Searcharxiv⌕ Search

arXiv subjects

Yifei Zhao

Publications and source records attributed to Yifei Zhao.

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

Morphogenetic Assembly and Adaptive Control for Heterogeneous Modular Robots

This paper presents a closed-loop automation framework for heterogeneous modular robots, covering the full pipeline from morphological construction to adaptive control. In this framework, a mobile manipulator handles heterogeneous functional modules including structural, joint, and wheeled modules to dynamically assemble diverse robot configurations and provide them with immediate locomotion capability. To address the state-space explosion in large-scale heterogeneous reconfiguration, we propose a hierarchical planner: the high-level planner uses a bidirectional heuristic search with type-penalty terms to generate module-handling sequences, while the low level planner employs A* search to compute optimal execution trajectories. This design effectively decouples discrete configuration planning from continuous motion execution. For adaptive motion generation of unknown assembled configurations, we introduce a GPU accelerated Annealing-Variance Model Predictive Path Integral (MPPI) controller. By incorporating a multi stage variance annealing strategy to balance global exploration and local convergence, the controller enables configuration-agnostic, real-time motion control. Large scale simulations show that the type-penalty term is critical for planning robustness in heterogeneous scenarios. Moreover, the greedy heuristic produces plans with lower physical execution costs than the Hungarian heuristic. The proposed annealing-variance MPPI significantly outperforms standard MPPI in both velocity tracking accuracy and control frequency, achieving real time control at 50 Hz. The framework validates the full-cycle process, including module assembly, robot merging and splitting, and dynamic motion generation.

cs.RO↗
arXiv2026

HAFO: A Force-Adaptive Control Framework for Humanoid Robots in Intense Interaction Environments

Reinforcement learning (RL) controllers have made impressive progress in humanoid locomotion and light-weight object manipulation. However, achieving robust and precise motion control with intense force interaction remains a significant challenge. To address these limitations, this paper proposes HAFO, a dual-agent reinforcement learning framework that concurrently optimizes both a robust locomotion strategy and a precise upper-body manipulation strategy via coupled training. We employ a constrained residual action space to improve dual-agent training stability and sample efficiency. The external tension disturbances are explicitly modeled using a spring-damper system, allowing for fine-grained force control through manipulation of the virtual spring. In this process, the reinforcement learning policy autonomously generates a disturbance-rejection response by utilizing environmental feedback. The experimental results demonstrate that HAFO achieves whole-body control for humanoid robots across diverse force-interaction environments using a single dual-agent policy, delivering outstanding performance under load-bearing and thrust-disturbance conditions, while maintaining stable operation even under rope suspension state.

cs.RO↗
arXiv2025

The quantum torus as an $\mathbb E_M$-category

Given an oriented $2$-manifold $M$, a locally constant sheaf of lattices $Λ$ over $M$, and a pointed morphism $q : \textsf B^2Λ\rightarrow \textsf B^4\mathbf C^{\times}$, we define an $\mathbb E_M$-category $\mathrm{Rep}_q(\check T)$ which we call the "quantum torus" at level $q$. We explain why this terminology is deserved and calculate the factorization homology of $\mathrm{Rep}_q(\check T)$. When $M$ arises from a global complex curve, we confirm (a version of) a conjecture of Ben-Zvi and Nadler for tori.

math.RT↗
arXiv2025

PSDiffusion: Harmonized Multi-Layer Image Generation via Layout and Appearance Alignment

Transparent image layer generation plays a significant role in digital art and design workflows. Existing methods typically decompose transparent layers from a single RGB image using a set of tools or generate multiple transparent layers sequentially. Despite some promising results, these methods often limit their ability to model global layout, physically plausible interactions, and visual effects such as shadows and reflections with high alpha quality due to limited shared global context among layers. To address this issue, we propose PSDiffusion, a unified diffusion framework that leverages image composition priors from pre-trained image diffusion model for simultaneous multi-layer text-to-image generation. Specifically, our method introduces a global layer interaction mechanism to generate layered images collaboratively, ensuring both individual layer quality and coherent spatial and visual relationships across layers. We include extensive experiments on benchmark datasets to demonstrate that PSDiffusion is able to outperform existing methods in generating multi-layer images with plausible structure and enhanced visual fidelity.

cs.CV↗
arXiv2025

GAP: Graph-Based Agent Planning with Parallel Tool Use and Reinforcement Learning

