token. This additional computational step enables the LM to refine its prediction within unconstrained continuous space. Our experiments demonstrate that, at an identical inference cost, a LM that generates one additional latent thought per token outperforms a standard model with double the parameters. For instance, our PonderLM-2-Pythia-1.4B, pretrained on 300B tokens from the Pile, significantly surpasses the vanilla Pythia-2.8B trained on the same data on both language modeling and a range of general downstream tasks. Furthermore, increasing the number of latent thoughts generated before each actual token-forming a chain analogous to CoT-consistently improves the model's performance. The code is available at https://github.com/LUMIA-Group/PonderLM-2.

cs.CL↗

arXiv2026

Nonlinear Nanophotonic Chip-space Interfaces: On-chip Generation of Structured, Topological and Spatiotemporal Lights Via Nonlinear Čerenkov Radiation

Miniaturized and reconfigurable interfaces between confined optical modes within integrated photonic chips and structured light propagating in free space would serve as a cornerstone for fundamental optical science and modern photonic technology. In this work, we exploit the anisotropic nonlinear susceptibility tensors associated with thin-film lithium niobate to construct nanophotonic chip-space interfaces capable of flexibly generating and multi-dimensionally engineering structured light via injections of photons to on-chip waveguides. By harnessing the nonlinear Čerenkov radiation in integrated nonlinear microring resonators, we successfully tailor the spatial profile, polarization state, emission wavelength, topological charge and temporal wave packet of structured optical vortices, exhibiting reconfigurabilities and tuning ranges far beyond the state-of-the-art. To further showcase the capabilities of our platform, we use a single pump to generate tunable optical skyrmions via the spin-orbit coupling and multi-state integrated vortex microcombs in the visible range via synergistic $χ^{(2)}$ and $χ^{(3)}$ nonlinear optical processes. Our work bridges the research fields of structured light and integrated nonlinear optics, providing unprecedented opportunities for spatiotemporal light generation and on-chip multidimensional nonlinear optics.

physics.optics↗

arXiv2026

KANO: Kolmogorov-Arnold Neural Operator

We introduce Kolmogorov--Arnold Neural Operator (KANO), a dual-domain neural operator jointly parameterized by both spectral and spatial bases with intrinsic symbolic interpretability. We theoretically demonstrate that KANO overcomes the pure-spectral bottleneck of Fourier Neural Operator (FNO): KANO remains expressive over generic position-dependent dynamics (variable coefficient PDEs) for any physical input, whereas FNO stays practical only for spectrally sparse operators and strictly imposes a fast-decaying input Fourier tail. We verify our claims empirically on position-dependent differential operators, for which KANO robustly generalizes but FNO fails to. In the quantum Hamiltonian learning benchmark, KANO reconstructs ground-truth Hamiltonians in closed-form symbolic representations accurate to the fourth decimal place in coefficients and attains $\approx 6\times10^{-6}$ state infidelity from projective measurement data, substantially outperforming that of the FNO trained with ideal full wave function data, $\approx 1.5\times10^{-2}$, by orders of magnitude.

cs.LG↗

arXiv2026

PonderLM: Pretraining Language Models to Ponder in Continuous Space

Humans ponder before articulating complex sentence elements, enabling deeper cognitive processing through focused effort. In this work, we introduce this pondering process into language models by repeatedly invoking the forward process within a single token generation step. During pondering, instead of generating an actual token sampled from the prediction distribution, the model ponders by yielding a weighted sum of all token embeddings according to the predicted token distribution. The generated embedding is then fed back as input for another forward pass. We show that the model can learn to ponder in this way through self-supervised learning, without any human annotations. Experiments across three widely used open-source architectures-GPT-2, Pythia, and LLaMA-and extensive downstream task evaluations demonstrate the effectiveness and generality of our method. On 9 downstream benchmarks, our pondering-enhanced Pythia models significantly outperform the official Pythia models. Notably, our PonderPythia models demonstrate remarkable effectiveness: PonderPythia-2.8B surpasses Pythia-6.9B and rivals Pythia-12B, while our PonderPythia-1B matches TinyLlama-1.1B, a model trained on 10 times more data. The code is available at https://github.com/LUMIA-Group/PonderingLM.

cs.CL↗

arXiv2026

Enhancing Inverse Reinforcement Learning through Encoding Dynamic Information in Reward Shaping

