arXiv subjects

Rui Li

Publications and source records attributed to Rui Li.

At least 19 recordsLinked to original sources

arXiv2026

Dreaming in Flow: Bidirectional Self-Refinement for Unified Multimodal Models

Unified multimodal models can generate images and interpret them, but their two capabilities are usually optimized with separate supervision. We introduce Bidirectional Self-Refinement via Flow-Direction Feedback and Semantic Replay (BiSe), a post-training method that uses self-generated visual episodes to construct feedback for both pathways. Understanding-guided generation aligns the prompt-conditioned flow direction with an image-conditioned prediction and with the model's response to a repair instruction. Generation-guided understanding replays the source intent through captioning and image generation, and learns a stable semantic evidence profile across the replayed images while separating an independent episode. Flow-matching and source-prompt objectives remain active, and frozen initial-model predictions limit drift. BiSe requires no paired image--text data during post-training. It improves text-to-image compositionality across three unified backbones and visual understanding on the two backbones evaluated for that task.

cs.CV↗

arXiv2026

Cumulative Marginal Matching for One-Step Diffusion

The sampling procedures of diffusion and flow-matching models involve hundreds of denoising steps incurring substantial inference time, which has motivated a growing body of work on training schemes for few-step and one-step generation. Existing approaches primarily seek to align the generation trajectory of few-step and base models, yet their training objectives only involve discrepancies at a single point in the trajectory, encouraging only local \emph{point-level} alignment. While recent methods such as \emph{Shortcut-CSL} introduce cumulative objective that enforce alignment over the entire trajectory, this is suboptimal as they do not directly enforce alignment between the distributions induced by the two models throughout the generation process. We propose \emph{Cumulative Marginal Matching} (CMM), a principled objective for few-step generation: by viewing flow-matching as a sequence of intermediate marginal distributions connecting the prior and data distributions, CMM directly optimizes the cumulative discrepancy between the base and few-step marginal distributions over the entire sequence. Specifically, we formulate this discrepancy using the Wasserstein-2 distance. By exploiting the pushforward representation of flow-matching, we transform an intractable cumulative Wasserstein objective into a tractable surrogate that can be optimized using samples from the intermediate marginal, which are readily generated through the conditional interpolation process of flow-matching. Furthermore, we show that the existing \emph{Shortcut-CSL} objective arises as a first-order approximation of that of the CMM, establishing a theoretical connection between the two objectives. Experiments on standard benchmarks demonstrate improvement in generation quality over \emph{Shortcut-CSL} and other state-of-the-art one- and few-step baselines under identical training budgets.

cs.LG↗

arXiv2026

SPA: Structure-Preserving Approximation of Gated Activations under FHE

Nonlinear activation evaluation remains a major bottleneck in neural inference under fully homomorphic encryption (FHE). FHE supports efficient addition and multiplication, but lacks native operations for exponential functions, reciprocals, and value-dependent selection. High-degree polynomial approximations on fixed intervals are therefore commonly used to replace nonlinear activations. They can achieve small errors but often incur substantial multiplicative depth and latency. We observe that several widely used nonlinear activations share the algebraic structure $A(x)=xg(x)$ with $g(-x)=1-g(x)$. This class includes GELU, SiLU, fixed-parameter Swish, and the scalar nonlinear branches of GEGLU/SwiGLU. We propose \SPA{} (\emph{Structure-Preserving Approximation}) to exploit this shared structure. \SPA{} retains the exact odd component $x/2$ and approximates only the even residual as a polynomial in $x^2$, yielding a comparison-free arithmetic kernel. This construction removes redundant odd degrees of freedom and enables lower-cost polynomial evaluation at a given error target. Plaintext experiments with Gemma-4-E2B and Llama-3.2-3B show that SPA largely preserves downstream task performance without weight updates. Fixed-interval CKKS benchmarks show favorable error--latency trade-offs. On $[-5,5]$, \SPA{}-balanced achieves a $3.25\times$ speedup over MOAI (ICLR'~26) for erf-GELU with comparable error. On the same interval, \SPA{}-accurate achieves $79.22\%$ lower root mean squared error (RMSE) and a $10.40\times$ speedup over Euston (S\&P'~26) for QuickGELU. For SiLU, \SPA{}-accurate achieves $19.02\%$ lower RMSE and a $2.06\times$ speedup over THOR (CCS'~25).

cs.CR↗

arXiv2026

Non-Monotonic Exciton Transport Efficiency in Multi-layer Ring-shaped Quantum Networks inspired by Photosystem I

