Search arXiv⌕ Search

arXiv subjects

Han Yu

Publications and source records attributed to Han Yu.

At least 19 recordsLinked to original sources
arXiv2026

EnvHarness: Awakening Static Worlds for Agent Learning

LLM agents learn by interacting with environments, yet these environments are hand-built and static: blind to an agent's weaknesses, and quickly left behind as it improves. While recent environment generation methods attempt to address this, they require domain-specific pipelines, rely on expensive or unreliable verifiers, and still produce static environments. To alleviate the engineering burden of rebuilding environments from scratch, we propose Environment Harness (EnvHarness), a programmable layer of plug-in components that wraps a static environment to reshape its behavior without modifying the underlying logic. Operating through standard interfaces, EnvHarness applies across diverse domains while ensuring every reshaped environment retains its original verifier. To automate this process, we introduce EnvRigger, which treats the target policy as a black box, observing its execution trajectories to synthesize EnvHarness components targeting diagnosed flaws, and validating them via fresh rollouts. Across five benchmarks in four domains, EnvHarness outperforms both original environments and domain-specific environment generation pipelines, achieving up to a 9.0-point improvement on held-out instances with 9.8% fewer execution steps. Furthermore, EnvHarness provides a superior optimization signal for reinforcement learning, enabling continuous, targeted co-evolution of the policy and its environment.

cs.AI↗
arXiv2026

BICPO-VLA: Behavior-Identified Continuation Preference Optimization for Smooth Asynchronous Vision-Language-Action Control

The request-to-handoff gap has three coupled sources: ambiguity about the behavior intended at request time, physical-state drift accumulated during action generation, and residual incompatibility when the new action finally assumes control. BICPO-VLA addresses them in sequence. First, an instruction-aware causal history encoder identifies the behavior supported by the command and current task progress. Second, sequential Haar subspace generation decomposes each action chunk into complementary pairwise scaffold and residual coefficients, enabling two specialized generation stages followed by exact reconstruction. By reducing iterative refinement in the original action space, it shortens the interval over which the robot continues moving before the new chunk becomes available. Finally, BICPO rolls the known outgoing actions to the actual handoff state and applies reference-relative Flow-DPO among behaviorally matched candidates, adapting the generated chunk to the remaining request-to-handoff mismatch without changing its intended behavior.

cs.RO↗
arXiv2026

Positioning with Flexible Reflectors: Solution and Performance Analysis

Flexible reflectors (FRs) have emerged as a low-cost and energy-efficient solution for reshaping electromagnetic propagation environments across a wide range of applications. This paper investigates FR-swarm-assisted target localization in scenarios where line-of-sight (LoS) paths are unavailable. By leveraging the virtual LoS paths created by the FRs, a simple yet accurate estimator is proposed for localization under severe blockage conditions. To characterize the performance limits of the proposed scheme, we derive the Cramer-Rao lower bound (CRLB) and use it to optimize the positions and orientations of the FRs. Furthermore, by accounting for random FR deployment, we characterize the CRLB distribution and reveal how different network configurations affect localization accuracy. Simulation results demonstrate that the developed scheme closely approaches the CRLB performance, while the derived analytical results provide useful guidelines for FR deployment and network design.

cs.IT↗
arXiv2026

IP Protection in the Era of Visual Generative AI: A Survey

The rapid evolution of visual generative AI has introduced a wide range of intellectual property risks, spanning the unauthorized learning, reproduction, extraction, misuse, and redistribution of protected data and model assets. To address these risks, a growing body of technical defenses has been proposed. However, existing surveys typically organize this literature by lifecycle stage or technical mechanism, which can obscure the protective intent of different methods. This survey presents a two-dimensional taxonomy for IP protection in visual generative models. The primary axis is a Control Logic View, which classifies methods into Information Exposure Control, Generative Behavior Constraint, and Attribution & Accountability according to the risk variable they regulate. The secondary axis distinguishes Data IP from Model IP as cross-cutting asset dimensions. Under this framework, we systematically review protection methods, align evaluation protocols with protection objectives, and discuss open challenges including proactive model-level safeguards, standardized evaluation, robustness against adaptive attacks, and explainable evidence. This survey aims to offer a principled, systematic, and easy-to-follow overview for both new and experienced researchers in visual generative AI IP protection.

cs.CV↗
arXiv2026

PBD-AG: Persistent Baseline-Delta Active Graphs with Uncertainty-Aware Inspection for Long-Horizon Service Robots

