arXiv subjects

Fan Feng

Publications and source records attributed to Fan Feng.

At least 19 recordsLinked to original sources

arXiv2026

CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?

End-to-end automation of realistic healthcare operations stresses three capabilities underrepresented in current benchmarks: policy density, decisions must be grounded in a large library of medical, insurance, and operational rules; Multi-role composition: a single task requires the agent to play multiple roles with handoffs; and multilateral interaction: intermediate workflow steps are multi-turn dialogs, such as peer-to-peer review and patient outreach. We introduce $χ$-Bench, a benchmark of long-horizon healthcare workflows across three domains: provider prior authorization, payer utilization management, and care management. Each task hands the agent a clinical case in a high-fidelity simulator of 20 healthcare apps exposed via 87 MCP tools, which it must drive to a terminal status through tool calls and writing the role's artifacts, guided by a 1,290+ document managed-care operations handbook skill. Across 30 agent harness/models configurations, the best agent resolves only 28.0% of tasks, no agent clears 20% on strict pass^3, and executing all tasks in a single session slumps the performance to 3.8%. These results raise the hypothesis that similar gaps are likely to surface in other policy-dense, role-composed, irreversible enterprise domains.

cs.CL↗

arXiv2026

Video2World: Benchmarking Coding Agents for Interactive World Modeling from Embodied Videos

Building interactive simulators from real-world observations is a promising way to scale embodied data, but current pipelines still rely heavily on manual environment construction and calibration. We study whether frontier foundation models and coding agents can automate this process end to end. We formulate \emph{autonomous video-to-simulation} as a software engineering task in which an agent observes an embodied video, constructs the corresponding simulated environment and robot behavior, and iteratively refines the result through execution feedback. To evaluate this capability, we introduce \textbf{Video2World}, a benchmark comprising 222 reconstruction instances derived from 189 robot and human demonstration videos. Video2World measures reconstructed worlds along geometric fidelity, dynamic fidelity, and functional correctness, capturing spatial perception, physical reasoning, and executable interaction. Evaluating 9 frontier coding-agent systems reveals a sharp improvement in Task success beginning with Claude Opus 5, rising from below 5\% to over 15\%, while substantial gaps to human-assisted reconstruction remain. We further find that worlds that look better could work worse: better visual fidelity does not always lead to higher task success. This echoes the broader gap between perceptual realism and factual correctness observed in generative models.

cs.CV↗

arXiv2026

Kinetic Reduction and Hydrodynamics under Time-Dependent Helical Symmetry

Translational and helical symmetries provide idealized descriptions of many flows encountered in nature and engineering. While translationally symmetric flows have been extensively studied within kinetic and continuum frameworks, the corresponding connection between microscopic dynamics and continuum behaviour for helical flows is less developed. Here, we investigate this connection for a dilute gas with time-dependent helical symmetry. Using the objective molecular dynamics framework, we derive a reduced Boltzmann equation that retains radial variations while preserving the classical collision operator under helical symmetry. Formal hydrodynamic expansions then yield the corresponding Euler and Navier-Stokes-Fourier equations. The resulting models describe how cylindrical geometry and axial deformation influence transport, including viscous transport and heat conduction, and how deformation generates unequal stresses along and across the axis in competition with collisional relaxation. Comparisons with direct simulation Monte Carlo solutions examine both spatially uniform and radially varying flows. In the cases studied, the first-order fluid corrections capture the leading stress anisotropy and improve the prediction of radial temperature variations, while larger Knudsen numbers expose limitations of the approximation.

physics.flu-dyn↗

arXiv2026

TAFFY: A Task-Adaptive Tabular Foundation Model with In-Context Diversity

Recent progress in tabular foundation models suggests that training on synthetic tasks can substantially improve in-context learning capabilities, with overall performance largely depending on how well models can infer task-specific predictive relationships from the available context during inference. In this paper, we introduce TAFFY, a tabular foundation model with an In-Context Diversity Prior and a Task-Conditioned Looped Transformer that strengthen this ability. Specifically, to construct each synthetic pretraining context, the In-Context Diversity Prior samples from multiple related environments derived via controlled interventions and distribution shifts on a shared causal process. This in-context diversity encourages the model to learn a more comprehensive and task-specific representation. Moreover, the Task-Conditioned Looped Transformer iteratively and selectively applies a shared group of Transformer blocks to refine contextual representations, with a task-conditioned gate modulating the final hidden-state update. This enables task-adaptive iterative refinement. Together, these components encourage the model to identify predictive relationships from contextual contrasts during pretraining and dynamically modulate context integration for each task. Across six classification and five regression benchmark datasets, TAFFY attains the lowest average rank.

