arXiv subjects

Haoran Lin

Publications and source records attributed to Haoran Lin.

At least 19 recordsLinked to original sources

arXiv2026

TONAV: Task-Oriented Navigation and Action-Velocity Chunk Learning for Articulated Object Quadrupedal Mobile Manipulation

Quadruped mobile manipulation requires two tightly coupled capabilities: reaching manipulation-ready configurations and maintaining stable contact throughout articulated-object interaction. However, existing methods often terminate navigation near the target, leaving a gap between reachability and manipulation readiness, while tracking lag, motion jitter, and contact instability limit continuous interaction. To address these challenges, we present TONAV, a unified framework integrating task-oriented navigation with action-velocity chunk learning. First, we introduce a position-velocity-coupled teleoperation framework that explicitly captures motion dynamics to improve master-follower consistency and collect smooth, temporally consistent demonstrations. Next, task-oriented navigation leverages vision-language reasoning to decompose high-level instructions into executable subgoals and adaptively refine the robot base toward a manipulation-ready configuration. Finally, action-velocity chunk learning jointly models joint positions and their temporal transitions under velocity supervision, enabling smooth and stable sustained-contact manipulation. Real-world experiments across diverse articulated-object tasks demonstrate that TONAV achieves higher success rates in both task-oriented navigation and complete mobile manipulation, mitigating the navigation-manipulation gap and improving continuous-contact interaction. The project page is at https://haochen611.github.io/TONAV.

cs.RO↗

arXiv2026

Signatures of a light-induced exciton condensate exhibiting BEC-BCS crossover

Exciton condensates provide a platform to study quasiparticle pairing, Bose-Einstein condensation-Bardeen-Cooper-Schrieffer (BEC-BCS) crossover, and excitonic topological phenomena. Achieving a nonequilibrium exciton condensate allows the ultimate tunability of these emergent phenomena. Yet, evidence of a light-induced, nonequilibrium exciton condensate and its BEC-BCS crossover remains elusive. Here, we use time- and angle-resolved photoemission spectroscopy to demonstrate signatures of a non-equilibrium exciton condensate and its BEC-BCS crossover in monolayer MnBi2Te4. Following optical excitation, a distinctive hole-like dispersion representing excitons emerges and persists for >20 ps. Strikingly, energy-domain sharpening in the valence band occurs 2 ps after time zero and exhibits a sharp onset at a threshold pump fluence of 0.84 mJ/cm2. The delayed and strongly nonlinear response is difficult to reconcile with transient field effects or conventional carrier-induced band shifts but is consistent with a model of exciton condensation governed by a Berezinskii-Kosterlitz-Thouless transition. The estimated threshold exciton density agrees quantitatively with the Nelson-Kosterlitz critical density. At higher fluences, the exciton feature develops a camel-back-shaped dispersion, consistent with the BEC-BCS crossover in the condensate framework. Our work establishes ultrathin MnBi2Te4 as a model system for studying nonequilibrium exciton condensates with a connection to superconductivity and exciton-driven topological phases.

cond-mat.mes-hall↗

arXiv2026

Safe and Adaptive Cloud Healing: Verifying LLM-Generated Recovery Plans with a Neural-Symbolic World Model

As the scale and complexity of cloud-based AI systems continue to escalate, ensuring service reliability through rapid fault detection and adaptive recovery has become a critical challenge. While existing approaches integrate Large Language Models (LLMs) for semantic understanding and Deep Reinforcement Learning (DRL) for policy optimization, they often rely on sequential, loosely coupled architectures that underutilize the generative and reasoning capabilities of LLMs. In this paper, we propose a paradigm shift with PASE, a Planning-Aware Semantic self-healing engine, a novel fault self-healing framework that reconceptualizes recovery as a neuro-symbolic program synthesis task. PASE employs an LLM as a core Plan Synthesis Engine to generate structured recovery plans from a library of semantic primitives. A Neural-Symbolic World Model verifies plan feasibility through simulation, while a Meta-Prompt Optimizer, trained via DRL, learns to generate optimal prompts that guide the LLM's planning process. This tight reason-plan-verify-adapt loop enables dynamic, context-aware recovery strategy generation beyond predefined action spaces. Experiments on a real-world cloud fault injection dataset demonstrate that PASE significantly outperforms state-of-the-art methods, reducing average system recovery time by over 40% and improving fault detection accuracy in unknown fault scenarios. Our framework advances autonomous system management by unifying LLM-based reasoning with model-assisted verification and meta-learned guidance.

