Searcharxiv⌕ Search

arXiv subjects

Cheng Chen

Publications and source records attributed to Cheng Chen.

At least 37 records · Page 2Linked to original sources
arXiv2026

AnyWorld: Factorized Egocentric World Models for Cross-Embodiment Generalization

Collecting contact-rich robot experiences at scale remains a major bottleneck for generalizable manipulation. Beyond data quantity, robot learning also requires diverse experiences across embodiments, viewpoints, and scenes. Human egocentric videos provide abundant physical interactions, but each video captures only a narrow slice of experience under a single body, camera trajectory, and environment. We propose AnyWorld, a cross-embodiment world modeling framework that expands a single human interaction into diverse robot-native rollouts without paired human-robot demonstrations. Our model factorizes an interaction into action, camera, and embodiment: action controls capture the motion structure, camera controls specify viewpoint evolution, and the target embodiment context defines the acting body and its interaction geometry. This formulation enables independent recomposition of embodiment, viewpoint, and scene factors, allowing a single model to generate many robot-domain experiences while preserving the underlying dynamics and object interactions. We train the model with large-scale human interaction pretraining followed by mixed-embodiment fine-tuning. Experiments show that our model supports controllable recomposition across embodiments, viewpoints, and scenes, and we further demonstrate that the generated data can improve manipulation performance on the RoboCasa GR1 tabletop benchmark and a real IRON humanoid robot. Beyond aggregate gains, we test whether unpaired human experience can be recomposed into robot-native video-action pairs that target a policy gap. Controlled IRON interventions correct a spurious completion prior and establish language-grounded spatial target selection; an action-only counterfactual intervention fails to learn the latter reliably, showing that both action calibration and visual recomposition are necessary.

cs.RO↗
arXiv2026

From Language to Behavior: Scaling Sequence Transformers for Industrial Recommendation Ranking with Rec-Native Designs

Scaling Transformers has driven large gains in language modeling, but transplanting this to behavior-sequence modeling in production ranking is challenging: recommendation differs in signal quality, where behavior sequences are noisy, temporally irregular, and sparsely supervised, and in computation asymmetry, where each request scores many candidates against one shared user history under tight latency budgets. We propose ReST, a recommendation-native Transformer scaling framework. For signal quality, it introduces a sequence encoder with dual-gated attention, rotary positional and temporal embedding, stabilized residual normalization, and training-only auxiliary objectives. For computation asymmetry, it factorizes ranking into a heavy reusable encoder and a lightweight cross decoder with projection-free KV attention and token-specific parameterization, coupling user-level shared-prefix training with shared-prefix serving for compute-once, decode-many-times ranking. Across industrial and public benchmarks, ReST achieves higher accuracy and scales more consistently along sequence length, depth, and width, where LLM-style Transformer blocks saturate. A one-week online A/B test on a production advertising platform improves online AUC by 1.31% and lifts a core revenue metric by 11.93% within a 50 ms P99 budget; ReST has since been fully deployed in production, showing that behavior-sequence scaling remains a promising, under-exploited axis for production ranking.

cs.IR↗
arXiv2026

On the formation of retrograde S-type planets in binaries with a polar circumbinary disk

Retrograde S-type planets have been observed in several binary systems, yet their formation pathway remains poorly understood. With high-resolution hydrodynamic simulations, we demonstrate that a polar circumbinary disk around an eccentric, unequal-mass binary can form and sustain a retrograde mini disk around the primary star. This provides a direct in-situ formation channel for retrograde S-type planets. The mini disk forms via a sub-Keplerian accretion stream that is slightly misaligned from the polar disk. The mini disk initially undergoes von Zeipel-Kozai-Lidov (ZKL) oscillations, driving coupled eccentricity and inclination evolution. Rather than oscillating indefinitely, the inner mini disk evolves past the critical ZKL inclination, decouples from the outer disk, and settles into a stable retrograde orbit. This evolution is sensitive to numerical resolution: the retrograde configuration is absent in previous lower-resolution simulations, where the mini disk accretion timescale is too short to sustain ZKL-driven evolution. For a higher disk viscosity, the mini disk remains near-polar due to a shorter accretion timescale. Since protoplanetary disks typically have low viscosity, our results suggest that retrograde S-type planets can form in-situ from retrograde mini disks around polar circumbinary disks, and their occurrence rate may be higher than currently estimated.

