arXiv subjects

Junfeng Li

Publications and source records attributed to Junfeng Li.

At least 19 recordsLinked to original sources

arXiv2026

Entirely nonlocal quantum magic without entanglement

Nonstabilizerness, or magic, is an archetypal \emph{quantum} resource that is necessary for quantum computational advantage. Here we uncover a phenomenon seemingly at odds with the quantum nature of magic: entirely nonlocal magic (ENM)---magic present only in correlations and absent from each party's marginal---can live without entanglement. We systematically study this separation and show it is universal and operationally reversible: every magical state or channel can be encoded into and recovered from a separable ENM realization using only local stabilizer processing and classical communication. We leverage this mechanism to devise an activation key protocol in which a classical key controls access to non-Clifford operations. We further formulate magic secret sharing, in which computational power inaccessible to any party alone becomes accessible through cooperation. On a superconducting quantum processor, we experimentally demonstrate activation key and network computing primitives, together with separable ENM state preparation and extraction protocols. Together, our results establish that magic can be classically activated, localized, and secret-shared without entanglement, providing new resource-control primitives for distributed quantum computation.

quant-ph↗

arXiv2026

SG-WAM: Text-Grounded and Spatial-aware Semantic Guidance for World-Action Models

World-Action Models (WAMs) have emerged as a promising paradigm for robotic manipulation. However, most existing WAMs generate future videos and actions by relying mainly on visual cues rather than language instructions, since off-the-shelf text encoders embed instructions independently of visual observations. As a result, the videos predicted by these WAMs are often semantically misaligned with their corresponding language instructions, which degrades the accuracy of the predicted actions. To overcome this limitation, we propose SG-WAM, a semantic guidance method for world-action models that leverages a vision-language model (VLM) as a semantic planner to enhance the instruction-grounding capacity of world-action models. Specifically, we train a VLM-based planner to predict text-grounded and spatial-aware semantic foresight. The text-grounded semantic foresight grounds the instruction by identifying the correct target objects, and the spatial-aware semantic foresight provides the scene geometry for precise manipulation. We then inject this foresight into the world-action model as high-level semantic guidance, ensuring that both future-video generation and action prediction faithfully follow the language instruction. Extensive experiments in simulation and the real world demonstrate the superiority of our semantic guidance method, showcasing precise manipulation and strong instruction-following capabilities.

cs.RO↗

arXiv2026

Is Forward Prediction Enough? Physical State Grounding for JEPA World Models

Learning structured and control-relevant latent representations remains a key challenge for world models. Recent JEPA-based world models learn action-conditioned predictive latent dynamics from observation sequences. However, their forward-prediction objectives do not explicitly enforce reliable identifiability of robot-centric physical state from individual latents or state changes from latent pairs, which can limit downstream planning and policy performance. We propose PSG-JEPA, a physically grounded JEPA world model that shapes its latent space with two complementary grounding objectives beyond forward prediction: grounding individual latents in robot proprioceptive state, and grounding latent pairs in multi-horizon joint-angle changes. Both objectives are applied only during training, leaving the inference architecture and computational cost unchanged. To comprehensively evaluate PSG-JEPA, we conduct experiments at three levels: (1) latent identifiability via probing, (2) goal-conditioned planning on frozen latents, and (3) policy learning in simulation and on a real robot. Experiments demonstrate that our PSG-JEPA consistently outperforms state-of-the-art latent world-model baselines at all three levels.

cs.RO↗

arXiv2026

Robust-WAM: Bridging Generative Pretraining and Semantic Foresight in World-Action Models

Mainstream World-Action Models (WAMs) adapt pretrained video generation models (VGMs) for robot control, transferring their learned dynamics prior for action prediction. These VGMs are typically trained in a variational autoencoder (VAE) latent space. However, the VAE latent space is optimized for pixel reconstruction, which rewards fine appearance detail and leaves the action prediction fragile under visual shifts. Recent works build WAMs in semantic latent space, which are more robust to appearance shifts. However, these models cannot leverage the large-scale VGM pretraining that exists only in VAE space. To overcome this dilemma, we propose Robust-WAM, a general post-training method for video-generation-based WAMs that preserves the VAE-based generative path and adds a lightweight semantic foresight alignment objective on the action stream. This retains the large-scale VGM pretraining while grounding actions in appearance-invariant dynamics that stay reliable under illumination shifts and other visual out-of-distribution conditions. Specifically, we employ learnable query tokens to bring future-scene semantics into the action stream by aligning their output hidden states with the semantic foresight of future ground-truth frames. To establish the temporal correspondence between each query and the future step it describes, we give it the positional encoding of the matching action tokens. Experiments on out-of-distribution generalization simulation benchmarks and a real-robot setup show that our Robust-WAM consistently improves the success rates of multiple WAM baselines without sacrificing in-distribution performance.

