arXiv subjects

Yun Du

Publications and source records attributed to Yun Du.

16 recordsLinked to original sources

arXiv2026

InstructMove: A Text-Indispensable Benchmark for Instruction-Following Manipulation

Vision-language-action (VLA) models have made general-purpose robot manipulation increasingly plausible by conditioning robot actions on natural-language instructions. A key test of such generality is whether policies actually follow language instructions. Yet many manipulation benchmarks leave this ability underdetermined: the intended object or destination is often visually salient or uniquely feasible, allowing policies to succeed without grounding the instruction. We argue that instruction-following evaluation should be text-indispensable: multiple actions should be visually and physically plausible, while only one should be consistent with the language instruction. We introduce InstructMove, a text-indispensable benchmark for instruction-following manipulation. InstructMove instantiates this principle in pick-and-place scenes with semantic distractors, decomposing instruction following into category identification, attribute discrimination, spatial reasoning, and compositional pick-and-place. InstructMove supports a train-eval protocol with InstructMove training data and held-out evaluation tasks, with additional diagnostics for language dependence. Experiments with representative VLA policies show that InstructMove provides a controlled testbed for diagnosing visual shortcuts and that InstructMove simulation data can improve real-world instruction-following manipulation performance. Code: https://github.com/HorizonRobotics/RoboOrchardSim

cs.RO↗

arXiv2026

HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory

LLM agents follow a practical execution loop in digital environments: they reason over structured states, invoke tools, inspect feedback, and revise actions. Extending this loop to physical robots is difficult because physical execution is continuous, embodiment-dependent, uncertain, and constrained by safety. Existing embodied-AI systems have advanced manipulation, spatial understanding, navigation, and humanoid control, but these capabilities often remain specialized modules or loosely coupled decision loops. In this work, we introduce HoloAgent-0, a unified embodied agent framework for real-world robot deployment. Embodied AgentOS converts language instructions into executable skill graphs, schedules robot resources, monitors execution, and triggers clarification or re-planning from runtime feedback. HoloAgent-0 organizes heterogeneous robot models and controllers through three coupled layers: Embodied AgentOS for closed-loop execution, 3D spatial memory for physical world grounding, and embodied skills for robot action. We deploy HoloAgent-0 on real hardware and evaluate its spatial memory, long-horizon navigation, and closed-loop execution across motion generation, object search, cross-robot coordination, and mobile manipulation.

cs.RO↗

arXiv2026

ManiSplat: Manipulation Trajectory Synthesis from Monocular Video via Decoupled 3D Gaussian Splatting

Reconstructing dynamic and interactive 3D scenes from real-world observations remains a fundamental challenge in computer vision and robotics. While recent advances in 3D Gaussian Splatting have enabled high-fidelity static reconstruction, extending it to interactive environments with articulated robots and manipulable objects remains difficult due to complex contact interactions and abrupt pose changes. To address these challenges, we introduce ManiSplat, a unified framework that reconstructs controllable and decoupled Gaussian digital twins directly from monocular ego-view robotic videos. Our method introduces a Graph-Structured Disentangled Representation that separates the robot, objects, and background into independently optimizable Gaussian subfields organized within a scene graph. To ensure stability, we propose a Task-Oriented Spatio-Temporal Alignment module that leverages the inherent logic of manipulation tasks-alternating between Motion and Skill phases-to construct accurate pseudo-ground-truth trajectories. Finally, a joint photometric-geometric optimization ensures the reconstructed scenes are temporally coherent, physically consistent, and simulation-ready. Extensive experiments demonstrate that our approach reconstructs interaction-driven dynamic scenes with high fidelity and controllability, effectively supporting downstream robotic tasks and policy learning.

cs.CV↗

arXiv2026

HoloBrain-0 Technical Report

In this work, we introduce HoloBrain-0, a comprehensive Vision-Language-Action (VLA) framework that bridges the gap between foundation model research and reliable real-world robot deployment. The core of our system is a novel VLA architecture that explicitly incorporates robot embodiment priors, including multi-view camera parameters and kinematic descriptions (URDF), to enhance 3D spatial reasoning and support diverse embodiments. We validate this design through a scalable ``pre-train then post-train" paradigm, achieving state-of-the-art results on simulation benchmarks such as RoboTwin 2.0, LIBERO, and GenieSim, as well as strong results on challenging long-horizon real-world manipulation tasks. Notably, our efficient 0.2B-parameter variant rivals significantly larger baselines, enabling low-latency on-device deployment. To further accelerate research and practical adoption, we fully open-source the entire HoloBrain ecosystem, which includes: (1) powerful pre-trained VLA foundations; (2) post-trained checkpoints for multiple simulation suites and real-world tasks; and (3) RoboOrchard, a full-stack VLA infrastructure for data curation, model training and deployment. Together with standardized data collection protocols, this release provides the community with a complete, reproducible path toward high-performance robotic manipulation.