astro-ph.EP↗
arXiv2026

Observation of electron spin interactions between Rydberg atoms

We report the observation of electron spin interactions between Rydberg atoms, which are driven by spin-orbit coupling through second-order dipole perturbation and exhibit a spatial anisotropy governed by the atomic configuration. Specifically, we observe coherent electron spin exchange dynamics, with the measured coupling strength agreeing well with both numerical calculations and theoretical models. Furthermore, we show that global microwave dressing enables active engineering and dynamical freezing of the spin exchange by introducing a differential AC Stark shift between the participating states. Additionally, we achieve tunability of the interaction by applying a stronger magnetic field, which effectively modifies the energy contributions of the underlying spin-orbit coupling channels. Finally, measuring spin dynamics in one-dimensional multi-atom chains aligned parallel or perpendicular to the magnetic field provides a self-consistent validation of the anisotropic XXZ framework. This electron spin interaction natively features spin-position coupling and enables a natural mapping onto the Heisenberg-Kitaev model. Our findings reveal a new class of electron spin-spin interactions among Rydberg atoms, expanding the scope of quantum simulation with Rydberg atom arrays.

cond-mat.quant-gas↗
arXiv2026

Adapting Dense Vision-Language Relationships for Multi-label Classification with Partial Label

Learning multi-label image classification with incomplete annotations is a challenging task that has been widely studied for its superior trade-off between high efficiency and less labor consumption on large-scale datasets. Predominant methods rely on strong prior assumptions to recover the missing semantics from partial annotations. However, these statistic priors suffer from unstable semantic mistakes and thus lead to catastrophic overfitting. Toward this end, we propose a Language-driven Dense Semantic Adaptor (LDSA) that excavates prior-adaptive relationships from multimodal pretrained CLIP models. In our approach, the densely contrastive adaptor is first proposed to construct dense visual contrastive constraints, transferring the task-specific knowledge to visual domains. We then propose a language-driven interactive decoder with the help of class-specific prompt tuning, which adapts language proxies with visual domains. With the collaborative learning of proposed modules, experimental results demonstrate our proposed LDSA achieves a new state of the art on public multi-label classification benchmarks, and interpretable analyses reveal that our LDSA discovers implicit semantic relationships with the prior-adaptive learning scheme.

cs.CV↗
arXiv2026

VLZip: Unified Visual and Textual Compression for Interleaved Long-Context Modeling

Vision Language Models (VLMs) face significant challenges with ultra-long, interleaved image-text sequences due to the quadratic complexity of self-attention. Current solutions either resort to aggressive token pruning, risking irreversible information loss, or adopt efficient but less precise architectures, while largely ignoring the equally vital textual component. We introduce VLZip, a framework that unifies visual and textual compression for high-fidelity reasoning within a pure Transformer. At its core, VLZip hierarchically distills visual and textual segments into compact, layer-specific "soft prefixes" and injects them into each decoder layer's hidden states, drastically shortening the attention sequence while preserving fine-grained global context. To address deficient evaluations in the field, we also introduce LongVLBench, a new benchmark derived from video narratives that demands holistic, narrative-level reasoning. Extensive experiments show VLZip achieves leading performance on long-context multimodal reasoning, enabling training up to 120K tokens, a 6x increase over the baseline, and inference beyond 280K tokens with significantly reduced memory, while demonstrating the memory scalability to handle up to 2M tokens. By excelling at extreme context lengths where existing methods collapse, VLZip establishes an efficient and powerful new standard for long-context multimodal AI. Code is available at https://github.com/ShareLab-SII/VLZip.

cs.CV↗
arXiv2026

Kimi K2.5: Visual Agentic Intelligence

We introduce Kimi K2.5, an open-source multimodal agentic model designed to advance general agentic intelligence. K2.5 emphasizes the joint optimization of text and vision so that two modalities enhance each other. This includes a series of techniques such as joint text-vision pre-training, zero-vision SFT, and joint text-vision reinforcement learning. Building on this multimodal foundation, K2.5 introduces Agent Swarm, a self-directed parallel agent orchestration framework that dynamically decomposes complex tasks into heterogeneous sub-problems and executes them concurrently. Extensive evaluations show that Kimi K2.5 achieves state-of-the-art results across various domains including coding, vision, reasoning, and agentic tasks. Agent Swarm also reduces latency by up to $4.5\times$ over single-agent baselines. We release the post-trained Kimi K2.5 model checkpoint to facilitate future research and real-world applications of agentic intelligence.