cs.AI↗

arXiv2026

DexPIE: Stable Dexterous Policy Improvement from Real-World Experience

Dexterous manipulation presents substantial challenges for imitation learning due to its high-dimensional action space and complex contact-rich dynamics. Policies trained purely from demonstrations often suffer from compounding errors during deployment and require large amounts of expert data to achieve reliable performance. To move beyond the limitations of demonstration data, in this work, we propose DexPIE, a post-training framework for dexterous policy improvement from experience collected through real-world deployment. First, DexPIE enables effective exploration coverage through a dexterous-hand-adapted intervention system and multi-stage DAgger-style data collection across initial and intermediate task stages, providing reliable supervision for accurate policy evaluation. To reduce temporal noise between post-training rollouts and demonstration data, we introduce asynchronous inference in the relative action space, which better aligns rollout data with demonstrated behavior and allows the critic to learn a value function induced by a more consistent underlying policy. Finally, DexPIE improves the policy through conditioning on a continuous optimality indicator, allowing the policy to leverage the quality of data in a more fine-grained manner. Across three challenging real-world dexterous manipulation tasks, DexPIE achieves a 37% improvement in success rate over the demonstration-based reference policy, outperforming all baseline methods and demonstrating stronger robustness. The source code and dataset will be made publicly available.

cs.RO↗

arXiv2026

WASD: Locating Critical Neurons as Sufficient Conditions for Explaining and Controlling LLM Behavior

Precise behavioral control of large language models (LLMs) is critical for complex applications. However, existing methods often incur high training costs, lack natural language controllability, or compromise semantic coherence. To bridge this gap, we propose WASD (unWeaving Actionable Sufficient Directives), a novel framework that explains model behavior by identifying sufficient neural conditions for token generation. Our method represents candidate conditions as neuron-activation predicates and iteratively searches for a minimal set that guarantees the current output under input perturbations. Experiments on SST-2 and CounterFact with the Gemma-2-2B model demonstrate that our approach produces explanations that are more stable, accurate, and concise than conventional attribution graphs. Moreover, through a case study on controlling cross-lingual output generation, we validated the practical effectiveness of WASD in controlling model behavior.

cs.CL↗

arXiv2026

Millimeter-Scale, Atomically Controlled 2D Topological Insulators Revealed by Multimodal Spectroscopy

Quantum spin Hall insulators, or synonymously known as 2D topological insulators, are crucial 2D systems hosting topologically protected edge states. The working temperature of this topological quantum phase is dictated by the inverted bandgap. However, the previously identified large-gap 2D topological insulators are either extremely chemically unstable, or cannot be made with atomistic precision over macroscopic scales. Here, we establish two-quintuple-layer Bi2Te3 and MnBi2Te4/Bi2Te3 heterostructures as atomically controlled, millimeter-scale 2D topological insulators, enabled by precision layer-by-layer growth that yields a carpet-like morphology extending coherently over macroscopic distances. This carpet-like growth mode renders the films amenable to mechanical exfoliation and subsequent wet or dry transfer. Multimodal spectroscopies and microscopies reveal the integer-layer tuned electronic structure of (Bi2Te3)n with excellent agreement to theory. Photon-energy-dependent photoemission and time-resolved photoemission identify band inversion and band dynamics, respectively, while scanning tunneling spectroscopy resolves topological edge states, characteristic of the 2D topological insulator phase. Thickness- and photon-energy-dependent photoemission further validates MnBi2Te4/Bi2Te3 as a robust 2D topological insulator. The large inverted gaps of ~100 meV in (Bi2Te3)2 and ~150 meV in MnBi2Te4/Bi2Te3 suggest operation near ambient temperature. These results define a scalable materials platform for next-generation, low-loss quantum and energy-efficient devices.

cond-mat.mtrl-sci↗

arXiv2025

HAP: Hybrid Adaptive Parallelism for Efficient Mixture-of-Experts Inference

Current inference systems for Mixture-of-Experts (MoE) models primarily employ static parallelization strategies. However, these static approaches cannot consistently achieve optimal performance across different inference scenarios, as they lack the flexibility to adapt to varying computational requirements. In this work, we propose HAP (Hybrid Adaptive Parallelism), a novel method that dynamically selects hybrid parallel strategies to enhance MoE inference efficiency. The fundamental innovation of HAP lies in hierarchically decomposing MoE architectures into two distinct computational modules: the Attention module and the Expert module, each augmented with a specialized inference latency simulation model. This decomposition promotes the construction of a comprehensive search space for seeking model parallel strategies. By leveraging Integer Linear Programming (ILP), HAP could solve the optimal hybrid parallel configurations to maximize inference efficiency under varying computational constraints. Our experiments demonstrate that HAP consistently determines parallel configurations that achieve comparable or superior performance to the TP strategy prevalent in mainstream inference systems. Compared to the TP-based inference, HAP-based inference achieves speedups of 1.68x, 1.77x, and 1.57x on A100, A6000, and V100 GPU platforms, respectively. Furthermore, HAP showcases remarkable generalization capability, maintaining performance effectiveness across diverse MoE model configurations, including Mixtral and Qwen series models.