cs.RO↗

arXiv2025

SnapStream: Efficient Long Sequence Decoding on Dataflow Accelerators

The proliferation of 100B+ parameter Large Language Models (LLMs) with 100k+ context length support have resulted in increasing demands for on-chip memory to support large KV caches. Techniques such as StreamingLLM and SnapKV demonstrate how to control KV cache size while maintaining model accuracy. Yet, these techniques are not commonly used within industrial deployments using frameworks like vLLM or SGLang. The reason is twofold: on one hand, the static graphs and continuous batching methodology employed by these frameworks make it difficult to admit modifications to the standard multi-head attention algorithm, while on the other hand, the accuracy implications of such techniques on modern instruction-following and reasoning models are not well understood, obfuscating the need for implementing these techniques. In this paper, we explore these accuracy implications on Llama-3.1-8B-Instruct and DeepSeek-R1, and develop SnapStream, a KV cache compression method that can be deployed at scale. We demonstrate the efficacy of SnapStream in a 16-way tensor-parallel deployment of DeepSeek-671B on SambaNova SN40L accelerators running at 128k context length and up to 1832 tokens per second in a real production setting. SnapStream enables $4\times$ improved on-chip memory usage and introduces minimal accuracy degradation on LongBench-v2, AIME24 and LiveCodeBench. To the best of our knowledge, this is the first implementation of sparse KV attention techniques deployed in a production inference system with static graphs and continuous batching.

cs.AI↗

arXiv2025

FineGrasp: Towards Robust Grasping for Delicate Objects

Recent advancements in robotic grasping have led to its integration as a core module in many manipulation systems. For instance, language-driven semantic segmentation enables the grasping of any designated object or object part. However, existing methods often struggle to generate feasible grasp poses for small objects or delicate components, potentially causing the entire pipeline to fail. To address this issue, we propose a novel grasping method, FineGrasp, which introduces improvements in three key aspects. First, we introduce multiple network modifications to enhance the ability of to handle delicate regions. Second, we address the issue of label imbalance and propose a refined graspness label normalization strategy. Third, we introduce a new simulated grasp dataset and show that mixed sim-to-real training further improves grasp performance. Experimental results show significant improvements, especially in grasping small objects, and confirm the effectiveness of our system in semantic grasping.

cs.RO↗

arXiv2024

SambaLingo: Teaching Large Language Models New Languages

Despite the widespread availability of LLMs, there remains a substantial gap in their capabilities and availability across diverse languages. One approach to address these issues has been to take an existing pre-trained LLM and continue to train it on new languages. While prior works have experimented with language adaptation, many questions around best practices and methodology have not been covered. In this paper, we present a comprehensive investigation into the adaptation of LLMs to new languages. Our study covers the key components in this process, including vocabulary extension, direct preference optimization and the data scarcity problem for human alignment in low-resource languages. We scale these experiments across 9 languages and 2 parameter scales (7B and 70B). We compare our models against Llama 2, Aya-101, XGLM, BLOOM and existing language experts, outperforming all prior published baselines. Additionally, all evaluation code and checkpoints are made public to facilitate future research.

cs.CL↗

arXiv2024

Relative Difficulty Distillation for Semantic Segmentation

Current knowledge distillation (KD) methods primarily focus on transferring various structured knowledge and designing corresponding optimization goals to encourage the student network to imitate the output of the teacher network. However, introducing too many additional optimization objectives may lead to unstable training, such as gradient conflicts. Moreover, these methods ignored the guidelines of relative learning difficulty between the teacher and student networks. Inspired by human cognitive science, in this paper, we redefine knowledge from a new perspective -- the student and teacher networks' relative difficulty of samples, and propose a pixel-level KD paradigm for semantic segmentation named Relative Difficulty Distillation (RDD). We propose a two-stage RDD framework: Teacher-Full Evaluated RDD (TFE-RDD) and Teacher-Student Evaluated RDD (TSE-RDD). RDD allows the teacher network to provide effective guidance on learning focus without additional optimization goals, thus avoiding adjusting learning weights for multiple losses. Extensive experimental evaluations using a general distillation loss function on popular datasets such as Cityscapes, CamVid, Pascal VOC, and ADE20k demonstrate the effectiveness of RDD against state-of-the-art KD methods. Additionally, our research showcases that RDD can integrate with existing KD methods to improve their upper performance bound.