cs.CV↗

arXiv2026

DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for Cross-Embodiment Manipulation

Vision-Language-Action (VLA) models have become a powerful paradigm for robot manipulation, but training a single generalist policy for heterogeneous robot embodiments remains an open problem. Existing methods have two main limitations. First, they underuse dynamics priors shared across diverse visual and interaction data, limiting cross-embodiment transfer. Second, they require extensive manual preprocessing to convert embodiment-specific actions into a common format. To overcome these limitations, we propose DyPES-VLA, a cross-embodiment VLA that learns shared Dynamics Priors and Embodiment-Specific control. First, we learn shared dynamics priors by training the vision-language model (VLM) with a future-prediction objective on cross-embodiment data, driving the shared query representation to capture object motion, contact, and interaction-induced scene changes. Second, an embodiment-specific Mixture-of-Experts (MoE) action head translates these shared dynamics priors into executable controls directly in each embodiment's native action space, without manually pre-aligning heterogeneous actions into a common format. This head shares attention layers to capture common temporal action structures, while its embodiment-specific feed-forward experts resolve the unique kinematic constraints and control semantics of distinct embodiments. As a generalist policy, our \ourmethod achieves state-of-the-art performance across simulation and real-world evaluations, reaching 98.0% success on LIBERO, 59.25% on RoboCasa-GR1, and 89.02% on RoboTwin~2.0.

cs.RO↗

arXiv2026

A Romanoff-type theorem for $P_2$+{$a^a$: a$\ge$ 1}

Let $\Omega(n)$ denote the number of prime factors of $n$, counted with multiplicity, and put $P_2$={$m$ $\ge$ 1:$\Omega(m)$ $\le$ 2}. We prove that the sumset $P_2$+{$a^a$: a$\ge$ 1} has positive lower density. The proof uses the Romanoff second moment method, in the spirit of Li and Pan's theorem on $P_2$+$2^{\mathcal P}$. The main new ingredient is the following average estimate for the singular factor \[ \frac{1}{K(K-1)} \sum_{\substack{1\le a,b\le K\\a\ne b}} \prod_{p\mid a^a-b^b}\left(1+\frac{\kappa}{p}\right) \le C_\kappa \] for some constant $C_\kappa>0$, which is valid for all $K \ge 2$ and any fixed $\kappa>0$. This estimate controls the average arithmetic correlation among the shifts $a^a$ and allows the Romanoff argument to be carried out.

math.NT↗

arXiv2026

GD$^2$PO: Mitigating Multi-Reward Conflicts via Group-Dynamic reward-Decoupled Policy Optimization

As LLMs advance, post-training reinforcement learning (RL) increasingly relies on multi-dimensional rewards to cultivate comprehensive capabilities. This shift demands new algorithms capable of optimizing diverse and potentially competing objectives simultaneously. To address this, existing methods such as Group reward-Decoupled Policy Optimization (GDPO) decompose the overall score into independent reward groups, then compute the RL loss separately within each group. However, this strategy still encounters multi-reward conflicts: a single rollout can yield positive advantages on certain reward dimensions but negative ones on others, causing opposing signals to cancel each other out during aggregation, further hindering RL training efficiency. Inspired by Dynamic sAmpling Policy Optimization (DAPO), which improves RL training efficiency by filtering out ineffective rollouts with near-zero advantages, we propose Group-Dynamic reward-Decoupled Policy Optimization (GD$^2$PO). Specifically, GD$^2$PO employs a conflict-aware filtering mechanism to mask out rollouts suffering from severe reward-wise disagreement. By preventing conflicting signals from canceling each other out, this masking strategy preserves and enhances the magnitude of effective RL advantages, thereby significantly accelerating learning efficiency. Furthermore, we introduce query-level reweighting to dynamically adjust the update intensity of each query based on its overall reward consensus. Experiments on various multi-reward scenarios, including tool calling and human preference alignment, demonstrate that GD$^2$PO consistently and significantly outperforms existing baselines. The code is available at https://github.com/Qwen-Applications/GD2PO.

cs.LG↗

arXiv2026

An AI Proof of 18-Variable Undecidability for Diophantine Equations over $\mathbb Z[i]$