Long-horizon service robots require persistent world models that can be built autonomously in unseen environments and revised as task-relevant objects change. Existing methods rely on online mapping, which accumulates localization and observation errors, static scene representations that cannot capture persistent object changes, or holistic vision-language predictions that lack verifiable 3D geometric evidence. We present PBD-AG, a persistent baseline-delta active graph framework that decouples robot-verified stable fixtures from revisable dynamic object events. Under our framework, the robot autonomously bootstraps the structural baseline from onboard exploration and inspects discovered fixtures to ground hierarchical object beliefs. PBD-AG maintains reliability-weighted object states over geometry, semantics, identity, existence, and support relations, utilizing a geometric visibility gate to mitigate false deletions under occlusion. Inspection viewpoints are selected by a graph-conditioned policy that balances target coverage, travel cost, collision risk, and redundant observation. Simulation experiments in multiple environments and under controlled dynamic evaluation show higher aggregate coarse-fixture F1 than capability-matched controls, as well as stronger identity continuity and event recall. A qualitative physical-robot demonstration further illustrates integration with onboard sensing, providing a traceable world model for long-horizon robotic perception. The project page of PBD-AG is available at https://shuobao214.github.io/PBD-AG/

cs.RO↗
arXiv2026

Extreme-RGMT: Continual Learning of Highly Dynamic Skills for Robust Generalist Humanoid Control

Humans can progressively acquire highly dynamic motor skills while preserving reliable everyday motor abilities. In contrast, existing humanoid controllers face a trade-off between generalist and specialist capabilities: generalist motion tracking policies struggle to reliably execute rare highly dynamic motions, whereas specialist training can degrade previously acquired behaviors. We introduce Extreme-RGMT, a two-stage continual learning framework for robust generalist humanoid control. The method first learns a generalist motion-tracking base policy from diverse multi-source motion data, then employs an asymmetric skill acquisition and capability consolidation mechanism to constrain policy drift on mastered motions while emphasizing difficult dynamic segments. To address the scarcity of highly dynamic motions, their high failure rates, and the resulting shortage of informative samples, Extreme-RGMT combines difficulty-aware sampling with advantage-prioritized trajectory resampling to emphasize critical segments. Experiments show that Extreme-RGMT achieves state-of-the-art generalist whole-body motion-tracking performance, including substantially improved completion of challenging highly dynamic motions. The resulting controller directly executes diverse unseen highly dynamic motions under fixed references and online inertial motion-capture inputs, advancing generalist whole-body motion-tracking controllers toward highly dynamic motor capabilities at the human-expert level.

cs.RO↗
arXiv2026

Ego-Dynamics-Augmented World Model for Autonomous Driving with Zero-Shot Cross-Chassis Adaptation

World model (WM)-based reinforcement learning enables sample-efficient end-to-end autonomous driving learning by imagining long-horizon trajectories in latent space. However, most driving WMs operate on bird's-eye-view (BEV) representations that are inherently egocentric: the transition between consecutive frames entangles the ego vehicle's own motion with scene dynamics. As a result, the WM devotes significant capacity to recovering ego-motion from warped observations, at the cost of scene modeling fidelity and imagination accuracy. This work proposes DynaDreamer, a dynamics-augmented Dreamer-style reinforcement learning method to address this problem by augmenting the WM with an explicit ego-dynamics prior. A physics-informed ego-dynamics encoder-decoder extracts the ego-state history into a compact and identifiable context, which modulates a causal Transformer WM to condition both its prior and posterior latents. During imagination, the ego-dynamics predictor propagates this context forward to keep the ego-dynamics prior synchronized with the rollout. An information-theoretic analysis shows that conditioning on this context reduces both the predictive entropy of the observation transition and the prior--posterior Kullback--Leibler divergence, confining the WM's modeling burden to the scene dynamics beyond ego-motion. An additional benefit is zero-shot cross-chassis adaptation: the ego-dynamics context depends on identifiable chassis parameters, so that a vehicle with previously unseen dynamic characteristics can adapt the WM to the new chassis without retraining. Experiments demonstrate that DynaDreamer improves task success rates over the strongest baseline by 28% and 61% in urban and highway driving scenarios, respectively, with the advantage rising to 73% when extrapolating to unseen chassis.

