arXiv subjects

Ye Wang

Publications and source records attributed to Ye Wang.

At least 91 records · Page 5Linked to original sources

arXiv2026

High-Performance Quantum Frequency Conversion from Ultraviolet to Telecom Band

Quantum frequency conversion (QFC) is essential for bridging the spectral gap between stationary qubits and low-loss optical communication channels. In this work, we demonstrate a short-wavelength-pumping QFC with the first-order quasi-phase matching period of 3.07 um on thin-film lithium niobate, converting ultraviolet photons to the telecom C-band. By constructing a theoretical model that correlates the normalized conversion efficiency with domain defects in the short-period phase-matched waveguide, we found the critical tolerance of domain defects along the waveguide should be $\le 2$ (excluding the ends). Based on this, we achieved a theoretical limit normalized conversion efficiency of 839%/(W*cm^2) for the fundamental guided mode through fabrication optimization. Furthermore, we propose a robust noise suppression strategy for short-wavelength pumping by utilizing the counter-tuning behaviors of difference-frequency generation and spontaneous parametric down-conversion. By combining these advances with ultra-narrowband filtering, we achieve a record-high external efficiency of 28.8% and an ultra-low noise of 35 counts per second. This high-performance QFC connecting ultraviolet and telecom bands satisfies the stringent requirements for long-lived remote ion-ion entanglement in scalable quantum networks [W.-Z. Liu et al., Nature (2026)].

quant-ph↗

arXiv2026

ICPRL: Acquiring Physical Intuition from Interactive Control

VLMs excel at static perception but falter in interactive reasoning in dynamic physical environments, which demands planning and adaptation to dynamic outcomes. Existing physical reasoning methods often depend on abstract symbolic inputs or lack the ability to learn and adapt from direct, pixel-based visual interaction in novel scenarios. We introduce ICPRL (In-Context Physical Reinforcement Learning), a framework inspired by In-Context Reinforcement Learning (ICRL) that empowers VLMs to acquire physical intuition and adapt their policies in-context. Our approach trains a vision-grounded policy model via multi-turn Group Relative Policy Optimization (GRPO) over diverse multi-episode interaction histories. This enables the agent to adapt strategies by conditioning on past trial-and-error sequences, without requiring any weight updates. This adaptive policy works in concert with a separately trained world model that provides explicit physical reasoning by predicting the results of potential actions. At inference, the policy proposes candidate actions, while the world model predicts outcomes to guide a root-node PUCT search to select the most promising action. Evaluated on the diverse physics-based puzzle-solving tasks in the DeepPHY benchmark, ICPRL demonstrates significant improvements across both its I. policy-only, and II. world-model-augmented stages. Notably, these gains are retained in unseen physical environments, demonstrating that our framework facilitates genuine in-context acquisition of the environment's physical dynamics from interactive experience.

cs.LG↗

arXiv2026

Embedding Morphology into Transformers for Cross-Robot Policy Learning

Cross-robot policy learning -- training a single policy to perform well across multiple embodiments -- remains a central challenge in robot learning. Transformer-based policies, such as vision-language-action (VLA) models, are typically embodiment-agnostic and must infer kinematic structure purely from observations, which can reduce robustness across embodiments and even limit performance within a single embodiment. We propose an embodiment-aware transformer policy that injects morphology via three mechanisms: (1) kinematic tokens that factorize actions across joints and compress time through per-joint temporal chunking; (2) a topology-aware attention bias that encodes kinematic topology as an inductive bias in self-attention, encouraging message passing along kinematic edges; and (3) joint-attribute conditioning that augments topology with per-joint descriptors to capture semantics beyond connectivity. Across a range of embodiments, this structured integration consistently improves performance over a vanilla pi0.5 VLA baseline, indicating improved robustness both within an embodiment and across embodiments.

cs.RO↗

arXiv2026

Joint-Aligned Latent Action: Towards Scalable VLA Pretraining in the Wild

Despite progress, Vision-Language-Action models (VLAs) are limited by a scarcity of large-scale, diverse robot data. While human manipulation videos offer a rich alternative, existing methods are forced to choose between small, precisely-labeled datasets and vast in-the-wild footage with unreliable hand tracking labels. We present JALA, a pretraining framework that learns Jointly-Aligned Latent Actions. JALA bypasses full visual dynamic reconstruction, instead learns a predictive action embedding aligned with both inverse dynamics and real actions. This yields a transition-aware, behavior-centric latent space for learning from heterogeneous human data. We scale this approach with UniHand-Mix, a 7.5M video corpus (>2,000 hours) blending laboratory and in-the-wild footage. Experiments demonstrate that JALA generates more realistic hand motions in both controlled and unconstrained scenarios, significantly improving downstream robot manipulation performance in both simulation and real-world tasks. These results indicate that jointly-aligned latent actions offer a scalable pathway for VLA pretraining from human data.