cs.DC↗

arXiv2025

UniFucGrasp: Human-Hand-Inspired Unified Functional Grasp Annotation Strategy and Dataset for Diverse Dexterous Hands

Dexterous grasp datasets are vital for embodied intelligence, but mostly emphasize grasp stability, ignoring functional grasps needed for tasks like opening bottle caps or holding cup handles. Most rely on bulky, costly, and hard-to-control high-DOF Shadow Hands. Inspired by the human hand's underactuated mechanism, we establish UniFucGrasp, a universal functional grasp annotation strategy and dataset for multiple dexterous hand types. Based on biomimicry, it maps natural human motions to diverse hand structures and uses geometry-based force closure to ensure functional, stable, human-like grasps. This method supports low-cost, efficient collection of diverse, high-quality functional grasps. Finally, we establish the first multi-hand functional grasp dataset and provide a synthesis model to validate its effectiveness. Experiments on the UFG dataset, IsaacSim, and complex robotic tasks show that our method improves functional manipulation accuracy and grasp stability, demonstrates improved adaptability across multiple robotic hands, helping to alleviate annotation cost and generalization challenges in dexterous grasping. The project page is at https://haochen611.github.io/UFG.

cs.RO↗

arXiv2025

Interaction-driven flat band and charge order in Fe5GeTe2

Flat electronic bands enable fascinating emergent phenomena such as superconductivity and charge orders. A prevailing approach to realizing flat bands is to engineer lattice geometric constraints in twisted or kagome-like materials. An alternative approach is to utilize purely electronic-interaction-driven flat bands, yet a fundamental challenge is that extreme flatness requires ultrastrong interaction strength, which often leads to incoherent states. Here we demonstrate the concurrent formation of an interaction-driven flat band at the Fermi level and a $\sqrt{3}\times\sqrt{3}\,R30^\circ$ charge order in a van der Waals magnet Fe5GeTe2 using high-resolution angle-resolved photoemission spectroscopy. This charge order is manifested by band folding within 30 meV below the Fermi level, with its nesting driven by flat bands. The presence of this flat band throughout the Brillouin zone and the logarithmic temperature dependence of its spectral weight suggest a Kondo-like, coherent Fermi liquid emerging from strong correlations. Our work establishes a paradigm where an interaction-driven flat band promotes large-scale electronic ordering.

cond-mat.str-el↗

arXiv2025

Spectroscopic evidence of intra-unit-cell charge redistribution in charge-neutral magnetic topological insulator Sb-doped MnBi6Te10

The magnetic topological insulator MnBi$_{6}$Te$_{10}$ has emerged as a promising candidate for realizing the quantum anomalous Hall effect (QAHE), owing to its ability to retain ferromagnetism through precise control of anti-site defects. The next important task for realizing the QAHE is to tune the chemical potential into the energy gap formed by the broken time-reversal symmetry. Here we reveal an intra-unit-cell charge redistribution even when the overall doping suggests a near-charge-neutral condition. By performing time- and angle-resolved photoemission spectroscopy (trARPES) on the optimally 18% Sb-doped MnBi$_{6}$Te$_{10}$, we observe transient surface photovoltage (SPV) effects on both the MnBi$_{2}$Te$_{4}$ and single-Bi$_{2}$Te$_{3}$ terminations. Furthermore, we observe a time-dependent splitting of the band structure indicating multiple SPV shifts with different magnitudes. This observation suggests that adjacent plateaus with nominally the same terminating layer exhibit a strong intra-unit-cell charge redistribution, resulting in spontaneous electrical polarization. This is consistent with static micro-ARPES measurements revealing significant doping deviations from the charge-neutral configuration. Our findings underscore the challenges of engineering the family of Mn-Bi-Te materials to realize QAHE purely through chemical doping. Achieving the desired topological quantum phase requires both a uniform carrier doping and a ferromagnetic ground state. Furthermore, the light-induced polarization within each unit cell of ferromagnetic Mn(Bi$_{0.82}$Sb$_{0.18}$)$_{6}$Te$_{10}$ may open new possibilities for optoelectronic and spintronics.