cs.CL↗
arXiv2026

Beyond Retrieval: Analytic Memory for Multimodal Agents

Long-term multimodal memory must support not only retrieving relevant information but also computing over observations accumulated across interactions. Existing systems largely emphasize \emph{retrieval memory}, organizing interaction histories through summaries and indexes to return query-relevant information at multiple granularities, from high-level abstractions to underlying records. In this paper, we formulate \emph{analytic memory} as a complementary abstraction that organizes recurring multimodal observations into queryable structures supporting filtering, aggregation, ranking, and temporal comparison. We present AdaMM, a framework that jointly supports retrieval and analytic memory. Rather than relying on application-defined schemas, AdaMM extracts provenance-linked attribute-value observations from dialogue, images, and contextual metadata, discovers recurring field structures, and materializes them for analytical access. At inference time, a memory-aware planner decomposes queries into retrieval and analytic operations and routes each operation to the appropriate tools. Experiments on two long-term multimodal memory benchmarks, MemEye and MemGallery, show that AdaMM improves performance by up to 11.3\% and 6.9\%, respectively.

cs.AI↗
arXiv2026

MoWorld: A Flash World Model

The future of World Models depends not only on scaling model capability, but also on scaling practicality and inference efficiency. High-frame-rate inference enables responsive perception, planning, and control in real-world autonomous systems. To this end, we present MoWorld, a cost-effective yet high-performance Flash World Model with an end-to-end framework spanning data generation, pre-training, distillation, and efficient inference, enabling up to 50 FPS real-time interaction with cinematic visual quality without the need of high-end GPUs. To enable large-scale real-world deployment, MoWorld jointly optimizes model capability and cost throughout the entire development pipeline. Specifically, unlike existing approaches that primarily rely on large-scale video corpora, MoWorld is built upon a scalable 3D-native data engine accumulated from our large-scale 3D vision and generative modeling pipeline, enabling the efficient construction of geometrically consistent training data across diverse real-world and synthetic environments. Based on this foundation, a curriculum cross-frame pre-training strategy for stable and scalable World Model learning, an efficient denoising-step distillation algorithm to reduce diffusion training cost, and a mixed-precision parallel inference framework for low-cost real-time deployment. MoWorld is the first real-time interactive World Model built on the Neural Processing Unit (NPU) and can achieves up to 50 FPS in such the devices, enabling practical and efficient deployment at scale. Comprehensive evaluations demonstrate that MoWorld achieves leading performance; notably, its average inference cost is only 30\%-50\% of that of existing World Models, providing a practical foundation for large-scale real-world applications of World Models. We also demonstrate diverse applications of MoWorld.

cs.CV↗
arXiv2026

iMontage: Unified, Versatile, Highly Dynamic Many-to-many Image Generation

Pre-trained video models learn powerful priors for generating high-quality, temporally coherent content. While these models excel at temporal coherence, their dynamics are often constrained by the continuous nature of their training data. We hypothesize that by injecting the rich and unconstrained content diversity from image data into this coherent temporal framework, we can generate image sets that feature both natural transitions and a far more expansive dynamic range. To this end, we introduce iMontage, a unified framework designed to repurpose a powerful video model into an all-in-one image generator. The framework consumes and produces variable-length image sets, unifying a wide array of image generation and editing tasks. To achieve this, we propose an elegant and minimally invasive adaptation strategy, complemented by a tailored data curation process and training paradigm. This approach allows the model to acquire broad image manipulation capabilities without corrupting its invaluable original motion priors. iMontage excels across several mainstream many-in-many-out tasks, not only maintaining strong cross-image contextual consistency but also generating scenes with extraordinary dynamics that surpass conventional scopes. Find our homepage at: https://kr1sjfu.github.io/iMontage-web/.

cs.CV↗
arXiv2026

Perturbed Dyadic Cubes and Quantitative Estimates for Schrödinger Operators with Potentials in $RH^{n/2}$