cs.LG↗

arXiv2026

Causally Debiased Latent Action Model for Embodied Action-Conditioned World Models

Action-conditioned world models (ACWMs) aim to simulate future observations conditioned on embodied actions. However, learning controllable ACWMs requires large-scale action-labeled data, which remains costly to collect. Latent action models (LAMs) mitigate this bottleneck by inferring latent actions from videos without executable action labels, but existing LAMs are typically trained with reconstruction-only objectives and therefore entangle action-relevant dynamics with action-irrelevant visual confounders. With such objectives, LAM favors visual context over action dynamics, therefore downstream ACWMs exhibit residual motion under zero actions and fail to reproduce supplied dynamics. We propose CD-LAM, a causally debiased framework for LAM-based ACWMs. CD-LAM introduces three debiasing objectives: embodiment-centric reconstruction, action-centric contrastive learning, and latent space calibration, which together encourage embodiment-focused, action-aware, and well-calibrated representations. CD-LAM reduces action-following error by up to 42% and 35% when conditioned on latent and robot actions, while improving fidelity, and matching DreamDojo reference with over 12$\times$ fewer adaptation updates. CD-LAM also adapts LTX-2.3-22B into an ACWM with competitive quality using only 3.2% of DreamDojo-14B's sample exposures. On X-VLA, pretraining with CD-LAM instead of baseline LAM consistently improves success rates by up to 5 percent point across LIBERO, LIBERO-Plus, and RoboTwin C2R.

cs.CV↗

arXiv2026

TimeBraid: Unifying Time Series and Language for Understanding and Forecasting

We present TimeBraid, a series of unified time-series and language models that align pretrained language models and pretrained time-series foundation models through interleaved global residual attention layers. Each model inherits knowledge, instruction following, and reasoning from one side, continuous-signal perception and zero-shot forecasting from the other, and fuses the two in a shared representation space where both modalities are understood and generated. We study the design choices that make such unified modeling work: where to align the two representation spaces, how to ground language in temporal structure, how to balance understanding with generation, and how to keep joint optimization stable. The resulting recipe combines a unified prompting scheme for diverse time-series and text tasks, stabilized joint training, and supervision from 2.2M curated series--text pairs and 4.9M instruction-tuning samples. Across benchmarks spanning time-series perception, understanding, reasoning, and both context-aided and unimodal forecasting, TimeBraid remains competitive with far larger general-purpose models and task-specific counterparts.

cs.CL↗

arXiv2026

In-Context Robot Learning with VLM Agents

Enabling robots to adapt to unfamiliar environments as readily as humans remains a moonshot goal of embodied AI. No finite collection of demonstrations can cover every task and situation a robot will encounter, making the ability to learn from context at deployment essential for generalization. Such in-context learning (ICL), however, remains largely beyond the reach of existing robotic policies. The broad agentic capabilities of commercial vision-language models (VLMs), such as GPT-6 Astra, raise a compelling question: can these models learn from demonstrations, examples, and interaction feedback, then translate that information into executable and verifiable robot behavior from a new initial state without gradient updates or persistent changes to task-specific parameters? We introduce GPT-Policy, a general-agent framework for in-context robot learning. GPT-Policy integrates a context compiler that preserves task-relevant visual transitions, a VLM that proposes robot-tool actions, and a constrained controller that verifies and executes each action and reports its outcome. We evaluate its reliability and limitations through task success and efficiency metrics, matched comparisons across models, and controlled context ablations. In real-robot trials, human video demonstrations improve task completion even without robot action labels, while aligned action references yield further gains on contact-sensitive tasks. These findings position GPT-Policy as a step toward robot adaptation through in-context learning, providing an empirical foundation for translating the general-purpose capabilities of VLMs into physical behavior and clarifying the challenges that must be overcome for reliable deployment.

cs.CV↗

arXiv2026

Emergent chiral symmetry breaking in moiré domain wall networks redirects topological boundary states in bilayer graphene

Lattice-mismatched bilayer graphene self-organizes into a moiré network of one-dimensional domain walls that conduct electrons with low dissipation, attractive for low-power electronics. We show these channels are not always straight: as the lattice relaxes to minimize strain, the network can spontaneously curve into chiral morphology. Atomistic simulations map a phase diagram in which the strain-flexibility balance selects one of three stable domain wall morphologies-straight, mono-chiral, or dual-chiral. Electronic structure calculations show that this morphology controls where low-energy electrons accumulate: straight channels concentrate states at the domain-wall-connecting nodes, while chiral channels shift that weight onto the domain walls themselves. This network geometric switch lets the same moiré material support either localized electronic hot spots or directional conducting channels-two strategies for guiding electrons in low-power graphene devices.