cs.RO↗
arXiv2026

A Survey on Federated Causal Discovery and Inference

Causal reasoning, which encompasses the discovery of causal structures and the inference of causal effects, is fundamental to data-driven decision making. In practice, data for reliable causal analysis are often distributed across institutions and cannot be centralized due to privacy regulations or communication constraints. Federated learning (FL) addresses this by enabling collaborative analysis without raw data sharing, giving rise to the rapidly growing field of federated causal discovery (FCD) and inference (FCI). However, the interdisciplinary nature of this field and the absence of a comprehensive survey present barriers to entry for researchers. This paper bridges that gap by providing a systematic review through multi-dimensional taxonomies. Grounded in the three core design decisions underlying any FCD solution, namely how structures are learned, how data are partitioned, and what structural knowledge each party obtains, we organize FCD along three axes: methodological paradigm, federation topology, and structural scope. We further examine key practical dimensions, including temporal dynamics, data heterogeneity, missing data, and non-identical variable sets. For FCI, we categorize methods by target estimand (average versus individualized/conditional treatment effects) and by estimation strategy, from classical weighting methods to modern deep generative architectures. Unlike prior works that treat FCD and FCI separately, we formalize their connection as complementary stages of a unified federated causal reasoning pipeline, where FCD supplies the structural knowledge required for valid effect estimation in FCI. Finally, we highlight their shared concerns regarding privacy, communication efficiency, theoretical guarantees, and application domains, and conclude by identifying open challenges for future research.

cs.LG↗
arXiv2026

Fourier transform of nonlinear images of self-similar measures: qualitative aspects

The goal of this paper is to establish polynomial Fourier decay for images of self-similar measures $\mu$ on $\mathbb{R}^k$ under sufficiently nonlinear real-analytic maps $f \colon \mathbb{R}^k \to \mathbb{R}^d$. For example, we prove that if $f$ is analytic on $\mathbb{R}^k$, its graph does not lie in an affine hyperplane in $\mathbb{R}^{k+d}$, and $\mu$ is not supported in an affine hyperplane in $\mathbb{R}^k$, then the image measure has polynomial Fourier decay. Key steps in the proof include establishing a uniform Lojasiewicz-type inequality for self-similar measures, and using the decay of the Fourier transform of $\mu$ outside a very small exceptional set of frequencies. As an application of our results, we prove polynomial Fourier decay for self-conformal measures on $\mathbb{C}$ for a large class of complex analytic IFSs which are not self-similar but are conjugate to a linear IFS via an analytic map.

math.DS↗
arXiv2026

LimiX-2M: Mitigating Low-Rank Collapse and Attention Bottlenecks in Tabular Foundation Models

Tabular foundation models (TFMs) increasingly rival tree ensembles, but their performance is often compute-inefficient: with standard affine scalar tokenization, each feature injects value variation through an essentially one-dimensional channel, and feature IDs/positional signals cannot increase within-feature value degrees of freedom, yielding weak early-layer value sensitivity and redundant hidden states. We present a unified tokenize-and-route framework for strong TFMs: RaBEL expands each scalar into compact localized RBF features (optionally exponent-gated) to improve conditioning and shallow-layer effective rank, while a reordered bidirectional block S->N->F aligns computation with the readout by aggregating cross-sample context before feature mixing and using attention pooling. Together, these changes yield LimiX-2M, a 2M-parameter model that outperforms larger TabPFN-v2 and TabICL baselines on widely used tabular benchmarks while reducing training and inference costs. These results highlight value-aware tokenization and readout-aligned routing as key levers for improving the accuracy--efficiency trade-off in TFMs. Model checkpoints and inference code are available at https://github.com/limix-ldm-ai/LimiX.

cs.LG↗
arXiv2026

Foundation Protocol: A Coordination Layer for Agentic Society

Autonomous agents are moving from tools into a layer of social infrastructure: they browse, purchase, deploy software, manage systems, and increasingly interact with one another. As these systems scale, the bottleneck shifts away from raw model capability toward coordination. Agents need to form reliable relationships, organize multi-agent work, exchange value, support an AI economy, and stay safe and accountable under real-world oversight. This paper introduces the Foundation Protocol (FP), a graph-first coordination layer for an emerging human-AI society. FP unifies heterogeneous entities, including agents, tools, resources, humans, institutions, and organizations, and supports native multi-party organization and event-based collaboration. It also provides economic primitives for metering, receipts, and settlement, and treats policy, provenance, and audit as first-class concerns. FP is designed to wrap and bridge existing protocols rather than replace them, enabling incremental adoption while reducing integration and governance overhead. The aim is to keep autonomous agency composable while keeping accountability non-negotiable, so that coordination itself can become shared infrastructure for a human-AI society that is open, pluralistic, and governable.