In this paper, we aim to tackle the limitation of the Adversarial Inverse Reinforcement Learning (AIRL) method in stochastic environments where theoretical results cannot hold and performance is degraded. To address this issue, we propose a novel method which infuses the dynamics information into the reward shaping with the theoretical guarantee for the induced optimal policy in the stochastic environments. Incorporating our novel model-enhanced rewards, we present a novel Model-Enhanced AIRL framework, which integrates transition model estimation directly into reward shaping. Furthermore, we provide a comprehensive theoretical analysis of the reward error bound and performance difference bound for our method. The experimental results in MuJoCo benchmarks show that our method can achieve superior performance in stochastic environments and competitive performance in deterministic environments, with significant improvement in sample efficiency, compared to existing baselines.

cs.LG↗

arXiv2026

Next Concept Prediction in Discrete Latent Space Leads to Stronger Language Models

We propose Next Concept Prediction (NCP), a generative pretraining paradigm built on top of Next Token Prediction (NTP). NCP predicts discrete concepts that span multiple tokens, thereby forming a more challenging pretraining objective. Our model, ConceptLM, quantizes hidden states using Vector Quantization and constructs a concept vocabulary. It leverages both NCP and NTP to drive parameter updates and generates a concept to guide the generation of the following tokens. We train ConceptLM from scratch at scales ranging from 70M to 1.5B parameters with up to 300B training data, including Pythia and GPT-2 backbones. Results on 13 benchmarks show that NCP yields consistent performance gains over traditional token-level models. Furthermore, continual pretraining experiments on an 8B-parameter Llama model indicate that NCP can further improve an NTP-trained model. Our analysis suggests that NCP leads to more powerful language models by introducing a harder pretraining task, providing a promising path toward better language modeling.

cs.CL↗

arXiv2026

FreqKV: Key-Value Compression in Frequency Domain for Context Window Extension

Existing key-value (KV) cache compression methods for large language models (LLMs) often rely on token eviction, which risks losing critical local information in both long prefilling and decoding scenarios. When extrapolating beyond the pretrained context length, their performance degrades sharply on long-context benchmarks. Motivated by the observation in the frequency domain that the context information is concentrated in the low-frequency components, we propose FreqKV, a parameter-free and architecture-agnostic approach. It iteratively compresses the increasing KV cache in the frequency domain, allowing models to process lengthy contexts efficiently. With minimal training at 8K length, FreqKV extends the context window of LLaMA-2-7B up to 256K tokens while maintaining stable perplexity. Extensive experiments across prefilling and decoding demonstrate that FreqKV enables robust context window extension and consistently outperforms existing KV cache compression methods on LLaMA-2 and LLaMA-3, highlighting its effectiveness for both understanding and generation in long contexts.

cs.CL↗

arXiv2026

Riemannian Lyapunov Optimizer: A Unified Framework for Optimization

We introduce Riemannian Lyapunov Optimizers (RLOs), a family of optimization algorithms that unifies classic optimizers within one geometric framework. Unlike heuristic improvements to existing optimizers, RLOs are systematically derived from a novel control-theoretic framework that reinterprets optimization as an extended state discrete-time controlled dynamical system on a Riemannian parameter manifold. Central to this framework is the identification of a Normally Attracting Invariant Manifold (NAIM), which organizes training dynamics into two distinct stages: rapid alignment of the speed state to a target graph, followed by controlled evolution within it. We formalize this by constructing a strict Lyapunov function that certifies convergence to a target manifold. This perspective yields a constructive ``optimizer generator" that not only recovers classic algorithms but enables the principled design of RLOs. We validate our theory via geometric diagnostics and demonstrate that grounding optimizer design in control theory yields state-of-the-art performance in large-scale benchmarks. Overall, RLOs bridge control theory and modern machine learning optimization, providing a unified language and a systematic toolkit for designing stable, effective optimizers.

cs.LG↗

arXiv2026

FC-PINO: High Precision Physics-Informed Neural Operators via Fourier Continuation