cond-mat.mes-hall↗

arXiv2026

Hamiltonian Two-Way Coupling of Nonlinear Waves and 3D Flows

Simulating large-scale free-surface water by coupling a localized 3D fluid solver to a cheaper 2D surface model has long faced a mismatch in wave dynamics: efficient 2D wave models used in graphics are typically either linear or non-dispersive. These models are fast, simple, and accurate for calm, small-amplitude seas, but coupling them with strongly nonlinear 3D solvers produces visible reflections and artifacts at the 2D--3D interface. We address this problem by introducing a nonlinear and dispersive 2D wave model based on the canonical Zakharov formulation. Its Hamiltonian structure, in which the surface elevation and surface potential form a canonical pair ($η$, $ψ$) governed by the wave energy, enables a canonically consistent two-way coupling scheme, allowing information to pass smoothly across the 2D--3D interface. Our 2D solver reduces mean wave-height error by 1.7--5$\times$ over SWE, BEM, and Airy baselines while running more than $10^3\times$ faster than BEM; it achieves greater nonlinear accuracy and coupling fidelity than SWE and Airy, with minor losses in speed and stability. Coupling it with a 3D Navier--Stokes solver yields a full system that suppresses visible seam artifacts across a range of experiments, including dispersion-matching and Kelvin-wake tests, and runs over 4$\times$ faster than a pure GPU NB-FLIP simulation on the same domain.

cs.GR↗

arXiv2026

Lattice parameter engineering for reversible martensitic materials using simplified cofactor conditions

Cofactor conditions (CCs) provide a geometric criterion for achieving highly compatible austenite/twinned martensite interfaces and have served as a guiding principle for the design of highly reversible martensitic materials. However, their standard ten- sorial form obscures the direct connection between crystallographic lattice parameters and compatibility. In this work, we reformulate the CCs in the eigenspace of the trans- formation stretch tensor. The resulting expressions reduce the compatibility problem to algebraic relations involving the principal stretches and the crystallographic twin axes. Simplified conditions are derived for both Type I/II and Compound 1/2 twins. We find that, for Type I/II twins, once the middle-eigenvalue condition λ2 = 1 is imposed, the remaining conditions reduce to sets of paired curves in the (λ1, λ3) eigenvalue space, while for Compound twins, the conditions lead to admissible domains, revealing a broader compatibility design window than the curve-like conditions associated with Type I/II twins. We further specialize the formulation to cubic to tetragonal, cubic to orthorhombic, and cubic to monoclinic I/II transformations, thereby establishing explicit links between lattice parameters, eigenvalues, and compatibility. The analysis suggests two main conclusions: (i) cubic to monoclinic transformations provide additional design freedom through the monoclinic angle, offering a promising route to simultaneously achieve low hysteresis and large transformation strain; and (ii) Compound twins provide an important design pathway because they occupy finite admissible regions rather than isolated solution curves.

cond-mat.mtrl-sci↗

arXiv2026

Learning Task-Sufficient World Models by Synergizing Agentic Exploration and Structured Modeling

Learning and planning in imagination using world models provides an effective paradigm for training agents for decision-making. However, existing approaches often rely on high-dimensional latent spaces or generic visual embeddings that retain many factors irrelevant to control, limiting efficiency and generalization across tasks. To this end, we study how agents can learn world models with representations that are task-specific, minimal, and sufficient for decision-making. We achieve this via a closed-loop synergy between the agent and the world model, in which structured world-model learning distills task-sufficient representations from informative interaction data. On the agent side, agents actively probe the environment to collect informative trajectories that expose task-relevant latent factors, guided by an adaptive curriculum. On the world-model side, we learn structured representations over observations to distill compact, task-sufficient latent states from the collected interaction data. This synergy enables the empirical recovery of task-sufficient latent representations that capture all control-relevant factors. Leveraging these representations, the resulting policies achieve improved sample efficiency and generalization, including generalization across skills, object-skill compositions, and previously unseen tasks on standard continuous-control and robotic-manipulation benchmarks.