cs.CV↗

arXiv2024

SambaNova SN40L: Scaling the AI Memory Wall with Dataflow and Composition of Experts

Monolithic large language models (LLMs) like GPT-4 have paved the way for modern generative AI applications. Training, serving, and maintaining monolithic LLMs at scale, however, remains prohibitively expensive and challenging. The disproportionate increase in compute-to-memory ratio of modern AI accelerators have created a memory wall, necessitating new methods to deploy AI. Composition of Experts (CoE) is an alternative modular approach that lowers the cost and complexity of training and serving. However, this approach presents two key challenges when using conventional hardware: (1) without fused operations, smaller models have lower operational intensity, which makes high utilization more challenging to achieve; and (2) hosting a large number of models can be either prohibitively expensive or slow when dynamically switching between them. In this paper, we describe how combining CoE, streaming dataflow, and a three-tier memory system scales the AI memory wall. We describe Samba-CoE, a CoE system with 150 experts and a trillion total parameters. We deploy Samba-CoE on the SambaNova SN40L Reconfigurable Dataflow Unit (RDU) - a commercial dataflow accelerator architecture that has been co-designed for enterprise inference and training applications. The chip introduces a new three-tier memory system with on-chip distributed SRAM, on-package HBM, and off-package DDR DRAM. A dedicated inter-RDU network enables scaling up and out over multiple sockets. We demonstrate speedups ranging from 2$\times$ to 13$\times$ on various benchmarks running on eight RDU sockets compared with an unfused baseline. We show that for CoE inference deployments, the 8-socket RDU Node reduces machine footprint by up to 19$\times$, speeds up model switching time by 15$\times$ to 31$\times$, and achieves an overall speedup of 3.7$\times$ over a DGX H100 and 6.6$\times$ over a DGX A100.

cs.AR↗

arXiv2023

Gauss-Bonnet modification to Hawking evaporation of AdS black holes in massive gravity

The Stefan-Boltzmann law can estimate particle emission power and lifetime of a black hole. However, in modified gravity theories, new parameters in the action can cause qualitative changes in thermodynamic quantities, thus obtaining specific thermodynamic properties often requires complicated calculation with higher degree equations. In this work, we aim to provide a general model-independent description of the evolution of AdS black holes, using Gauss-Bonnet massive gravity as an example. We prove that the impact factor of an infinitely large AdS black hole is equal to the effective AdS radius, and the black hole is able to evaporate an infinite amount of mass in finite time, so that the lifetime of the black hole depends on the final state temperature in the evaporation process. The black hole will evaporate out when the final state temperature diverges and will transform into a remnant when the temperature is zero. Since we have analyzed massive gravity in detail, we can introduce the Gauss-Bonnet term to study how it affects the thermodynamic quantities. We obtain the final states for different parameter intervals, study the qualitative properties of black hole evaporation, and classify the associated cases. This method can also be applied to other models.

gr-qc↗

arXiv2022

Noise-Robust Bidirectional Learning with Dynamic Sample Reweighting

Deep neural networks trained with standard cross-entropy loss are more prone to memorize noisy labels, which degrades their performance. Negative learning using complementary labels is more robust when noisy labels intervene but with an extremely slow model convergence speed. In this paper, we first introduce a bidirectional learning scheme, where positive learning ensures convergence speed while negative learning robustly copes with label noise. Further, a dynamic sample reweighting strategy is proposed to globally weaken the effect of noise-labeled samples by exploiting the excellent discriminatory ability of negative learning on the sample probability distribution. In addition, we combine self-distillation to further improve the model performance. The code is available at \url{https://github.com/chenchenzong/BLDR}.

cs.LG↗

arXiv2022

Exploring Contextual Relationships for Cervical Abnormal Cell Detection

Cervical abnormal cell detection is a challenging task as the morphological discrepancies between abnormal and normal cells are usually subtle. To determine whether a cervical cell is normal or abnormal, cytopathologists always take surrounding cells as references to identify its abnormality. To mimic these behaviors, we propose to explore contextual relationships to boost the performance of cervical abnormal cell detection. Specifically, both contextual relationships between cells and cell-to-global images are exploited to enhance features of each region of interest (RoI) proposals. Accordingly, two modules, dubbed as RoI-relationship attention module (RRAM) and global RoI attention module (GRAM), are developed and their combination strategies are also investigated. We establish a strong baseline by using Double-Head Faster R-CNN with feature pyramid network (FPN) and integrate our RRAM and GRAM into it to validate the effectiveness of the proposed modules. Experiments conducted on a large cervical cell detection dataset reveal that the introduction of RRAM and GRAM both achieves better average precision (AP) than the baseline methods. Moreover, when cascading RRAM and GRAM, our method outperforms the state-of-the-art (SOTA) methods. Furthermore, we also show the proposed feature enhancing scheme can facilitate both image-level and smear-level classification. The code and trained models are publicly available at https://github.com/CVIU-CSU/CR4CACD.