The physics-informed neural operator (PINO) is a machine learning paradigm that has demonstrated promising results for learning solutions to partial differential equations (PDEs). It leverages the Fourier Neural Operator to learn solution operators in function spaces and leverages physics losses during training to penalize deviations from known physics laws. Spectral differentiation provides an efficient way to compute derivatives for the physics losses, but it inherently assumes periodicity. When applied to non-periodic functions, this assumption can lead to significant errors, including Gibbs phenomena near domain boundaries which degrade the accuracy of both function representations and derivative computations. To overcome this limitation, we introduce the FC-PINO (Fourier-Continuation-based Physics-Informed Neural Operator) architecture which extends the accuracy and efficiency of PINO and spectral differentiation to non-periodic and non-smooth PDEs. In FC-PINO, we propose integrating Fourier continuation into the PINO framework, and test two different continuation approaches: FC-Legendre and FC-Gram. By transforming non-periodic signals into periodic functions on extended domains in a well-conditioned manner, Fourier continuation enables fast and accurate derivative computations. This approach avoids the discretization sensitivity of finite differences and the memory overhead of automatic differentiation. We demonstrate that standard PINO fails (without padding) or struggles (even with padding) to solve non-periodic and non-smooth PDEs with high precision, across challenging benchmarks. In contrast, the proposed FC-PINO provides accurate, robust, and scalable solutions, substantially outperforming PINO alternatives, and demonstrating that Fourier continuation is critical for extending PINO to a wider range of PDE problems when high-precision solutions are needed.

cs.LG↗

arXiv2026

Latent-Space Contrastive Reinforcement Learning for Stable and Efficient LLM Reasoning

While Large Language Models (LLMs) demonstrate exceptional performance in surface-level text generation, their nature in handling complex multi-step reasoning tasks often remains one of ``statistical fitting'' rather than systematic logical deduction. Traditional Reinforcement Learning (RL) attempts to mitigate this by introducing a ``think-before-speak'' paradigm. However, applying RL directly in high-dimensional, discrete token spaces faces three inherent challenges: sample-inefficient rollouts, high gradient estimation variance, and the risk of catastrophic forgetting. To fundamentally address these structural bottlenecks, we propose \textbf{DeepLatent Reasoning (DLR)}, a latent-space bidirectional contrastive reinforcement learning framework. This framework shifts the trial-and-error cost from expensive token-level full sequence generation to the continuous latent manifold. Specifically, we introduce a lightweight assistant model to efficiently sample $K$ reasoning chain encodings within the latent space. These encodings are filtered via a dual reward mechanism based on correctness and formatting; only high-value latent trajectories are fed into a \textbf{frozen main model} for single-pass decoding. To maximize reasoning diversity while maintaining coherence, we design a contrastive learning objective to enable directed exploration within the latent space. Since the main model parameters remain frozen during optimization, this method mathematically eliminates catastrophic forgetting. Experiments demonstrate that under comparable GPU computational budgets, DLR achieves more stable training convergence, supports longer-horizon reasoning chains, and facilitates the sustainable accumulation of reasoning capabilities, providing a viable path toward reliable and scalable reinforcement learning for LLMs.

cs.LG↗

arXiv2026

Seer Self-Consistency: Advance Budget Estimation for Adaptive Test-Time Scaling

Test-time scaling improves the inference performance of Large Language Models (LLMs) but also incurs substantial computational costs. Although recent studies have reduced token consumption through dynamic self-consistency, they remain constrained by the high latency of sequential requests. In this paper, we propose SeerSC, a dynamic self-consistency framework that simultaneously improves token efficiency and latency by integrating System 1 and System 2 reasoning. Specifically, we utilize the rapid System 1 to compute the answer entropy for given queries. This score is then used to evaluate the potential of samples for scaling, enabling dynamic self-consistency under System 2. Benefiting from the advance and accurate estimation provided by System 1, the proposed method can reduce token usage while simultaneously achieving a significant decrease in latency through parallel generation. It outperforms existing methods, achieving up to a 47% reduction in token consumption and a 43% reduction in inference latency without significant performance loss.

cs.CL↗

arXiv2026

Judge Q: Trainable Queries for Optimized Information Retention in KV Cache Eviction