cs.RO↗

arXiv2026

EasyMimic: A Low-Cost Framework for Robot Imitation Learning from Human Videos

Robot imitation learning is often hindered by the high cost of collecting large-scale, real-world data. This challenge is especially significant for low-cost robots designed for home use, as they must be both user-friendly and affordable. To address this, we propose the EasyMimic framework, a low-cost and replicable solution that enables robots to quickly learn manipulation policies from human video demonstrations captured with standard RGB cameras. Our method first extracts 3D hand trajectories from the videos. An action alignment module then maps these trajectories to the gripper control space of a low-cost robot. To bridge the human-to-robot domain gap, we introduce a simple and user-friendly hand visual augmentation strategy. We then use a co-training method, fine-tuning a model on both the processed human data and a small amount of robot data, enabling rapid adaptation to new tasks. Experiments on the low-cost LeRobot platform demonstrate that EasyMimic achieves high performance across various manipulation tasks. It significantly reduces the reliance on expensive robot data collection, offering a practical path for bringing intelligent robots into homes. Project website: https://zt375356.github.io/EasyMimic-Project/.

cs.RO↗

arXiv2026

Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization

While Vision-Language-Action (VLA) models show strong promise for generalist robot control, it remains unclear whether -- and under what conditions -- the standard "scale data" recipe translates to robotics, where training data is inherently heterogeneous across embodiments, sensors, and action spaces. We present a systematic, controlled study of VLA scaling that revisits core training choices for pretraining across diverse robots. Using a representative VLA framework that combines a vision-language backbone with flow-matching, we ablate key design decisions under matched conditions and evaluate in extensive simulation and real-robot experiments. To improve the reliability of real-world results, we introduce a Grouped Blind Ensemble protocol that blinds operators to model identity and separates policy execution from outcome judgment, reducing experimenter bias. Our analysis targets three dimensions of VLA scaling. (1) Physical alignment: we show that a unified end-effector (EEF)-relative action representation is critical for robust cross-embodiment transfer. (2) Embodiment mixture: we find that naively pooling heterogeneous robot datasets often induces negative transfer rather than gains, underscoring the fragility of indiscriminate data scaling. (3) Training regularization: we observe that intuitive strategies, such as sensory dropout and multi-stage fine-tuning, do not consistently improve performance at scale. Together, this study challenge some common assumptions about embodied scaling and provide practical guidance for training large-scale VLA policies from diverse robotic data. Project website: https://research.beingbeyond.com/rethink_vla

cs.RO↗

arXiv2026

Understanding Dynamic Compute Allocation in Recurrent Transformers

Token-level adaptive computation seeks to reduce inference cost by allocating more computation to harder tokens and less to easier ones. However, prior work is primarily evaluated on natural-language benchmarks using task-level metrics, where token-level difficulty is unobservable and confounded with architectural factors, making it unclear whether compute allocation truly aligns with underlying complexity. We address this gap through three contributions. First, we introduce a complexity-controlled evaluation paradigm using algorithmic and synthetic language tasks with parameterized difficulty, enabling direct testing of token-level compute allocation. Second, we propose ANIRA, a unified recurrent Transformer framework that supports per-token variable-depth computation while isolating compute allocation decisions from other model factors. Third, we use this framework to conduct a systematic analysis of token-level adaptive computation across alignment with complexity, generalization, and decision timing. Our results show that compute allocation aligned with task complexity can emerge without explicit difficulty supervision, but such alignment does not imply algorithmic generalization: models fail to extrapolate to unseen input sizes despite allocating additional computation. We further find that early compute decisions rely on static structural cues, whereas online halting more closely tracks algorithmic execution state.