Photosystem I (PSI) is one of the most efficient natural photosynthetic complexes and serves as a structural prototype for understanding excitation energy transfer (EET) and designing artificial light-harvesting systems. Here, we construct a multilayer elliptical-conical quantum network inspired by PSI to investigate how antenna architecture affects EET. Excitation dynamics are simulated using the quantum jump approach combined with coherent modified Redfield theory (QJA-CMRT). The transfer efficiency exhibits a pronounced nonmonotonic dependence on pigment density, reaching a maximum of 0.86 at x=6 in the five-layer network. The optimal configuration contains 91 pigments, close to the number of core antenna pigments in natural PSI. Our combined density-of-states and kinetic analyses reveal that this nonmonotonic dependence originates from a trade-off: at low pigment densities, enhanced band overlap promotes energy transfer, whereas at high densities, excessive exciton delocalization compromises funnel directionality and reduces efficiency; the optimal configuration is consequently achieved at x=6. These findings provide insights into PSI antenna organization and offer guidelines for designing efficient artificial light-harvesting systems.

cond-mat.soft↗

arXiv2026

Beyond Actions: Learning Future Operation Targets for Vision-Language-Action Models

Actions specify what a robot should execute, but do not describe what those commands will achieve. Vision-language-action (VLA) policies nevertheless learn primarily by predicting action chunks from images, language, and state. We introduce Action--Future Coupled Effect Learning (AFCE), which makes a future-grounded physical Effect the policy's primary prediction target. An offline tokenizer constructs temporally ordered Effect tokens from demonstrated actions, realized robot responses, and sparse visual transitions. Reconstruction objectives encourage the representation to retain executable detail and observed physical change. A flow-matching VLA predicts Effects from current images, instruction, and robot state, while a state-conditioned decoder converts them into action chunks. Embodiment-specific interfaces allow single- and dual-arm tasks to share the Effect model. On DexJoCo, AFCE achieves $50.97\%$ overall success, improving over $π_{0.5}$ and UVT by 11.82 and 15.21 percentage points, respectively. It leads on all five bimanual tasks, highlighting the value of Effect supervision for coordinated dexterous manipulation. The project page is available at https://afceproject.github.io/AFCE-Web/.

cs.RO↗

arXiv2026

AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned Prefixes

Generative recommendation models based on autoregressive decoding have been adopted as alternatives to classical ranking-based systems. While their modeling flexibility has been widely studied, their resilience to data poisoning remains largely unexplored. In this work, we ask: how susceptible are generative recommenders to conditional targeted poisoning? Under a realistic setting where an adversary can inject limited malicious interactions and query a recommendation API as a black box, our empirical study shows that generative recommenders are vulnerable: injecting fewer than 1% synthetic interactions significantly increases the likelihood of promoting malicious targets under the trigger. To characterize this vulnerability, we analyze the synthesis of malicious interactions from two perspectives: (i) decoding alignment and (ii) prefix popularity. We observe that short, context-aligned prefixes that are both common under the model's induced transition dynamics and strongly linked to a frequent trigger consistently amplify targeted promotion. Motivated by this analysis, we develop AlignPoison, a black-box framework that discovers and samples context-aligned prefixes from API outputs. We conduct experiments on benchmarks in unimodal and multimodal generative recommendation settings. Results show that these prefixes outperform competitive baselines, improving ASR lift by up to 16.3 percentage points and highlighting a decoding-dependent, context-sensitive poisoning vulnerability in generative recommendation models.

cs.AI↗

arXiv2026

CatPortrait: Taming Video Generative Models for Portrait Animation via In-context Conditioning

Recent video generative models have shown strong capability in synthesizing high-quality and temporally coherent videos. Yet, adapting them for controllable portrait animation remains nontrivial, as existing methods typically rely on architectural changes or extra conditioning modules, increasing complexity and reducing compatibility with pretrained models. We present CatPortrait, a simple and effective framework that enables controllable portrait animation via 3D face guidance as in-context conditioning. CatPortrait preserves the original video generation backbone and injects structured 3D facial guidance into the input space through spatial concatenation, allowing pretrained image-to-video models to animate portraits without sacrificing their pretrained priors. CatPortrait supports both video-driven and audio-driven portrait animation within a unified framework. Extensive experiments demonstrate that CatPortrait achieves competitive or superior results across both settings.

cs.CV↗

arXiv2026

Neural Estimation of Summary Statistics for Simulation-Based Inference with Non-Asymptotic Guarantees