This paper presents an AI proof that there is no algorithm deciding whether a polynomial equation over the Gaussian integers in $18$ unknowns has a solution. The proof improves the $20$-unknown theorem of Matiyasevich and Sun. It follows their rationality criterion and integer test, but saves two variables: the clearing-denominator variable is avoided by imposing two integer conditions, and the remaining nonzero condition is absorbed into the relation-combining lemma by the one-variable gadget $(2R+1)(3R+1)$.

math.NT↗

arXiv2026

Adjacent comparison bounds and extremal sets for Ruzsa numbers

Let $m$ be a positive integer and $\mathbb{Z}_m$ the residue class ring modulo $m$. The Ruzsa number $R_m$ is defined to be the least integer $r$ such that there is a subset $\mathcal{A}$ of $\mathbb{Z}_m$ satisfying $ 1\le \sigma_{\mathcal{A}}(n)\le r $ for any $n\in \mathbb{Z}_m$, where $$ \sigma_{\mathcal{A}}(n) =\#\big\{(a,a')\in\mathcal A^2: a+a'\equiv n\pmod{m}\big\}. $$ Motivated by a 2024 conjecture of Ding and Zhao, we prove $ | R_{m+1}-R_m|\le 144. $ Let $\mathcal{A}$ be a subset of $\mathbb{Z}_m$ satisfying $1\le \sigma_{\mathcal{A}}(n)\le R_m$ for any $n\in \mathbb{Z}_m$. We also give nontrivial bounds for the size of $\mathcal{A}$. Additionally, we provide exact values of $R_m$ for all $m\le 100$, which substantially extends the table of values given by S\'andor and Yang in 2017. Finally, we pose several related problems and prove some partial results.

math.NT↗

arXiv2026

Si/SiGe multi-channel superlattice structure epitaxial growth with segmented temperature control for Next-Generation Logic Devices

Stacking multiple SiSiGe channels in advanced logic devices faces severe thermal budget accumulation, which degrades interfaces via Ge-Si interdiffusion and strain relaxation.This strategy lowers the Ge diffusion coefficient to 5.6-7% of its value at 650C (Arrhenius estimate), suppressing interdiffusion and preserving pseudomorphic strain. The 4 + 4 channel stack exhibits clear XRD satellite peaks, fully coherent strain state (reciprocal space mapping), sharp interfaces (1.5-2.6 nm transition width) and low RMS roughness (0.08 nm). Quantitative analysis from bottom to top reveals that prolonged high-temperature exposure broadens bottom interfaces and dilutes Ge concentration (from 20% to 18.5%), while the top stack maintains design targets. This work provides a process-physics understanding of thermal budget effects in multi-channel superlattices and establishes a high-quality material foundation for advanced logic devices beyond 2 nm node.

cond-mat.mtrl-sci↗

arXiv2026

Structured Scaling of AI Discovery Across Diverse Scientific Domains

Scientific discovery often requires many cycles of proposing, testing, and refining candidate solutions. Language models can increasingly participate in these loops, but simply generating more attempts does not ensure progress: parallel searches may duplicate one another and iterative refinement may become trapped in poor directions. The central challenge is therefore not only to scale AI-driven discovery, but to structure that scaling so that evaluation signals compound over time. Here we introduce SimpleTES (Simple Test-time Evaluation-driven Scaling), a framework that focuses on the structured scaling of AI discovery loops, organizing evaluator queries across independent trajectories, iterative refinement, local candidate selection, and the selective reuse of evaluated histories. Drawing on structural features of scientific communities, SimpleTES uses a single open-source GPT-OSS model to establish new state-of-the-art solutions across 28 open-ended problems in diverse scientific domains ranging from quantum physics and astronomy to biology, AI, and mathematics. These include a 24.5% reduction in quantum circuit compilation overhead, up to 23% lower propulsive cost for deep-space trajectories, a 2.17x faster lasso-path solver, an 8.5% lower-error whole-brain neural-activity predictor, the fastest reported TriMul kernel, and new mathematical constructions beyond prior human or AI records. We further post-train the model for long-horizon discovery by assigning each attempt the final outcome of the trajectory it helped produce. This improves performance on both training and held-out mathematics problems, further advancing the frontier. Together, these results establish structured scaling as a general mechanism for advancing AI scientific discovery.

cs.LG↗

arXiv2026

Multigrain-aware Semantic Prototype Scanning and Tri-Token Prompt Learning Embraced High-Order RWKV for Pan-Sharpening