cs.CL↗

arXiv2026

A building block of quantum repeaters for scalable quantum networks

Quantum networks, integrating quantum communication, quantum metrology, and distributed quantum computing, could provide secure and efficient information transfer, high-resolution sensing, and an exponential speed-up in information processing. Deterministic entanglement distribution over long distances is a prerequisite for scalable quantum networks, enabling the utilization of device-independent quantum key distribution (DI-QKD) and quantum teleportation to achieve secure and efficient information transfer. However, the exponential photon loss in optical fibres prohibits efficient and deterministic entanglement distribution. Quantum repeaters, incorporating entanglement swapping and entanglement purification with quantum memories, offer the most promising means to overcome this limitation in fibre-based quantum networks. Despite numerous pioneering efforts toward realizing quantum repeaters, a critical bottleneck remains, as remote memory-memory entanglement suffers from decoherence more rapidly than it can be established and purified over long distances. We overcome this by developing long-lived trapped-ion memories, an efficient telecom interface, and a high-visibility single-photon entanglement protocol. This allows us to establish and maintain memory-memory entanglement over a 10 km fibre within the average entanglement establishment time for the same distance. As a direct application, we demonstrate metropolitan-scale DI-QKD, distilling 1,917 secret keys out of 4.05*10^5 Bell pairs over 10 km. We further report a positive key rate over 101 km in the asymptotic limit, extending the achievable distance by more than two orders of magnitude. Our work provides a critical building block for quantum repeaters and marks an important step toward scalable quantum networks.

quant-ph↗

arXiv2026

Reinforcement Learning-based Home Energy Management with Heterogeneous Batteries and Stochastic EV Behaviour

The widespread adoption of photovoltaic (PV), electric vehicles (EVs), and stationary energy storage systems (ESS) in households increases system complexity while simultaneously offering new opportunities for energy regulation. However, effectively coordinating these resources under uncertainties remains challenging. This paper proposes a novel home energy management framework based on deep reinforcement learning (DRL) that can jointly minimise energy expenditure and battery degradation while guaranteeing occupant comfort and EV charging requirements. Distinct from existing studies, we explicitly account for the heterogeneous degradation characteristics of stationary and EV batteries in the optimisation, alongside stochastic user behaviour regarding arrival time, departure time, and driving distance. The energy scheduling problem is formulated as a constrained Markov decision process (CMDP) and solved using a Lagrangian soft actor-critic (SAC) algorithm. This approach enables the agent to learn optimal control policies that enforce physical constraints, including indoor temperature bounds and target EV state of charge upon departure, despite stochastic uncertainties. Numerical simulations over a one-year horizon demonstrate the effectiveness of the proposed framework in satisfying physical constraints while eliminating thermal oscillations and achieving significant economic benefits. Specifically, the method reduces the cumulative operating cost substantially compared to two standard rule-based baselines while simultaneously decreasing battery degradation costs by 8.44%.

eess.SY↗

arXiv2026

Composable Visual Tokenizers with Generator-Free Diagnostics of Learnability

We introduce CompTok, a training framework for learning visual tokenizers whose tokens are enhanced for compositionality. CompTok uses a token-conditioned diffusion decoder. By employing an InfoGAN-style objective, where we train a recognition model to predict the tokens used to condition the diffusion decoder using the decoded images, we enforce the decoder to not ignore any of the tokens. To promote compositional control, besides the original images, CompTok also trains on tokens formed by swapping token subsets between images, enabling more compositional control of the token over the decoder. As the swapped tokens between images do not have ground truth image targets, we apply a manifold constraint via an adversarial flow regularizer to keep unpaired swap generations on the natural-image distribution. The resulting tokenizer not only achieves state-of-the-art performance on image class-conditioned generation, but also demonstrates properties such as swapping tokens between images to achieve high level semantic editing of an image. Additionally, we propose two metrics that measures the landscape of the token space that can be useful to describe not only the compositionality of the tokens, but also how easy to learn the landscape is for a generator to be trained on this space. We show in experiments that CompTok can improve on both of the metrics as well as supporting state-of-the-art generators for class conditioned generation.

cs.CV↗

arXiv2026

Unlocking Large Audio-Language Models for Interactive Language Learning

Achieving pronunciation proficiency in a second language (L2) remains a challenge, despite the development of Computer-Assisted Pronunciation Training (CAPT) systems. Traditional CAPT systems often provide unintuitive feedback that lacks actionable guidance, limiting its effectiveness. Recent advancements in audio-language models (ALMs) offer the potential to enhance these systems by providing more user-friendly feedback. In this work, we investigate ALMs for chat-based pronunciation training by introducing L2-Arctic-plus, an English dataset with detailed error explanations and actionable suggestions for improvement. We benchmark cascaded ASR+LLMs and existing ALMs on this dataset, specifically in detecting mispronunciation and generating actionable feedback. To improve the performance, we further propose to instruction-tune ALMs on L2-Arctic-plus. Experimental results demonstrate that our instruction-tuned models significantly outperform existing baselines on mispronunciation detection and suggestion generation in terms of both objective and human evaluation, highlighting the value of the proposed dataset.