Simulation-based inference (SBI) enables Bayesian inference when a simulator is available but the likelihood is intractable. Modern SBI methods train neural networks on simulated data to obtain an amortized approximation of the posterior, making inference for new observations fast. A central challenge in SBI is to extract informative, low-dimensional summary statistics; without them, performance can degrade rapidly as the observation dimension grows. We therefore highlight summary learning as a critical component of SBI and propose a general framework that jointly learns a neural-network-parameterized summary and an amortized posterior estimator. We further introduce a local refinement strategy that adapts the amortized estimator to a specific observation, thereby reducing the amortization gap. Both amortized training and local refinement accommodate implicit posterior estimators without tractable densities. On the theoretical side, we analyze the finite-sample behavior of the joint learning procedure through the lens of nonparametric conditional distribution estimation, and show on representative examples how low-dimensional summaries can mitigate the classical curse of dimensionality. Empirically, our results support this motivation and demonstrate favorable performance relative to existing methods.

stat.ME↗

arXiv2026

Learning from Evolving Errors: Adaptive Iterative Repair for On-Policy Distillation

On-policy self-distillation (OPSD) supplies dense token-level feedback on trajectories sampled from the student's own policy, a richer training signal than the outcome-level rewards of reinforcement learning. This feedback comes from a teacher conditioned on a full reference solution unavailable to the student. The reference solution specifies the target but not how to move from the student's current error toward it, creating a solution-conditioned shortcut risk. We introduce AIR-OPD, an adaptive iterative repair framework for on-policy distillation that provides error-to-repair supervision. Given a failed response, a guidance generator synthesizes repair guidance for the current error. The student samples an on-policy retry with this guidance. If the retry remains incorrect, the generator produces new repair guidance for the newly observed error. At each round, a fixed teacher receives the guidance as privileged context and supervises the student on an error-aligned region of its latest failed response. Outcome-aware stage weighting favors early repair stages and credits stages whose immediate retry passes verification. We train AIR-OPD on the DAPO-Math-17K dataset and evaluate on AIME24, AIME25, and HMMT25, alongside out-of-distribution tests on MMLU-Pro and GPQA. We examine two guidance sources, self-guidance from the current student policy and external guidance from a larger model. For both Qwen3-4B and Qwen3-8B, AIR-OPD attains the best mathematical-reasoning averages, improving over the strongest baseline by up to 3.6 points, while preserving base-model performance on the out-of-distribution benchmarks.

cs.LG↗

arXiv2026

RoboChemGym: A Protocol-Driven Generative Simulation Framework for Long-Horizon Chemical Manipulation

Wet-lab experimentation serves as the gold standard for hypothesis verification in scientific discovery; yet it is inherently labor-intensive, costly, and safety-critical. Embodied agents hold the promise of automating these tedious workflows, but their development is hindered by the scarcity of real-world training data. While simulation offers a scalable alternative for producing demonstrations, current methods primarily target relatively short-horizon tasks with loosely structured interactions, failing to meet the strict procedural constraints and fine-grained manipulation demands of chemical experiments. To bridge this gap, we introduce \textbf{RoboChemGym}, a framework that autonomously generates high-fidelity manipulation demonstrations aligned with real-world experiment protocols, featuring a \textit{self-improving task synthesis} mechanism to iteratively refine task execution and scene configurations, enabling the reliable generation of expert trajectories for complex, multi-object protocols exceeding 10 interaction steps. Furthermore, we introduce a hierarchical benchmark that systematically assesses performance across varying granularities, spanning from atomic operations to full-cycle experimental workflows. RoboChemGym sets a scalable paradigm for the automated data synthesis and capability evaluation of embodied agents in intricate chemical tasks, serving as a critical stepping stone toward fully intelligent laboratories.