Large language models (LLMs) utilize key-value (KV) cache to store historical information during sequence processing. The size of KV cache grows linearly as the length of the sequence extends, which seriously affects memory usage and decoding efficiency. Current methods for KV cache eviction typically utilize the last window from the pre-filling phase as queries to compute the KV importance scores for eviction. Although this scheme is simple to implement, it tends to overly focus on local information, potentially leading to the neglect or omission of crucial global information. To mitigate this issue, we propose Judge Q, a novel training method which incorporates a soft token list. This method only tunes the model's embedding layer at a low training cost. By concatenating the soft token list at the end of the input sequence, we train these tokens' attention map to the original input sequence to align with that of the actual decoded tokens. In this way, the queries corresponding to the soft tokens can effectively capture global information and better evaluate the importance of the keys and values within the KV cache, thus maintaining decoding quality when KV cache is evicted. Under the same eviction budget, our method exhibits less performance degradation compared to existing eviction approaches. We validate our approach through experiments conducted on models such as Llama-3.1-8B-Instruct and Mistral-7B-Instruct-v0.3, using benchmarks including LongBench, RULER, and Needle-in-a-Haystack. Results indicate an improvement of approximately 1 point on the LongBench and over 3 points on RULER. This proposed methodology can be seamlessly integrated into existing open-source models with minimal training overhead, thereby enhancing performance in KV cache eviction scenarios.

cs.CL↗

arXiv2026

$L^2$-based stability of blowup with log correction for semilinear heat equation

We propose an alternative proof of the classical result of Type-I blowup with log correction for the semilinear heat equation. Compared with previous proofs, we use a novel idea of enforcing stable normalizations for perturbations around the approximate profile and we establish a weighted $H^k$ stability, thereby avoiding the use of a topological argument and the analysis of a linearized spectrum. Consequently, this approach can be adopted even if we only have a numerical profile and do not have explicit information on the spectrum of its linearized operator. This result generalizes the $L^2$-based stability framework beyond exactly self-similar blowup and can be adapted to higher dimensions. Numerical results corroborate the effectiveness of our normalization, even in the large perturbation regime beyond our theoretical setting.

math.AP↗

arXiv2025

Quantum Noise Spectroscopy of Nanoscale Charge Defects in Silicon Carbide at Room Temperature

The nanoscale charge environment critically influences semiconductor physics and device performance. While conventional bulk characterization techniques provide volume-averaged defect properties, they lack the spatial resolution to resolve nanoscale charge heterogeneity and identify microscopic noise sources. Here, we utilize single PL5 centers in 4H-SiC as room-temperature broadband quantum sensors to fill in the gap. We report the first real-time, nanoscale observation of singlecharge tunneling dynamics in a commercial semiconductor at room temperature, by monitoring the random telegraph noise using optically detected magnetic resonance (ODMR). This capability enables an electrical noise imaging technique, showing distinct noise variations across different wafer substrates. By employing dynamical decoupling, we extend noise spectroscopy from near-DC to MHz frequencies, uncovering significant noise spectral density correlations across frequency bands. Finally, we probe MHz-GHz noise and identify its origin via T1 relaxation spectroscopy, obtaining the first nanoscale electron paramagnetic resonance (EPR) spectroscopic fingerprint of charge defects in SiC. These techniques open avenues for characterizing noise environments in semiconductor devices, providing critical insights for optimizing SiC fabrication processes, defect control, and advancing quantum technologies.

quant-ph↗

arXiv2025

Lyapunov-Based Kolmogorov-Arnold Network (KAN) Adaptive Control

Recent advancements in Lyapunov-based Deep Neural Networks (Lb-DNNs) have demonstrated improved performance over shallow NNs and traditional adaptive control for nonlinear systems with uncertain dynamics. Existing Lb-DNNs rely on multi-layer perceptrons (MLPs), which lack interpretable insights. As a first step towards embedding interpretable insights in the control architecture, this paper develops the first Lyapunov-based Kolmogorov-Arnold Networks (Lb-KAN) adaptive control method for uncertain nonlinear systems. Unlike MLPs with deep-layer matrix multiplications, KANs provide interpretable insights by direct functional decomposition. In this framework, KANs are employed to approximate uncertain dynamics and embedded into the control law, enabling visualizable functional decomposition. The analytical update laws are constructed from a Lyapunov-based analysis for real-time learning without prior data in a KAN architecture. The analysis uses the distinct KAN approximation theorem to formally bound the reconstruction error and its effect on the performance. The update law is derived by incorporating the KAN's learnable parameters into a Jacobian matrix, enabling stable, analytical, real-time adaptation and ensuring asymptotic convergence of tracking errors. Moreover, the Lb-KAN provides a foundation for interpretability characteristics by achieving visualizable functional decomposition. Simulation results demonstrate that the Lb-KAN controller reduces the function approximation error by 20.2% and 18.0% over the baseline Lb-LSTM and Lb-DNN methods, respectively.

eess.SY↗