In this work, we propose a Multigrain-aware Semantic Prototype Scanning paradigm for pan-sharpening, built upon a high-order RWKV architecture and a tri-token prompting mechanism derived from semantic clustering. Specifically, our method contains three key components: 1) Multigrain-aware Semantic Prototype Scanning. Although RWKV offers a efficient linear-complexity alternative to Transformers, its conventional bidirectional raster scanning is still semantic-agnostic and prone to positional bias. To address this issue, we introduce a semantic-driven scanning strategy that leverages locality-sensitive hashing to group semantically related regions and construct multi-grain semantic prototypes, enabling context-aware token reordering and more coherent global interaction. 2) Tri-token Prompt Learning. We design a tri-token prompting mechanism consisting of a global token, cluster-derived prototype tokens, and a learnable register token. The global and prototype tokens provide complementary semantic priors for RWKV modeling, while the register token helps suppress noisy and artifact-prone intermediate representations. 3) Invertible Q-Shift. To counteract spatial details, we apply center difference convolution on the value pathway to inject high-frequency information, and introduce an invertible multi-scale Q-shift operation for efficient and lossless feature transformation without parameter-heavy receptive field expansion. Experimental results demonstrate the superiority of our method.

cs.CV↗

arXiv2026

The Material Point Method (MPM) for simulating hypervelocity impact on asteroids

Shock-physics numerical codes are essential tools for describing the short but extreme fragmentation stage of the hypervelocity impact process on asteroids. However, accurately representing complex interior structures, surfaces, and contact mechanics in these events remains a significant challenge for traditional hydrocodes. This study introduces and validates an innovative yet underutilized technique, i.e., the Material Point Method (MPM), to simulate hyper-velocity impacts on asteroids. This approach offers new perspectives and solutions for capturing complex interfaces and handling the contact and boundary conditions in asteroid impact simulations. Our MPM implementation incorporates critical improvements to material models, including a pressure-dependent C^1 continuous yield criterion with quantifiable plastic strain, and a resolution-independent Grady-Kipp fragmentation model, to capture the complex physics of geological materials under extreme conditions. The framework is rigorously validated against laboratory impact experiments and benchmarked with smoothed particle hydrodynamics (SPH) simulations, confirming its robustness and precision. Crucially, when applied to asteroid-scale collisions, our model successfully reproduces the formation of large, coherent fragments analogous to (433) Eros. This work establishes MPM as a validated and powerful extension to the planetary scientist's toolkit, enabling the expansion of the parameter space and the treatment of complex contact and boundary conditions, which will enable more realistic simulations of asteroid evolution, family formation, and planetary defense scenarios.

astro-ph.EP↗

arXiv2026

VLA-OPD: Bridging Offline SFT and Online RL for Vision-Language-Action Models via On-Policy Distillation

Although pre-trained Vision-Language-Action (VLA) models exhibit impressive generalization in robotic manipulation, post-training remains crucial to ensure reliable performance during deployment. However, standard offline Supervised Fine-Tuning (SFT) suffers from distribution shifts and catastrophic forgetting of pre-trained capabilities, while online Reinforcement Learning (RL) struggles with sparse rewards and poor sample efficiency. In this paper, we propose On-Policy VLA Distillation (VLA-OPD), a framework bridging the efficiency of SFT with the robustness of RL. Instead of relying on sparse environmental rewards, VLA-OPD leverages an expert teacher to provide dense, token-level supervision on the student's self-generated trajectories. This enables active error correction on policy-induced states while preserving pre-trained general capabilities through gentle alignment. Crucially, we formulate VLA-OPD via a Reverse-KL objective. Unlike standard Forward-KL that induces mode-covering entropy explosion, or Hard-CE that causes premature entropy collapse, our bounded mode-seeking objective ensures stable policy learning by filtering out the teacher's epistemic uncertainty while maintaining action diversity. Experiments on LIBERO and RoboTwin2.0 benchmarks demonstrate that VLA-OPD significantly improves sample efficiency over RL and robustness over SFT, while effectively mitigating catastrophic forgetting during post-training.

cs.RO↗

arXiv2026

DualCoT-VLA: Visual-Linguistic Chain of Thought via Parallel Reasoning for Vision-Language-Action Models