cond-mat.mes-hall↗

arXiv2025

A Topological Superconductor Tuned by Electronic Correlations

A topological superconductor, characterized by either a chiral order parameter or a chiral topological surface state in proximity to bulk superconductivity, is foundational to topological quantum computing. As in other topological phases of matter, electronic correlations can tune topological superconductivity via modifications of the low-energy Fermiology. Such tuning has not been realized so far. Here we uncover a unique topological superconducting phase in competition with electronic correlations in 10-unit-cell thick FeTe$_{x}$Se$_{1-x}$ films grown on SrTiO$_{3}$ substrates. When the Te content $x$ exceeds $0.7$, we observe a rapid increase of the effective mass for the Fe $d_{xy}$ band, with the emergence of a superconducting topological surface state confirmed by high-resolution angle-resolved photoemission spectroscopy; however, near the FeTe limit, the system enters an incoherent regime where the topological surface state becomes unidentifiable and superconductivity is suppressed. Theory suggests that the electron-electron interactions in the odd-parity $xy^-$ band with a strong $d_{xy}$ character lead to an orbital-selective correlated phase. Our work establishes FeTe$_{x}$Se$_{1-x}$ thin films as a unique platform where electronic correlations sensitively modulate topological superconductivity, suggesting opportunities to use tunable electron-electron interactions to engineer new topological phases in a broad class of materials.

cond-mat.supr-con↗

arXiv2025

Valley2: Exploring Multimodal Models with Scalable Vision-Language Design

Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.

cs.CV↗

arXiv2024

Theoretical Insights into Layered Metamaterials with Enhanced Thermal and Mechanical Properties

The inherent trade-off between ultra-low thermal conductivity and high mechanical rigidity in natural materials limits their utility in advanced applications. Inspired by the unique architecture of layered honeycomb structures, this study introduces a new class of metamaterials designed to overcome these constraints. By systematically exploring unit cell configurations and stacking arrangements, we demonstrate that a zigzag internal geometry, analogous to rhombohedral graphene stacking, optimizes thermal insulation while maintaining relatively high mechanical rigidity. Our finite element simulations predict that these layered structures can achieve a thermal conductivity of 12.5 mW/(m.K) using zirconia as the constructing material, theoretically outperforming state-of-the-art ceramic aerogels while maintaining robust mechanical stability. This novel approach paves the way for designing next-generation super-insulating materials with customizable mechanical properties, enabling innovative applications in extreme environments, lightweight aerospace structures, and advanced thermal management systems.

physics.app-ph↗

arXiv2024

FastAttention: Extend FlashAttention2 to NPUs and Low-resource GPUs

FlashAttention series has been widely applied in the inference of large language models (LLMs). However, FlashAttention series only supports the high-level GPU architectures, e.g., Ampere and Hopper. At present, FlashAttention series is not easily transferrable to NPUs and low-resource GPUs. Moreover, FlashAttention series is inefficient for multi- NPUs or GPUs inference scenarios. In this work, we propose FastAttention which pioneers the adaptation of FlashAttention series for NPUs and low-resource GPUs to boost LLM inference efficiency. Specifically, we take Ascend NPUs and Volta-based GPUs as representatives for designing our FastAttention. We migrate FlashAttention series to Ascend NPUs by proposing a novel two-level tiling strategy for runtime speedup, tiling-mask strategy for memory saving and the tiling-AllReduce strategy for reducing communication overhead, respectively. Besides, we adapt FlashAttention for Volta-based GPUs by redesigning the operands layout in shared memory and introducing a simple yet effective CPU-GPU cooperative strategy for efficient memory utilization. On Ascend NPUs, our FastAttention can achieve a 10.7$\times$ speedup compared to the standard attention implementation. Llama-7B within FastAttention reaches up to 5.16$\times$ higher throughput than within the standard attention. On Volta architecture GPUs, FastAttention yields 1.43$\times$ speedup compared to its equivalents in \texttt{xformers}. Pangu-38B within FastAttention brings 1.46$\times$ end-to-end speedup using FasterTransformer. Coupled with the propose CPU-GPU cooperative strategy, FastAttention supports a maximal input length of 256K on 8 V100 GPUs. All the codes will be made available soon.

cs.LG↗

arXiv2024

Learning Granularity-Aware Affordances from Human-Object Interaction for Tool-Based Functional Dexterous Grasping