cs.AI↗
arXiv2026

PrismFlow: Residual Dynamics for Flow Matching in Time-Series Generation

Generating high-quality time-series data is challenging because real-world signals often exhibit multimodal patterns and multiscale dynamics, including oscillations and high-frequency variations. Flow Matching (FM) offers an efficient alternative to diffusion models, but practical implementations typically rely on a single finite-capacity global vector-field estimator. In such heterogeneous temporal distributions, distinct regimes may pass through nearby flow states while requiring incompatible conditional velocities. A monolithic estimator trained with the standard $\ell_2$ velocity-matching objective may therefore learn an overly smoothed approximation of the local transport field. This estimator-level smoothing can attenuate branch-specific dynamics, leading to spectral distortion and poor mode coverage. To address this, we propose PrismFlow, a new FM method with Koopman-inspired dynamical experts. Each expert learns residual corrections in a latent space where local nonlinear temporal evolution can be approximated by linear transitions. We further propose a confidence-aware Winner-Take-All (WTA) objective that updates only the expert best aligned with each sample while masking gradients to the others, encouraging mode-specific specialization. During sampling, the selected expert adds a residual dynamical correction to the global transport field, preserving FM stability while recovering fine-grained and high-frequency temporal structures. Across various benchmarks, PrismFlow effectively mitigates the spectral contraction in standard FM and achieves state-of-the-art performance, with a 15.6% gain in Context-FID and a 38.6% improvement in Discriminative Score, while remaining robust in low-data settings and effective for forecasting and imputation.

cs.LG↗
arXiv2026

Information-Theoretic Decentralized Secure Aggregation with User Dropouts

This paper investigates the fundamental limits of information-theoretic decentralized secure aggregation (DSA) with user dropouts. We consider a fully decentralized network where $K$ users communicate over broadcast channels without a trusted aggregation server. Each user holds a private input and aims to recover the sum of the surviving users' inputs (users may drop) while ensuring that no additional information about individual inputs is revealed to that user, even if it can collude with other users. A two-round communication protocol is considered, where we assume at least $U$ users survive and each user can collude with at most $T$ other users. For this setting, the optimal communication rate region is fully characterized: we show that DSA is infeasible if $U\le T+1$; otherwise, the optimal rate region is given by $R_1\geq 1$ and $R_2\geq \frac{1}{U-T-1}$, where $R_1$ and $R_2$ denote the first- and second-round communication rates, respectively. The proposed aggregation scheme is based on correlated secret keys constructed from $(T+1)$-private maximum distance separable (MDS) matrices, which simultaneously provide robustness against user dropouts and security against collusion. We also derive tight converse bounds that establish the optimality of the proposed scheme. Our result shows that the optimal second-round communication rate depends only on the effective redundancy level $U-T-1$ regardless the total number of users.

cs.IT↗
arXiv2026

Federated Nested Learning: Collaborative Training of Self-Referential Memories for Test-Time Adaptation

We rethink Federated Learning (FL) from a nested learning perspective, framing the core challenge as how to collaboratively learn optimization rules, not just static models, to tackle Non-IID client data. To address this, we propose Federated Nested Learning (FedNL), a novel framework that reformulates FL as a three-level nested optimization system. FedNL embeds Titans-based linear attention into FL, enabling clients to perform lightweight, zero-shot test-time adaptation by treating a delta rule as an online gradient step. Experiments on Non-IID MMLU and long-context benchmarks show that FedNL achieves competitive performance in short-context reasoning, enhances the performance of long-context retrieval and streaming Cross-Entropy, and maintains constant inference memory.

cs.LG↗
arXiv2026

Optimal Communication Rate of Secure Aggregation over Ring Networks with Pairwise Keys

Information-theoretic topological secure aggregation (TSA)\cite{zhang2026information_regular} enables distributed users to compute neighborhood sums over arbitrary networks without revealing individual inputs, while remaining communication-efficient. It has broad applications, including secure model aggregation in decentralized federated learning (FL). Existing TSA formulations rely on arbitrarily correlated keys generated by a trusted key server, which introduces a single point of failure. In this paper, we instead study TSA with \tit{pairwise} secret keys, where each user pair $(i,j)$ shares an independent key $S_{i,j}$. Such keys can be established through inter-user communication, eliminating the need for a key server and improving robustness. Focusing on a ring topology with $K$ users, we characterize the minimum per-user communication rate: \tit{to securely compute one bit of the desired input sum, each user must send at least $1$ bit to its neighbors when $K=3,4$, and at least $2$ bits for all $K\ge 5$}. The higher rate in larger networks arises because each user must simultaneously satisfy two independent key-alignment constraints from its two neighborhoods, which cannot be resolved within a single broadcast symbol under pairwise key independence. We propose a linear pairwise-masking scheme that achieves these rates and prove its optimality via tight entropic converse bounds that exploit the dependency structure of the keys. Notably, for all $K\ge 4$, only a subset of the $\binom{K}{2}$ pairwise keys -- specifically, those between users at ring distance $2$ -- is sufficient to achieve optimality, revealing a nontrivial role of topological sparsity in secure aggregation.