Let $L:=-Δ+V$ be a Schrödinger operator on the Euclidean space $\mathbb{R}^n$ with potential $V$ in the reverse Hölder class $RH^{n/2}$ satisfying some mild assumptions that $V$ neither decays too rapidly nor oscillates violently at infinity. In this paper, the authors construct a new system of dyadic cubes $\mathcal{D}^V$ that reflects the intrinsic geometry perturbed by $V$. Then using the quantitative geometric information of $\mathcal{D}^V$, the authors characterize the $L^p$ operator norm of the Riesz potential $L^{-α/2}$ for all $α\in (0,2]$ and $p\in (1,\infty)$. As applications, some quantitative spectral estimates for $L$ are given.

math.AP↗
arXiv2026

OrthKD: Extracting Generalized Clinical Knowledge from Heterogeneous Teachers for Lightweight Deployment

Deploying diabetic retinopathy (DR) screening models in primary care requires edge-efficient systems that remain accurate, safe, and reliable under domain shift. Multi-teacher knowledge distillation (KD) is a natural compression strategy, but existing approaches largely assume that all teachers provide equally trustworthy supervision. In our setting, this assumption fails: a strong CNN teacher (EfficientNet-B3, 0.876 QWK) and a weaker Transformer teacher (Swin-Base, 0.830 QWK) are complementary, yet the Transformer's logits can still mislead the student. We therefore propose OrthKD, a selective-trust distillation framework that transfers full supervision from the strong CNN, uses feature-only distillation from the weak ViT, and enforces orthogonality between teacher-specific student projections to encourage complementary rather than redundant evidence. This design preserves local lesion precision, injects global structural context, and improves robustness to distribution shift. On 132,049 retinal images, a 5.4M-parameter MobileNetV3 student reaches 0.885 QWK on EyePACS and improves zero-shot Messidor-2 performance from 0.507 to 0.728 QWK, while also achieving strong referral AUC and calibration. These results show that selectively distilling heterogeneous teachers can enable practical DR screening on resource-constrained devices.

cs.LG↗
arXiv2026

Electromagnetic form factors of singly charmed baryons $Σ_c$ and $Λ_c$ in a covariant quark-diquark model

We present a systematic study of the spacelike electromagnetic form factors of the ground-state singly charmed baryons, $Σ_c$ ($Σ_c^{++},Σ_c^+,Σ_c^0$) and $Λ_c^+$, within a covariant quark-diquark model. Based on this framework, we obtain their magnetic moments as well as electric charge and magnetic moment radii. Such observables are important to understand the internal structure and the inner dynamics of these heavy baryon states. Our theoretical calculations are in qualitative agreement with available lattice QCD results of $Σ_c^{++}$ and $Σ_c^0$ baryons. We also discuss the mechanism behind the dependence of the numerical results on the charm quark and light diquark. A key finding is that the electric form factors of the singly charmed baryons fall off much more slowly with momentum transfer $Q^2$ than that of the proton, indicating a more compact electric charge distribution, which is attributed to the heavy charm quark acting as a localized core. More importantly, we observe a striking difference in the magnetic structure: while the magnetic form factors of $Σ_c$ are dominated by the light axial-vector diquark, those of $Λ_c^+$ are unexpectedly governed by the charm quark due to the vanishing contribution of the scalar diquark. This highlights the decisive role of the light-diquark spin configuration in determining the magnetic properties. Finally, using an empirical asymptotic relation connecting the spacelike and timelike regions, we predict the total cross section for $e^+e^-\toΣ_c\barΣ_c$. Our findings can be tested at existing facilities including the BESIII, Belle II and LHCb, as well as the proposed Super Tau-Charm Facility.