cs.LG↗

arXiv2026

ReFPO: Reflow Regularization for Flow Matching Policy Gradients

We present Reflow-regularized Flow Matching Policy Gradients (ReFPO), a simple online RL method that adds explicit Reflow regularization to FPO for efficient flow-based control. We uncover a key structural property: the gradient updates in Flow Matching Policy Gradients (FPO) can be interpreted as an implicit advantage-weighted Reflow process, providing a new geometric perspective on flow-based policy gradients. Building on this insight, ReFPO introduces an explicit geometric regularizer that can be implemented with a single line of code change without incurring additional computational overhead or auxiliary distillation stages. By synergizing advantage-guided updates with path rectification, our method reduces CFM proxy-ratio spikes, stabilizes PPO-style training, and enables high-fidelity one-step inference that often matches or exceeds multi-step performance. We experimentally demonstrate that ReFPO improves average performance and discretization robustness across GridWorld, MuJoCo Playground, and high-dimensional Humanoid Control tasks, providing a scalable and stable approach for generative policies in complex physical simulations.

cs.RO↗

arXiv2026

MimicIK: Real-Time Generative Inverse Kinematics from Teleoperation with FK Consistency

Inverse kinematics (IK) remains a critical bottleneck for real-time robot manipulation. Classical numerical solvers achieve high geometric precision but often suffer from discontinuous branch switching and unstable behavior near kinematic singularities during closed-loop deployment. Meanwhile, learned IK approaches frequently struggle to balance spatial accuracy, motion smoothness, and real-time efficiency, particularly when trained on noisy human teleoperation data. We present \textbf{MimicIK}, a real-time generative inverse kinematics framework that learns smooth and robust joint-space motion priors from teleoperation demonstrations through conditional flow matching. Given the current joint configuration and a target end-effector pose, MimicIK predicts continuous delta-joint commands using an efficient two-step iterative refinement process based on a Minimal Iterative Policy (MIP) backbone. To enforce physical consistency, we further introduce an FK consistency loss, a differentiable forward-kinematics regularization that penalizes task-space deviations from the target pose during training. We evaluate MimicIK on a real-world 6-DOF robot dataset containing 8,848 teleoperation demonstrations. MimicIK achieves a mean position error of 4.65 mm, a 10 mm success rate of 92.01\%, and a trajectory spike rate of only 7.99\%. Compared with a UNet diffusion baseline, our method improves both spatial accuracy and motion smoothness while reducing inference latency from 21.66 ms to 6.74 ms. Furthermore, unlike deterministic MLP baselines that catastrophically diverge under out-of-distribution deployment, MimicIK remains stable near singular configurations and enables robust 20 Hz real-time control on deployment hardware.

cs.RO↗

arXiv2026

Acting While Understanding: Asynchronous Semantic-Action Decoupling for Real-Time Vision-Language-Action Models

Vision-Language-Action models (VLAs) have demonstrated strong task understanding and generalization in robotic manipulation, yet the high computational cost of full-model inference limits their deployment in low-latency, high-frequency closed-loop control. We propose an asynchronous semantic-action decoupling framework that separates semantic understanding from action generation along the internal semantic-action interface of existing VLAs, without redesigning the vision-language backbone or introducing an external planner. A low-frequency understanding module asynchronously updates reusable semantic conditions, while a high-frequency action module continuously outputs control actions without repeatedly invoking the full model. To mitigate the temporal mismatch between stale semantics and the current execution state, we further introduce historical action conditioning and time-misalignment training, which provide short-horizon execution context and improve feedback control robustness under stale semantic conditions. Experiments on LIBERO with $π_{0.5}$ and UniVLA, together with real-robot deployment using UniVLA, show that the proposed framework achieves up to 35.6 Hz server-side action-module inference throughput and offers a low-intrusion path to high-frequency closed-loop control without running full VLA inference at control rate.

cs.RO↗

arXiv2026

Elastic Queries Reinforcement Learning: Self-Aware Policy Execution for VLA Models

Vision-language-action (VLA) models are powerful action generators for robot manipulation, but they are typically executed with fixed inference and replanning schedules. This rigidity ignores the uneven difficulty of robot control: contact-rich or uncertain states may need more computation and fresher feedback, while easier states can often be handled with fewer inference steps and longer open-loop execution. We propose Elastic Queries Reinforcement Learning (EQRL), a framework that makes each VLA policy query elastic. A lightweight latent-schedule adaptor jointly selects the latent input, denoising budget, and action chunk length, without fine-tuning the underlying VLA model. To make scheduling difficulty-aware, EQRL trains a critic over the joint latent-schedule action and derives a state difficulty signal from critic ensemble disagreement. This signal guides compute toward difficult states, while a learned residual allows task-driven correction. We formulate variable chunk execution as query-level macro-action RL with chunk-dependent discounting and an amortized number-of-function-evaluations (NFE) budget. Across simulation and real-robot manipulation, EQRL reduces amortized inference cost while preserving or improving task success.