cs.SD↗

arXiv2026

Comparative Study of Large Language Models on Chinese Film Script Continuation: An Empirical Analysis Based on GPT-5.2 and Qwen-Max

As large language models (LLMs) are increasingly applied to creative writing, their performance on culturally specific narrative tasks warrants systematic investigation. This study constructs the first Chinese film script continuation benchmark comprising 53 classic films, and designs a multi-dimensional evaluation framework comparing GPT-5.2 and Qwen-Max-Latest. Using a "first half to second half" continuation paradigm with 3 samples per film, we obtained 303 valid samples (GPT-5.2: 157, 98.7% validity; Qwen-Max: 146, 91.8% validity). Evaluation integrates ROUGE-L, Structural Similarity, and LLM-as-Judge scoring (DeepSeek-Reasoner). Statistical analysis of 144 paired samples reveals: Qwen-Max achieves marginally higher ROUGE-L (0.2230 vs 0.2114, d=-0.43); however, GPT-5.2 significantly outperforms in structural preservation (0.93 vs 0.75, d=0.46), overall quality (44.79 vs 25.72, d=1.04), and composite scores (0.50 vs 0.39, d=0.84). The overall quality effect size reaches large effect level (d>0.8). GPT-5.2 excels in character consistency, tone-style matching, and format preservation, while Qwen-Max shows deficiencies in generation stability. This study provides a reproducible framework for LLM evaluation in Chinese creative writing.

cs.CL↗

arXiv2026

Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization

We introduce Being-H0.5, a foundational Vision-Language-Action (VLA) model designed for robust cross-embodiment generalization across diverse robotic platforms. While existing VLAs often struggle with morphological heterogeneity and data scarcity, we propose a human-centric learning paradigm that treats human interaction traces as a universal "mother tongue" for physical interaction. To support this, we present UniHand-2.0, the largest embodied pre-training recipe to date, comprising over 35,000 hours of multimodal data across 30 distinct robotic embodiments. Our approach introduces a Unified Action Space that maps heterogeneous robot controls into semantically aligned slots, enabling low-resource robots to bootstrap skills from human data and high-resource platforms. Built upon this human-centric foundation, we design a unified sequential modeling and multi-task pre-training paradigm to bridge human demonstrations and robotic execution. Architecturally, Being-H0.5 utilizes a Mixture-of-Transformers design featuring a novel Mixture-of-Flow (MoF) framework to decouple shared motor primitives from specialized embodiment-specific experts. Finally, to make cross-embodiment policies stable in the real world, we introduce Manifold-Preserving Gating for robustness under sensory shift and Universal Async Chunking to universalize chunked control across embodiments with different latency and control profiles. We empirically demonstrate that Being-H0.5 achieves state-of-the-art results on simulated benchmarks, such as LIBERO (98.9%) and RoboCasa (53.9%), while also exhibiting strong cross-embodiment capabilities on five robotic platforms.

cs.RO↗

arXiv2026

Role-Playing Agents Driven by Large Language Models: Current Status, Challenges, and Future Trends

In recent years, with the rapid advancement of large language models (LLMs), role-playing language agents (RPLAs) have emerged as a prominent research focus at the intersection of natural language processing (NLP) and human-computer interaction. This paper systematically reviews the current development and key technologies of RPLAs, delineating the technological evolution from early rule-based template paradigms, through the language style imitation stage, to the cognitive simulation stage centered on personality modeling and memory mechanisms. It summarizes the critical technical pathways supporting high-quality role-playing, including psychological scale-driven character modeling, memory-augmented prompting mechanisms, and motivation-situation-based behavioral decision control. At the data level, the paper further analyzes the methods and challenges of constructing role-specific corpora, focusing on data sources, copyright constraints, and structured annotation processes. In terms of evaluation, it collates multi-dimensional assessment frameworks and benchmark datasets covering role knowledge, personality fidelity, value alignment, and interactive hallucination, while commenting on the advantages and disadvantages of methods such as human evaluation, reward models, and LLM-based scoring. Finally, the paper outlines future development directions of role-playing agents, including personality evolution modeling, multi-agent collaborative narrative, multimodal immersive interaction, and integration with cognitive neuroscience, aiming to provide a systematic perspective and methodological insights for subsequent research.

cs.CL↗

arXiv2025

PhraseVAE and PhraseLDM: Latent Diffusion for Full-Song Multitrack Symbolic Music Generation