cs.RO↗

arXiv2026

RoboBridge: A Self-Evolving Embodied Agent Framework for Sim-to-Real Transfer

A key challenge in bringing embodied intelligence into the real world is transferring capabilities from simulation to reality and enabling agents to continually adapt after deployment. End-to-end vision-language-action policies provide strong manipulation capabilities, but their transfer to physical environments typically relies on calibrating simulated visual and dynamical conditions, collecting additional target-domain demonstrations, and optimizing the policy through further training. Tool-using embodied agents offer flexible task orchestration, yet existing systems primarily emphasize task execution and experience reuse within a given environment, with limited support for transferring procedural knowledge and continuously adapting it across simulation and reality. We propose RoboBridge, a framework that treats sim-to-real transfer as the continued adaptation of executable task skills. The agent represents task knowledge as procedures connecting task intent, observations, tool operations, and outcome verification. Interaction feedback is used to generate candidate skill revisions, which are evaluated before being persisted or rejected. A pretrained vision-language-action policy is exposed as a reusable action tool and enhanced with inference-time guidance, enabling fine-grained execution without retraining the underlying policy. RoboBridge grounds transferable skills in task semantics and interaction interfaces shared across simulation and reality. This representation preserves reusable task structure while allowing environment-dependent operations to be selectively revised through real-world execution feedback. We evaluate the framework on LIBERO-PRO and corresponding physical tasks, studying both skill evolution and post-transfer adaptation. Our framework provides a route from one-shot policy deployment to continual procedural learning across environments.

cs.RO↗

arXiv2026

EWAM: Emergent Depth-Wise Specialization in a Unified Embodied Model -- From Semantic Understanding through Visual Foresight to Action

Vision-language-action (VLA) policies emphasize semantic understanding, whereas world-action models (WAMs) learn predictive representations of environment dynamics. Systems that expose a policy to both sources often still concentrate action computation on a single expert. We present EWAM, an action-centric unified embodied model whose asymmetric joint attention lets action tokens read semantic, current-visual, predicted-future, and action information at every layer while the perceptual experts retain their distinct roles. Without layer-wise supervision, EWAM develops an emergent depth-wise specialization: action queries attend mainly to vision-language features in shallow layers, to predicted future frames in intermediate layers, and to action tokens themselves in deep layers. This handoff replicates across tasks and is stable across denoising steps. Checkpoint tracking and causal interventions show that it is learned and that action generation depends on it. EWAM is pretrained in two separate regimes, one on cross-embodiment robot trajectories and one on human egocentric video. In simulation and real-robot experiments, it surpasses existing VLA, WAM, and hybrid baselines. Human egocentric data improve both cross-embodiment transfer and real-robot robustness, and subtask-phase supervision improves long-horizon completion. Together, these results suggest that unified embodied learning can induce an ordered internal progression from semantic understanding, through visual foresight, to action formation.

cs.RO↗

arXiv2026

Low Mach number limit around the planar diffusion wave for 3D Navier-Stokes-Fourier equations

We investigate the low Mach number limit of the three-dimensional full compressible Navier-Stokes-Fourier (NSF) equations on $\mathbb{R}\times\mathbb{T}^2$ for two different classes of initial data, corresponding respectively to a well-prepared regime and an ill-prepared regime. The density and temperature are allowed to approach different asymptotic states at infinity. For the well-prepared regime, the solutions of compressible NSF equations converge to a planar diffusion wave solution globally in time as the Mach number tends to zero, where the difference between the states at the far fields is small independently of the Mach number and the non-zero modes of the initial perturbations are exponentially small. Moreover, the optimal time-decay rate can be obtained. It can be viewed as the first global-in-time result on the low Mach number limit of three-dimensional NSF equations with large temperature variations. The corresponding local-in-time result is proved while the difference between the states at the far fields can be arbitrarily large for the ill-prepared data.

math.AP↗

arXiv2026

Shock profiles for the Navier--Stokes--Fourier--Poisson system under the Boltzmann relation

The one-dimensional compressible Navier--Stokes--Fourier--Poisson system under the Boltzmann relation is studied for the existence of small-amplitude viscous shock profiles connecting the Rankine--Hugoniot states of the quasineutral Euler system with effective pressure $P_{\mathrm{eff}}=ρ(θ+1)$. This profile is unique up to translation, of Lax type, and orbitally stable in time without a zero-mass assumption on the perturbation. The proof is based on a center-manifold reduction of the traveling-wave ODE and on the method of $a$-contraction with shifts, together with a relative-entropy functional augmented by electric energy and a sharp estimate of the viscous flux associated with $P_{\mathrm{eff}}$.