cs.RO↗

arXiv2026

IB-HFN: Information Bottleneck-Driven SAR-Optical Fusion Network for High-Fidelity Cloud Removal

Synthetic aperture radar (SAR)-assisted optical cloud removal aims to recover surface information obscured by clouds in optical remote sensing images by exploiting complementary SAR observations. Existing multimodal fusion methods typically rely on direct spatial concatenation and pixel-wise supervision, which can propagate SAR speckle noise into optical reconstruction and lead to over-smoothed results. To address these limitations, we propose an Information Bottleneck-driven High-Fidelity Network (IB-HFN) for SAR-assisted optical cloud removal. IB-HFN employs a dual-stream backbone to preserve modality-specific representations before deep semantic fusion, thereby mitigating premature cross-modal contamination. At the fusion stage, we introduce a Spatial Information Bottleneck Fusion module that compresses SAR features through a channel-wise variational information bottleneck to suppress unstructured speckle noise. In parallel, a local-global gating mechanism predicts clear-sky regions and routes reliable optical details through a Dirac-initialized skip connection, decoupling noise suppression from texture preservation. We further develop a joint optimization strategy that integrates feature-level bottleneck regularization with image-level constraints on reconstruction accuracy, structural consistency, spectral fidelity, and contrastive sharpness. A dynamic weighting schedule balances these objectives to stabilize training and reduce hazy artifacts. Experiments on the SEN12MS-CR dataset under challenging spatio-temporal splits demonstrate that IB-HFN achieves superior structural preservation and spectral fidelity over existing methods.

cs.CV↗

arXiv2026

Learning Object Manipulation from Scratch via Contrastive Interaction

Contrastive Reinforcement Learning (CRL) has seen recent success in a wide variety of goal-conditioned robotics tasks by learning structured representations of the dynamics. However, despite its success in locomotion and simpler control domains, CRL often struggles in interaction-rich manipulation. We argue that a key source of this difficulty is object-centric interaction, such as contact or grasping, that induces distinct changes in the underlying dynamic modes. In this work, we formulate manipulation dynamics as a piecewise-smooth Markov process and show that interaction-induced mode changes create piecewise nonlinear reachability structures that are difficult for standard CRL energy functions to represent and plan over. Based on this analysis, we introduce Interaction-weighted Resampling (IWR). IWR performs interaction-aware resampling around phases before, during, and after interactions, encouraging the learned representation to preserve the mode boundaries that determine future reachability to capture multi-modal and piecewise nonlinear reachability. Across interaction-centric environments, including 2D dynamic control, robotic manipulation, and robot air hockey, IWR improves both sample efficiency and overall performance over prior CRL methods, with 19.8% average improvement in simulation. Finally, using a sim-to-real pipeline with policies trained by IWR, we demonstrate the first real-world goal-conditioned robot air hockey agent capable of hitting goals, improving success from 25% to 60%. Project Page: IWR-arxiv.github.io.

cs.RO↗

arXiv2026

On Information Self-Locking in Reinforcement Learning for Active Reasoning of LLM agents

Reinforcement learning (RL) has become a de facto paradigm for building LLM-based agents that act, interact, and reason over extended task horizons. However, in active reasoning where agents must elicit new observations through interaction with the environment to solve the task, we find that outcome-based RL can induce a systematic failure mode which we call information self-locking (SeL): agents fail both to elicit informative feedback and to internalize obtained evidence. To understand the issue, we trace agentic behaviors into two coupled capabilities: Action Selection (AS), which determines observation streams, and Belief Tracking (BT), which updates the agent's internal task understanding. Theoretical and empirical analyses reveal a bidirectional bottleneck that leads to SeL: weak BT obscures the credit of informative actions, while weak AS deprives BT of useful evidence. This coupling weakens the learning signal for both capabilities and leads to SeL. To mitigate this issue, we propose AREW, a simple yet effective Advantage Reweighting method that uses easy-to-obtain directional critiques to reallocate credit within trajectories. Extensive experiments across 9 agentic tasks of varying complexity show that AREW significantly mitigates SeL, yielding up to 60-point gains in final performance. Code is available at https://github.com/unimpor/T3.

cs.AI↗