cs.CV↗

arXiv2021

Semantically Contrastive Learning for Low-light Image Enhancement

Low-light image enhancement (LLE) remains challenging due to the unfavorable prevailing low-contrast and weak-visibility problems of single RGB images. In this paper, we respond to the intriguing learning-related question -- if leveraging both accessible unpaired over/underexposed images and high-level semantic guidance, can improve the performance of cutting-edge LLE models? Here, we propose an effective semantically contrastive learning paradigm for LLE (namely SCL-LLE). Beyond the existing LLE wisdom, it casts the image enhancement task as multi-task joint learning, where LLE is converted into three constraints of contrastive learning, semantic brightness consistency, and feature preservation for simultaneously ensuring the exposure, texture, and color consistency. SCL-LLE allows the LLE model to learn from unpaired positives (normal-light)/negatives (over/underexposed), and enables it to interact with the scene semantics to regularize the image enhancement network, yet the interaction of high-level semantic knowledge and the low-level signal prior is seldom investigated in previous methods. Training on readily available open data, extensive experiments demonstrate that our method surpasses the state-of-the-arts LLE models over six independent cross-scenes datasets. Moreover, SCL-LLE's potential to benefit the downstream semantic segmentation under extremely dark conditions is discussed. Source Code: https://github.com/LingLIx/SCL-LLE.

cs.CV↗

arXiv2020

Cloud detection in Landsat-8 imagery in Google Earth Engine based on a deep neural network

Google Earth Engine (GEE) provides a convenient platform for applications based on optical satellite imagery of large areas. With such data sets, the detection of cloud is often a necessary prerequisite step. Recently, deep learning-based cloud detection methods have shown their potential for cloud detection but they can only be applied locally, leading to inefficient data downloading time and storage problems. This letter proposes a method to directly perform cloud detection in Landsat-8 imagery in GEE based on deep learning (DeepGEE-CD). A deep neural network (DNN) was first trained locally, and then the trained DNN was deployed in the JavaScript client of GEE. An experiment was undertaken to validate the proposed method with a set of Landsat-8 images and the results show that DeepGEE-CD outperformed the widely used function of mask (Fmask) algorithm. The proposed DeepGEE-CD approach can accurately detect cloud in Landsat-8 imagery without downloading it, making it a promising method for routine cloud detection of Landsat-8 imagery in GEE.

eess.IV↗

arXiv2016

Deformed square resonator lasers for optical frequency comb generation

Deformed square resonators with the flat sides replaced by circular sides are proposed and demonstrated to enhance mode Q factors and adjust transverse mode intervals using the regular ray dynamic analysis and numerical simulations. Dual-transverse-mode emissions due to the ultrahigh-Q factors with different wavelength intervals are realized experimentally for AlGaInAs/InP circular-side square microlasers, and the stationary condition of the dual-mode emission is satisfied because the high-Q confined modes have totally different mode numbers. Furthermore, optical frequency combs are generated using the dual-mode lasing microlaser as a seeding light source by cascaded four-wave mixing in a highly nonlinear optical fiber.

physics.optics↗

arXiv2012

Effect of the momentum dependence of nuclear symmetry potential on the transverse and elliptic flows

In the framework of the isospin-dependent Boltzmann-Uehling-Uhlenbeck transport model, effect of the momentum dependence of nuclear symmetry potential on nuclear transverse and elliptic flows in the neutron-rich reaction $^{132}$Sn+$^{124}$Sn at a beam energy of 400 MeV/nucleon is studied. We find that the momentum dependence of nuclear symmetry potential affects the rapidity distribution of the free neutron to proton ratio, the neutron and the proton transverse flows as a function of rapidity. The momentum dependence of nuclear symmetry potential affects the neutron-proton differential transverse flow more evidently than the difference of neutron and proton transverse flows as well as the difference of proton and neutron elliptic flows. It is thus better to probe the symmetry energy by using the difference of neutron and proton flows since the momentum dependence of nuclear symmetry potential is still an open question. And it is better to probe the momentum dependence of nuclear symmetry potential by using the neutron-proton differential transverse flow and the rapidity distribution of the free neutron to proton ratio.

nucl-th↗