math.AP↗

arXiv2026

HandAnthro: Automated Hand Anthropometry from a Single Image

Hand anthropometry supports protective-glove design, but existing measurement methods often require trained operators, specialized hardware, or manual landmarking. We present HandAnthro, which estimates 44 projected hand dimensions from a smartphone photograph of a palm-up hand on US letter-size paper. The pipeline reconstructs wrist-occluded paper boundaries for rectification, whitens non-hand pixels, and refines 41 anthropometry-specific landmarks from a fine-tuned You Only Look Once (YOLO) pose model using image-specific geometry and contours. Controlled evaluation comprised 720 captures from 45 held-out participants, each contributing 16 images across two smartphones, two backgrounds, two angles, and two nominal illumination settings. HandAnthro produced complete outputs for 704 captures (97.8%); among these, mean absolute error (MAE) was 3.80 mm per dimension against two trained operators' caliper measurements. Regional MAEs were 2.48 mm for non-thumb fingers, 6.04 mm for thumbs, and 6.17 mm for palm and wrist. In a researcher-assisted mobile-app pilot, automated batch processing returned all 44 dimensions for 260 of 268 retained, researcher-screened firefighter images (97.0%). A descriptive, unpaired comparison with an independent national firefighter reference yielded a mean absolute difference of 2.40 mm across 28 sex-by-dimension group-mean contrasts. These results characterize controlled measurement performance and researcher-assisted field feasibility for future distributed hand-anthropometry studies.

cs.CV↗

arXiv2026

Evolution of Magnetism in Ce$_4$Ge$_7$ under Magnetic Field and Pressure

We report the magnetic and transport properties of single-crystalline Ce$_4$Ge$_7$, which crystallizes in a non-centrosymmetric orthorhombic structure (space group C222$_1$). It undergoes an antiferromagnetic transition at $T_N$ = 7.3 K. When a magnetic field is applied along the $b$-axis, a metamagnetic transition occurs at 1.3 T (at 2 K). Correspondingly, this transition gives rise to an anomalous Hall effect, which is dominated by the intrinsic Karplus-Luttinger mechanism. Under hydrostatic pressure, $T_N$ of Ce$_4$Ge$_7$ initially increases slightly and is then gradually suppressed. For pressures above $P_c$ = 10.2 GPa, no magnetic order is observed. The divergence of the $A$ coefficient, the maximum of the residual resistivity, and the non-Fermi liquid behavior around $P_c$ indicate the possible existence of an antiferromagnetic quantum critical point in Ce$_4$Ge$_7$.

cond-mat.str-el↗

arXiv2026

DuoOPD: Learning from Joint Teacher-Student Outcomes for Multi-Task On-Policy Distillation

On-policy distillation (OPD) trains a student on its own responses with token-level feedback from a stronger teacher, yet the teacher can fail on questions the student already answers correctly, and how often each model succeeds varies across tasks. OPD ignores these outcomes and, on average, pushes down even the student's correct responses; gating feedback by student correctness fixes the direction but uses the teacher in the same way whether or not it succeeded. We introduce DuoOPD, in which the student's outcome sets the direction of feedback and the joint teacher-student outcome decides how the teacher supports it: when only the teacher succeeds, its verified answer becomes context for scoring the student's failed response, and when only the student succeeds, a weight shared within the task reinforces the whole response. A single rule covers all four outcome combinations without task-specific settings. Across Qwen3 and Llama, DuoOPD outperforms all five baselines in mean macro accuracy, improving over OPD by 2.58 and 5.98 percentage points, and it also leads on two further task mixtures spanning scientific calculation, instruction following, and code generation. Ablations show that outcome-based direction alone stays near the gated baseline, while the joint-outcome designs supply most of the gain.

cs.LG↗

arXiv2026

Dependency- and Layer-Aware Microservice Workflow Offloading and Service Image Caching for Edge Environments

The microservice architecture has been applied broadly in many mainstream computing environments. As one of the most prevalent environments, edge computing also widely employs microservices to handle diverse requests and tasks. In practical scenarios, microservices usually constitute workflows that are built based on service dependencies to execute tasks. This offers an opportunity to explore the offloading technology to better harness resources and accelerate task execution. Unfortunately, this problem has not been paid attention to by existing research, and thus some valuable resources (e.g., microservice image cache) remain obscure. To fill this gap, we innovatively study the problem of joint optimization for microservice workflow offloading and service image caching in edge. This problem is challenging due to several issues such as the complexity of its solution space, intricate relationships among shared image layers, long-range dependencies between workflow tasks, and the absence of a real-world collection of service images. To address these issues, this paper proposes an innovative dependency- and layer-aware workflow offloading and image caching framework for microservices. We collected a real-world collection of microservice image layer data and published both this collection and experimental codes on GitHub. Extensive results demonstrate that our framework achieves superior performances, for example, a 22. 38\% reduction in average task completion time compared to baselines and significantly-increased image hit rates.

cs.DC↗