This technical report presents a new paradigm for full-song symbolic music generation. Existing symbolic models operate on note-attribute tokens and suffer from extremely long sequences, limited context length, and weak support for long-range structure. We address these issues by introducing PhraseVAE and PhraseLDM, the first latent diffusion framework designed for full-song multitrack symbolic music. PhraseVAE compresses an arbitrary variable-length polyphonic note sequence into a single compact 64-dimensional phrase-level latent representation with high reconstruction fidelity, allowing a well-structured latent space and efficient generative modeling. Built on this latent space, PhraseLDM generates an entire multi-track song in a single pass without any autoregressive components. The system eliminates bar-wise sequential modeling, supports up to 128 bars of music (8 minutes at 64 bpm), and produces complete songs with coherent local texture, idiomatic instrument patterns, and clear global structure. With only 45M parameters, our framework generates a full song within seconds while maintaining competitive musical quality and generation diversity. Together, these results show that phrase-level latent diffusion provides an effective and scalable solution to long-sequence modeling in symbolic music generation. We hope this work encourages future symbolic music research to move beyond note-attribute tokens and to consider phrase-level units as a more effective and musically meaningful modeling target.

cs.SD↗

arXiv2025

Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos

Vision-Language-Action (VLA) models provide a promising paradigm for robot learning by integrating visual perception with language-guided policy learning. However, most existing approaches rely on 2D visual inputs to perform actions in 3D physical environments, creating a significant gap between perception and action grounding. To bridge this gap, we propose a Spatial-Aware VLA Pretraining paradigm that performs explicit alignment between visual space and physical space during pretraining, enabling models to acquire 3D spatial understanding before robot policy learning. Starting from pretrained vision-language models, we leverage large-scale human demonstration videos to extract 3D visual and 3D action annotations, forming a new source of supervision that aligns 2D visual observations with 3D spatial reasoning. We instantiate this paradigm with VIPA-VLA, a dual-encoder architecture that incorporates a 3D visual encoder to augment semantic visual representations with 3D-aware features. When adapted to downstream robot tasks, VIPA-VLA achieves significantly improved grounding between 2D vision and 3D action, resulting in more robust and generalizable robotic policies.

cs.RO↗

arXiv2025

Joint Learning of Wording and Formatting for Singable Melody-to-Lyric Generation

Despite progress in melody-to-lyric generation, a substantial singability gap remains between machine-generated lyrics and those written by human lyricists. In this work, we aim to narrow this gap by jointly learning both wording and formatting for melody-to-lyric generation. After general-domain pretraining, our model acquires length awareness through an self-supervised stage trained on a large text-only lyric corpus. During supervised melody-to-lyric training, we introduce multiple auxiliary supervision objective informed by musicological findings on melody--lyric relationships, encouraging the model to capture fine-grained prosodic and structural patterns. Compared with naïve fine-tuning, our approach improves adherence to line-count and syllable-count requirements by 3.8% and 21.4% absolute, respectively, without degrading text quality. In human evaluation, it achieves 42.2% and 74.2% relative gains in overall quality over two task-specific baselines, underscoring the importance of formatting-aware training for generating singable lyrics.

cs.CL↗

arXiv2025

Seedream 4.0: Toward Next-generation Multimodal Image Generation

We introduce Seedream 4.0, an efficient and high-performance multimodal image generation system that unifies text-to-image (T2I) synthesis, image editing, and multi-image composition within a single framework. We develop a highly efficient diffusion transformer with a powerful VAE which also can reduce the number of image tokens considerably. This allows for efficient training of our model, and enables it to fast generate native high-resolution images (e.g., 1K-4K). Seedream 4.0 is pretrained on billions of text-image pairs spanning diverse taxonomies and knowledge-centric concepts. Comprehensive data collection across hundreds of vertical scenarios, coupled with optimized strategies, ensures stable and large-scale training, with strong generalization. By incorporating a carefully fine-tuned VLM model, we perform multi-modal post-training for training both T2I and image editing tasks jointly. For inference acceleration, we integrate adversarial distillation, distribution matching, and quantization, as well as speculative decoding. It achieves an inference time of up to 1.8 seconds for generating a 2K image (without a LLM/VLM as PE model). Comprehensive evaluations reveal that Seedream 4.0 can achieve state-of-the-art results on both T2I and multimodal image editing. In particular, it demonstrates exceptional multimodal capabilities in complex tasks, including precise image editing and in-context reasoning, and also allows for multi-image reference, and can generate multiple output images. This extends traditional T2I systems into an more interactive and multidimensional creative tool, pushing the boundary of generative AI for both creativity and professional applications. We further scale our model and data as Seedream 4.5. Seedream 4.0 and Seedream 4.5 are accessible on Volcano Engine https://www.volcengine.com/experience/ark?launch=seedream.

cs.CV↗