Autonomous agents powered by large language models (LLMs) have shown impressive capabilities in tool manipulation for complex task-solving. However, existing paradigms such as ReAct rely on sequential reasoning and execution, failing to exploit the inherent parallelism among independent sub-tasks. This sequential bottleneck leads to inefficient tool utilization and suboptimal performance in multi-step reasoning scenarios. We introduce Graph-based Agent Planning (GAP), a novel framework that explicitly models inter-task dependencies through graph-based planning to enable adaptive parallel and serial tool execution. Our approach trains agent foundation models to decompose complex tasks into dependency-aware sub-task graphs, autonomously determining which tools can be executed in parallel and which must follow sequential dependencies. This dependency-aware orchestration achieves substantial improvements in both execution efficiency and task accuracy. To train GAP, we construct a high-quality dataset of graph-based planning traces derived from the Multi-Hop Question Answering (MHQA) benchmark. We employ a two-stage training strategy: supervised fine-tuning (SFT) on the curated dataset, followed by reinforcement learning (RL) with a correctness-based reward function on strategically sampled queries where tool-based reasoning provides maximum value. Experimental results on MHQA datasets demonstrate that GAP significantly outperforms traditional ReAct baselines, particularly on multi-step retrieval tasks, while achieving dramatic improvements in tool invocation efficiency through intelligent parallelization. The project page is available at: https://github.com/WJQ7777/Graph-Agent-Planning.

cs.AI↗
arXiv2025

A Domain Adaptive Position Reconstruction Method for Time Projection Chamber based on Deep Neural Network

Transverse position reconstruction in a Time Projection Chamber (TPC) is crucial for accurate particle tracking and classification, and is typically accomplished using machine learning techniques. However, these methods often exhibit biases and limited resolution due to incompatibility between real experimental data and simulated training samples. To mitigate this issue, we present a domain-adaptive reconstruction approach based on a cycle-consistent generative adversarial network. In the prototype detector, the application of this method led to a 60.6% increase in the reconstructed radial boundary. Scaling this method to a simulated 50-kg TPC, by evaluating the resolution of simulated events, an additional improvement of at least 27% is achieved.

hep-ex↗
arXiv2025

PRO: Enabling Precise and Robust Text Watermark for Open-Source LLMs

Text watermarking for large language models (LLMs) enables model owners to verify text origin and protect intellectual property. While watermarking methods for closed-source LLMs are relatively mature, extending them to open-source models remains challenging, as developers cannot control the decoding process. Consequently, owners of open-source LLMs lack practical means to verify whether text was generated by their models. A core difficulty lies in embedding watermarks directly into model weights without hurting detectability. A promising idea is to distill watermarks from a closed-source model into an open one, but this suffers from (i) poor detectability due to mismatch between learned and predefined patterns, and (ii) fragility to downstream modifications such as fine-tuning or model merging. To overcome these limitations, we propose PRO, a Precise and Robust text watermarking method for open-source LLMs. PRO jointly trains a watermark policy model with the LLM, producing patterns that are easier for the model to learn and more consistent with detection criteria. A regularization term further simulates downstream perturbations and penalizes degradation in watermark detectability, ensuring robustness under model edits. Experiments on open-source LLMs (e.g., LLaMA-3.2, LLaMA-3, Phi-2) show that PRO substantially improves both watermark detectability and resilience to model modifications.

cs.CR↗
arXiv2025

Securing Transformer-based AI Execution via Unified TEEs and Crypto-protected Accelerators

Recent advances in Transformer models, e.g., large language models (LLMs), have brought tremendous breakthroughs in various artificial intelligence (AI) tasks, leading to their wide applications in many security-critical domains. Due to their unprecedented scale and prohibitively high development cost, these models have become highly valuable intellectual property for AI stakeholders and are increasingly deployed via machine learning as a service (MLaaS). However, MLaaS often runs on untrusted cloud infrastructure, exposing data and models to potential breaches. Mainstream protection mechanisms leverage trusted execution environments (TEEs) where confidentiality and integrity for secretive data are shielded using hardware-based encryption and integrity checking. Unfortunately, running model inference entirely within TEEs is subject to non-trivial slowdown, which is further exacerbated in LLMs due to the substantial computation and memory footprint involved. Recent studies reveal that the hybrid TEE-based scheme offloading partial model inference operations to the untrusted accelerators (e.g., GPU) is a promising solution. However, prior offloading schemes fail to ensure dual protection of data and model in Transformer inference, as they cannot securely offload critical operations, i.e., Attention and SoftMax, forcing these computations to remain confined within TEEs. To address these challenges, we propose TwinShield, a framework enabling secure Transformer inference in heterogeneous TEE and accelerator systems with dual protection for both model and data. TwinShield offloads ~87% of computation to GPUs and delivers 4.0x - 6.1x speedups over previous approaches across various Transformer models.

cs.CR↗
arXiv2025

What are the extended pure inner forms of a cover?