To enable robots to use tools, the initial step is teaching robots to employ dexterous gestures for touching specific areas precisely where tasks are performed. Affordance features of objects serve as a bridge in the functional interaction between agents and objects. However, leveraging these affordance cues to help robots achieve functional tool grasping remains unresolved. To address this, we propose a granularity-aware affordance feature extraction method for locating functional affordance areas and predicting dexterous coarse gestures. We study the intrinsic mechanisms of human tool use. On one hand, we use fine-grained affordance features of object-functional finger contact areas to locate functional affordance regions. On the other hand, we use highly activated coarse-grained affordance features in hand-object interaction regions to predict grasp gestures. Additionally, we introduce a model-based post-processing module that transforms affordance localization and gesture prediction into executable robotic actions. This forms GAAF-Dex, a complete framework that learns Granularity-Aware Affordances from human-object interaction to enable tool-based functional grasping with dexterous hands. Unlike fully-supervised methods that require extensive data annotation, we employ a weakly supervised approach to extract relevant cues from exocentric (Exo) images of hand-object interactions to supervise feature extraction in egocentric (Ego) images. To support this approach, we have constructed a small-scale dataset, Functional Affordance Hand-object Interaction Dataset (FAH), which includes nearly 6K images of functional hand-object interaction Exo images and Ego images. Extensive experiments on the dataset demonstrate that our method outperforms state-of-the-art methods. The source code and the established dataset are available at https://github.com/yangfan293/GAAF-DEX.

cs.RO↗

arXiv2024

Distinguishing Surface and Bulk Electromagnetism via Their Dynamics in an Intrinsic Magnetic Topological Insulator

The indirect exchange interaction between local magnetic moments via surface electrons has been long predicted to bolster the surface ferromagnetism in magnetic topological insulators (MTIs), which facilitates the quantum anomalous Hall effect. This unconventional effect is critical to determining the operating temperatures of future topotronic devices. However, the experimental confirmation of this mechanism remains elusive, especially in intrinsic MTIs. Here we combine time-resolved photoemission spectroscopy with time-resolved magneto-optical Kerr effect measurements to elucidate the unique electromagnetism at the surface of an intrinsic MTI MnBi2Te4. Theoretical modeling based on 2D Ruderman-Kittel-Kasuya-Yosida interactions captures the initial quenching of a surface-rooted exchange gap within a factor of two but over-estimates the bulk demagnetization by one order of magnitude. This mechanism directly explains the sizable gap in the quasi-2D electronic state and the nonzero residual magnetization in even-layer MnBi2Te4. Furthermore, it leads to efficient light-induced demagnetization comparable to state-of-the-art magnetophotonic crystals, promising an effective manipulation of magnetism and topological orders for future topotronics.

cond-mat.str-el↗

arXiv2024

Kilometer-Level Coupled Modeling Using 40 Million Cores: An Eight-Year Journey of Model Development

With current and future leading systems adopting heterogeneous architectures, adapting existing models for heterogeneous supercomputers is of urgent need for improving model resolution and reducing modeling uncertainty. This paper presents our three-week effort on porting a complex earth system model, CESM 2.2, to a 40-million-core Sunway supercomputer. Taking a non-intrusive approach that tries to minimizes manual code modifications, our project tries to achieve both improvement of performance and consistency of the model code. By using a hierarchical grid system and an OpenMP-based offloading toolkit, our porting and parallelization effort covers over 80% of the code, and achieves a simulation speed of 340 SDPD (simulated days per day) for 5-km atmosphere, 265 SDPD for 3-km ocean, and 222 SDPD for a coupled model, thus making multi-year or even multi-decadal experiments at such high resolution possible.

cs.DC↗

arXiv2023

O2ATH: An OpenMP Offloading Toolkit for the Sunway Heterogeneous Manycore Platform

The next generation Sunway supercomputer employs the SW26010pro processor, which features a specialized on-chip heterogeneous architecture. Applications with significant hotspots can benefit from the great computation capacity improvement of Sunway many-core architectures by carefully making intensive manual many-core parallelization efforts. However, some legacy projects with large codebases, such as CESM, ROMS and WRF, contain numerous lines of code and do not have significant hotspots. The cost of manually porting such applications to the Sunway architecture is almost unaffordable. To overcome such a challenge, we have developed a toolkit named O2ATH. O2ATH forwards GNU OpenMP runtime library calls to Sunway's Athread library, which greatly simplifies the parallelization work on the Sunway architecture.O2ATH enables users to write both MPE and CPE code in a single file, and parallelization can be achieved by utilizing OpenMP directives and attributes. In practice, O2ATH has helped us to port two large projects, CESM and ROMS, to the CPEs of the next generation Sunway supercomputers via the OpenMP offload method. In the experiments, kernel speedups range from 3 to 15 times, resulting in 3 to 6 times whole application speedups.Furthermore, O2ATH requires significantly fewer code modifications compared to manually crafting CPE functions.This indicates that O2ATH can greatly enhance development efficiency when porting or optimizing large software projects on Sunway supercomputers.

cs.PL↗