Vision-Language-Action (VLA) models map visual observations and language instructions directly to robotic actions. While effective for simple tasks, standard VLA models often struggle with complex, multi-step tasks requiring logical planning, as well as precise manipulations demanding fine-grained spatial perception. Recent efforts have incorporated Chain-of-Thought (CoT) reasoning to endow VLA models with a ``thinking before acting'' capability. However, current CoT-based VLA models face two critical limitations: 1) an inability to simultaneously capture low-level visual details and high-level logical planning due to their reliance on isolated, single-modal CoT; 2) high inference latency with compounding errors caused by step-by-step autoregressive decoding. To address these limitations, we propose DualCoT-VLA, a visual-linguistic CoT method for VLA models with a parallel reasoning mechanism. To achieve comprehensive multi-modal reasoning, our method integrates a visual CoT for low-level spatial understanding and a linguistic CoT for high-level task planning. Furthermore, to overcome the latency bottleneck, we introduce a parallel CoT mechanism that incorporates two sets of learnable query tokens, shifting autoregressive reasoning to single-step forward reasoning. Extensive experiments demonstrate that our DualCoT-VLA achieves state-of-the-art performance on the LIBERO and RoboCasa GR1 benchmarks, as well as in real-world platforms.

cs.CV↗

arXiv2026

Global Rotation Equivariant Phase Modeling for Speech Enhancement with Deep Magnitude-Phase Interaction

While deep learning has advanced speech enhancement (SE), effective phase modeling remains challenging, as conventional networks typically operate within a flat Euclidean feature space, which is not easy to model the underlying circular topology of the phase. To address this, we propose a magnitude-phase dual-stream framework that aligns the phase stream with its intrinsic circular geometry by enforcing Global Rotation Equivariance (GRE) characteristic. Specifically, we introduce a Magnitude-Phase Interactive Convolutional Module (MPICM) for modulus-based information exchange and a Hybrid-Attention Dual Feed-Forward Network (HADF) bottleneck for unified feature fusion, both of which are designed to preserve GRE in the phase stream. Comprehensive evaluations are conducted across phase retrieval, denoising, dereverberation, and bandwidth extension tasks to validate the superiority of the proposed method over multiple advanced baselines. Notably, the proposed architecture reduces Phase Distance by over 20\% in the phase retrieval task and improves PESQ by more than 0.1 in zero-shot cross-corpus denoising evaluations. The overall superiority is also established in universal SE tasks involving mixed distortions. Qualitative analysis further reveals that the learned phase features exhibit distinct periodic patterns, which are consistent with the intrinsic circular nature of the phase. The source code is available at https://github.com/wangchengzhong/GRE-Net.

cs.SD↗

arXiv2026

Physics Guided Exponential Model Design of High Ge Content SiGe Selective Epitaxy for Gate All Around Source/Drain Applications

High germanium content silicon germanium (SiGe) epitaxy is critical for strain engineering in advanced gate all around (GAA) transistors. This paper demonstrates a physics guided exponential function model that quantitatively links selective epitaxial growth (SEG) parameters to Ge incorporation kinetics in nanoscale trenches. By coupling surface diffusion limited transport, gradient strain, and competitive adsorption dynamics, the model predicts optimal conditions for bottom-up filling with maximal Ge content. For trenches with widths of approximately 60 nm, the optimized process achieved a maximum Ge content of 57.93% and demonstrated 100% selectivity against silicon nitride (SiN) and silicon dioxide (SiO). Cross sectional TEM and EDS analyses reveal a graded Ge profile that minimizes interfacial defects and strain energy. Our results show that the established process physics correlation will significantly facilitate the development of GAA devices with 5nm CMOS technology nodes and beyond.

physics.app-ph↗

arXiv2025

Open-world Hand-Object Interaction Video Generation Based on Structure and Contact-aware Representation

Generating realistic hand-object interactions (HOI) videos is a significant challenge due to the difficulty of modeling physical constraints (e.g., contact and occlusion between hands and manipulated objects). Current methods utilize HOI representation as an auxiliary generative objective to guide video synthesis. However, there is a dilemma between 2D and 3D representations that cannot simultaneously guarantee scalability and interaction fidelity. To address this limitation, we propose a structure and contact-aware representation that captures hand-object contact, hand-object occlusion, and holistic structure context without 3D annotations. This interaction-oriented and scalable supervision signal enables the model to learn fine-grained interaction physics and generalize to open-world scenarios. To fully exploit the proposed representation, we introduce a joint-generation paradigm with a share-and-specialization strategy that generates interaction-oriented representations and videos. Extensive experiments demonstrate that our method outperforms state-of-the-art methods on two real-world datasets in generating physics-realistic and temporally coherent HOI videos. Furthermore, our approach exhibits strong generalization to challenging open-world scenarios, highlighting the benefit of our scalable design. Our project page is https://hgzn258.github.io/SCAR/.

cs.CV↗