cs.IT↗
arXiv2026

DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence

We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models -- DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSeek-V4-Flash with 284B parameters (13B activated) -- both supporting a context length of one million tokens. DeepSeek-V4 series incorporate several key upgrades in architecture and optimization: (1) a hybrid attention architecture that combines Compressed Sparse Attention (CSA) and Heavily Compressed Attention (HCA) to improve long-context efficiency; (2) Manifold-Constrained Hyper-Connections (mHC) that enhance conventional residual connections; (3) and the Muon optimizer for faster convergence and greater training stability. We pre-train both models on more than 32T diverse and high-quality tokens, followed by a comprehensive post-training pipeline that unlocks and further enhances their capabilities. DeepSeek-V4-Pro-Max, the maximum reasoning effort mode of DeepSeek-V4-Pro, redefines the state-of-the-art for open models, outperforming its predecessors in core tasks. Meanwhile, DeepSeek-V4 series are highly efficient in long-context scenarios. In the one-million-token context setting, DeepSeek-V4-Pro requires only 27% of single-token inference FLOPs and 10% of KV cache compared with DeepSeek-V3.2. This enables us to routinely support one-million-token contexts, thereby making long-horizon tasks and further test-time scaling more feasible. The model checkpoints are available at https://huggingface.co/collections/deepseek-ai/deepseek-v4.

cs.CL↗
arXiv2026

Wafer-scale hybrid molecular beam epitaxy of BaTiO3 and SrTiO3 on silicon

The integration of epitaxial barium titanate (BTO) on silicon represents a highly promising pathway for next-generation, energy-efficient photonic integrated circuits due to BTO's exceptionally high Pockels coefficients. However, the scalable epitaxy of BTO on Si remains hindered by complex stoichiometric control and slow growth rates. In this work, we demonstrate the continuous, uniform wafer-scale growth of high-quality BTO films on SrTiO3 (STO)-buffered 4-inch Si(001) wafers using a fully hybrid molecular beam epitaxy (hMBE) approach. By utilizing titanium tetraisopropoxide as a titanium precursor, we achieve a self-regulating, adsorption-controlled layer-by-layer growth at rates exceeding 75 nm/h, while maintaining an atomically sharp and structurally coherent BTO/STO interface. We systematically compare the structural, ferroelectric, and electro-optic (EO) properties of fully hMBE-grown BTO with those deposited via pulsed laser deposition (PLD) on identical STO/Si templates. While both techniques yield high-quality c-domain dominated films, the optimized hMBE-grown BTO exhibits superior crystallinity and a larger effective EO coefficient of 248 pm/V, surpassing that of the PLD-grown films (220 pm/V). These results highlight the advantages of the fully hMBE approach as a scalable, deterministic, and high-performance materials platform for wafer-scale integrated ferroelectric photonics.

cond-mat.mtrl-sci↗
arXiv2026

From Transfer to Collaboration: A Federated Framework for Cross-Market Sequential Recommendation

Cross-market recommendation (CMR) aims to enhance recommendation performance across multiple markets. Due to its inherent characteristics, i.e., data isolation, non-overlapping users, and market heterogeneity, CMR introduces unique challenges and fundamentally differs from cross-domain recommendation (CDR). Existing CMR approaches largely inherit CDR by adopting the one-to-one transfer paradigm, where a model is pretrained on a source market and then fine-tuned on a target market. However, such a paradigm suffers from CH1. source degradation, where the source market sacrifices its own performance for the target markets, and CH2. negative transfer, where market heterogeneity leads to suboptimal performance in target markets. To address these challenges, we propose FeCoSR, a novel federated collaboration framework for cross-market sequential recommendation. Specifically, to tackle CH1, we introduce a many-to-many collaboration paradigm that enables all markets to jointly participate in and benefit from training. It consists of a federated pretraining stage for capturing shared behavior-level patterns, followed by local fine-tuning for market-specific item-level preferences. For CH2, we theoretically and empirically show that vanilla Cross-Entropy (CE) exacerbates market heterogeneity, undermining federated optimization. To address this, we propose a Semantic Soft Cross-Entropy (S^2CE) that leverages shared semantic information to facilitate collaborative behavioral learning across markets. Then, we design a market-specific adaptation module during fine-tuning to capture local item preferences. Extensive experiments on the real-world datasets demonstrate the advantages of FeCoSR over other methods.

cs.IR↗