Kottwitz suggested to study all extended pure inner forms together in the local Langlands correspondence for linear reductive groups. We extend this philosophy to a large class of covers, including those defined by Brylinski and Deligne, and explain its relation with Weissman's observation that L-packets for covers are sometimes empty.

math.RT↗
arXiv2025

Lifelong Safety Alignment for Language Models

LLMs have made impressive progress, but their growing capabilities also expose them to highly flexible jailbreaking attacks designed to bypass safety alignment. While many existing defenses focus on known types of attacks, it is more critical to prepare LLMs for unseen attacks that may arise during deployment. To address this, we propose a lifelong safety alignment framework that enables LLMs to continuously adapt to new and evolving jailbreaking strategies. Our framework introduces a competitive setup between two components: a Meta-Attacker, trained to actively discover novel jailbreaking strategies, and a Defender, trained to resist them. To effectively warm up the Meta-Attacker, we first leverage the GPT-4o API to extract key insights from a large collection of jailbreak-related research papers. Through iterative training, the first iteration Meta-Attacker achieves a 73% attack success rate (ASR) on RR and a 57% transfer ASR on LAT using only single-turn attacks. Meanwhile, the Defender progressively improves its robustness and ultimately reduces the Meta-Attacker's success rate to just 7%, enabling safer and more reliable deployment of LLMs in open-ended environments. The code is available at https://github.com/sail-sg/LifelongSafetyAlignment.

cs.CR↗
arXiv2025

Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Models

Standing in 2025, at a critical juncture in the pursuit of Artificial General Intelligence (AGI), reinforcement fine-tuning (RFT) has demonstrated significant potential in enhancing the reasoning capability of large language models (LLMs) and has led to the development of cutting-edge AI models such as OpenAI-o1 and DeepSeek-R1. Moreover, the efficient application of RFT to enhance the reasoning capability of multimodal large language models (MLLMs) has attracted widespread attention from the community. In this position paper, we argue that reinforcement fine-tuning powers the reasoning capability of multimodal large language models. To begin with, we provide a detailed introduction to the fundamental background knowledge that researchers interested in this field should be familiar with. Furthermore, we meticulously summarize the improvements of RFT in powering reasoning capability of MLLMs into five key points: diverse modalities, diverse tasks and domains, better training algorithms, abundant benchmarks and thriving engineering frameworks. Finally, we propose five promising directions for future research that the community might consider. We hope that this position paper will provide valuable insights to the community at this pivotal stage in the advancement toward AGI. Summary of works done on RFT for MLLMs is available at https://github.com/Sun-Haoyuan23/Awesome-RL-based-Reasoning-MLLMs.

cs.CL↗
arXiv2025

Half-integral levels

We construct equivalences among four notions associated to a reductive group scheme $G$: factorization super central extensions of the loop group of $G$ by $\mathbb G_m$ subject to a condition on the commutator, factorization super line bundles on the affine Grassmannian of $G$, rigidified sections of a quotient of $2$-truncated K-theory over the Zariski classifying stack of $G$, and combinatorial data defined by Brylinski and Deligne in a conjectural extension of their classification theorem.