hep-ph↗
arXiv2026

LaRec: Unleashing LLM-based Latent Reasoning for Generative Recommendation

Large Language Models (LLMs) have shown great promise in recommendation due to superior reasoning abilities. However, existing methods mainly rely on explicit Chain-of-Thought (CoT), resulting in verbose reasoning texts and inefficient response times. latent reasoning aims to balance efficiency by thinking within a continuous latent space, yet it faces two major challenges: (1) Lack of Fine-grained Supervision: Latent reasoning relies solely on feedback from the final labels, providing sparse supervisory signals that struggle to effectively guide the optimization of multiple hidden reasoning steps. (2) Single Reasoning Path: The deterministic nature of latent reasoning impedes the exploration of users' diverse interests and preferences, thereby limiting the recommendation capabilities of LLMs. To address these issues, we propose \textbf{$LaRec$}, an efficient generative recommendation framework designed to unleash the potential of latent reasoning in LLMs. $LaRec$ consists of two core stages: First, we design Latent Pre-training that empowers LLMs with latent reasoning capabilities by providing rich supervisory signals to the latent space reasoning via step-level alignment and process direction alignment. Second, we introduce Personalized RL-tuning. Specifically, we construct a personalized Gaussian Mixture Distribution for each user based on their historical interests. By randomly sampling distinct reasoning starting points from this distribution during training, we guide the LLMs to traverse diverse reasoning paths within the latent space, enabling efficient exploration of user's multi-faceted interests. Experiments on multiple datasets show that $LaRec$ significantly outperforms existing baselines with comparable efficiency.

cs.IR↗
arXiv2026

AgentJet: A Distributed Swarm Training Framework for Agentic Reinforcement Learning

Training reinforcement learning (RL) policies for large language model (LLM) agents requires optimizing multi-turn trajectories that interact with external environments. Existing training frameworks struggle with runtime failures, single-model constraints, incompatible task environments, and redundant context. We present AgentJet, a distributed swarm training framework based on a decoupled multi-node architecture. AgentJet treats the server--client topology as configurable: swarm servers host trainable models and perform optimization on GPU clusters, while detachable swarm clients execute arbitrary agents and communicate through OpenAI-compatible APIs. Reconfiguring this topology supports heterogeneous multi-model RL, mixed-task training with isolated runtimes, fault-tolerant execution, and live code iteration through hot-swappable clients. AgentJet also introduces context tracking with timeline merging, reducing actor-update time by 6.25x on AppWorld. The same detachable-client design supports an automated research system that conducts long-horizon, multi-day RL studies on large-scale clusters with limited human intervention. AgentJet is open-source and compatible with agent systems that issue standard LLM inference requests.

cs.AI↗
arXiv2026

HOMIE: Human-object Centric Video Personalization via Multimodal Intelligent Enhancement

Human-object centric video personalization (HOCVP) is a core task within subject-driven video generation. However, existing methods suffer from two key limitations. First, most approaches focusing on inter-subject personalization still struggle to strike a balance between high subject fidelity and accurate interaction patterns between humans and diverse objects, especially when objects represent abstract concepts such as logos. Second, while intra-subject references (e.g., OCR maps, multi-view inputs) are expected to enhance subject fidelity, most existing works lack mechanisms to understand such latent correspondence. To address both challenges, we propose HOMIE, an HOCVP framework that tackles both inter- and intra-subject input settings in a unified manner. Compared to previous approaches, HOMIE proposes a better MLLM integration strategy to extract knowledge of reference-level relationships without compromising the controllability of text encoders or incurring costly re-alignment. Specifically, we introduce global multimodal guidance within self-attention to better align MLLM-derived semantic features with VAE tokens. Furthermore, we propose modality-reference embedding to differentiate tokens from MLLM features and VAE tokens and associate intra-subject reference image tokens. Extensive experiments validate that our method achieves state-of-the-art performance across various HOCVP tasks. Project Page: https://yiyangcai.github.io/homie-page.github.io/

cs.CV↗
arXiv2026

Symmetric Rank-$k$ Methods

This paper proposes a novel class of block quasi-Newton methods for convex optimization which we call symmetric rank-$k$ (SR-$k$) methods. Each iteration of SR-$k$ incorporates the curvature information with~$k$ Hessian-vector products achieved from the greedy or random strategy. We prove that SR-$k$ methods have the local superlinear convergence rate of $\mathcal{O}\big((1-k/d)^{t(t-1)/2}\big)$ for minimizing smooth and strongly convex function, where $d$ is the problem dimension and $t$ is the iteration counter. This is the first explicit superlinear convergence rate for block quasi-Newton methods, and it successfully explains why block quasi-Newton methods converge faster than ordinary quasi-Newton methods in practice. We also leverage the idea of SR-$k$ methods to study the block BFGS and block DFP methods, showing their superior convergence rates.

math.OC↗