math.AG↗
arXiv2024

Reactor neutrino liquid xenon coherent elastic scattering experiment

Coherent elastic neutrino-nucleus scattering (CEvNS) provides a unique probe for neutrino properties Beyond the Standard Model (BSM) physics. REactor neutrino LIquid xenon Coherent Scattering experiment (RELICS), a proposed reactor neutrino program using liquid xenon time projection chamber (LXeTPC) technology, aims to investigate the CEvNS process of antineutrinos off xenon atomic nuclei. In this work, the design of the experiment is studied and optimized based on Monte Carlo (MC) simulations. To achieve a sufficiently low energy threshold for CEvNS detection, an ionization-only analysis channel is adopted for RELICS. A high emission rate of delayed electrons after a big ionization signal is the major background, leading to an analysis threshold of 120 photo-electrons in the CEvNS search. The second largest background, nuclear recoils induced by cosmic-ray neutrons, is suppressed via a passive water shield. The physics potential of RELICS is explored with a 32 kg*yr exposure at a baseline of 25 m from a reactor core with a 3 GW thermal power. In an energy range of 120 to 300 PE, corresponding to an average nuclear recoil from 0.63 to 1.36 keV considering the liquid xenon response and detector-related effect, we expect 4639.7 CEvNS and 1687.8 background events. The sensitivity of RELICS to the weak mixing angle is investigated at a low momentum transfer. Our study shows that RELICS can further improve the constraints on the non-standard neutrino interaction (NSI) compared to the current best results.

hep-ex↗
arXiv2024

Artistic Intelligence: A Diffusion-Based Framework for High-Fidelity Landscape Painting Synthesis

Generating high-fidelity landscape paintings remains a challenging task that requires precise control over both structure and style. In this paper, we present LPGen, a novel diffusion-based model specifically designed for landscape painting generation. LPGen introduces a decoupled cross-attention mechanism that independently processes structural and stylistic features, effectively mimicking the layered approach of traditional painting techniques. Additionally, LPGen proposes a structural controller, a multi-scale encoder designed to control the layout of landscape paintings, striking a balance between aesthetics and composition. Besides, the model is pre-trained on a curated dataset of high-resolution landscape images, categorized by distinct artistic styles, and then fine-tuned to ensure detailed and consistent output. Through extensive evaluations, LPGen demonstrates superior performance in producing paintings that are not only structurally accurate but also stylistically coherent, surpassing current state-of-the-art models. This work advances AI-generated art and offers new avenues for exploring the intersection of technology and traditional artistic practices. Our code, dataset, and model weights will be publicly available.

cs.CV↗
arXiv2024

OAKINK2: A Dataset of Bimanual Hands-Object Manipulation in Complex Task Completion

We present OAKINK2, a dataset of bimanual object manipulation tasks for complex daily activities. In pursuit of constructing the complex tasks into a structured representation, OAKINK2 introduces three level of abstraction to organize the manipulation tasks: Affordance, Primitive Task, and Complex Task. OAKINK2 features on an object-centric perspective for decoding the complex tasks, treating them as a sequence of object affordance fulfillment. The first level, Affordance, outlines the functionalities that objects in the scene can afford, the second level, Primitive Task, describes the minimal interaction units that humans interact with the object to achieve its affordance, and the third level, Complex Task, illustrates how Primitive Tasks are composed and interdependent. OAKINK2 dataset provides multi-view image streams and precise pose annotations for the human body, hands and various interacting objects. This extensive collection supports applications such as interaction reconstruction and motion synthesis. Based on the 3-level abstraction of OAKINK2, we explore a task-oriented framework for Complex Task Completion (CTC). CTC aims to generate a sequence of bimanual manipulation to achieve task objectives. Within the CTC framework, we employ Large Language Models (LLMs) to decompose the complex task objectives into sequences of Primitive Tasks and have developed a Motion Fulfillment Model that generates bimanual hand motion for each Primitive Task. OAKINK2 datasets and models are available at https://oakink.net/v2.

cs.CV↗
arXiv2023

Wildfire Detection Via Transfer Learning: A Survey

This paper surveys different publicly available neural network models used for detecting wildfires using regular visible-range cameras which are placed on hilltops or forest lookout towers. The neural network models are pre-trained on ImageNet-1K and fine-tuned on a custom wildfire dataset. The performance of these models is evaluated on a diverse set of wildfire images, and the survey provides useful information for those interested in using transfer learning for wildfire detection. Swin Transformer-tiny has the highest AUC value but ConvNext-tiny detects all the wildfire events and has the lowest false alarm rate in our dataset.

cs.CV↗
arXiv2023

Étale metaplectic covers of reductive group schemes

Given a reductive group scheme $G$, we give a linear algebraic description of reduced étale $4$-cocycles on its classifying stack $\mathrm B(G)$. These cocycles form a $2$-groupoid, which we interpret as parameters of metaplectic covers of $G$. We use our linear algebraic description to define the Langlands dual of a metaplectic cover.

math.AG↗