arXiv subjects

Li Hu

Publications and source records attributed to Li Hu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Dynamical friction can flip the hierarchical three-body system

In recent years, the long-term effects of non-linear perturbations were found to be important for the evolution of the hierarchical triple system, which, for the central third body of a larger mass, can significantly suppress the occurrences of orbital flip that changes the sign of angular momentum of inner binary. However, as the third-body mass increases significantly, the ambient dark matter spike becomes much more dense, rendering the effect of dynamical friction non-negligible. In this work, we take the dynamical friction into account for the first time in the hierarchical triple system up to the octupole order and find that the suppressed occurrences of orbital flip could be recovered, and as the spike index increases, the number of flips could increase over a period of time; meanwhile, as both the inner and outer semi-major axes increase while keeping their ratio fixed, the number of flips could also increase over the same number of outer orbital periods, making the detection of orbital flip a potential probe of the dark matter via observations of either electromagnetic waves or gravitational waves.

gr-qc↗

arXiv2025

Reminiscence Attack on Residuals: Exploiting Approximate Machine Unlearning for Privacy

Machine unlearning enables the removal of specific data from ML models to uphold the right to be forgotten. While approximate unlearning algorithms offer efficient alternatives to full retraining, this work reveals that they fail to adequately protect the privacy of unlearned data. In particular, these algorithms introduce implicit residuals which facilitate privacy attacks targeting at unlearned data. We observe that these residuals persist regardless of model architectures, parameters, and unlearning algorithms, exposing a new attack surface beyond conventional output-based leakage. Based on this insight, we propose the Reminiscence Attack (ReA), which amplifies the correlation between residuals and membership privacy through targeted fine-tuning processes. ReA achieves up to 1.90x and 1.12x higher accuracy than prior attacks when inferring class-wise and sample-wise membership, respectively. To mitigate such residual-induced privacy risk, we develop a dual-phase approximate unlearning framework that first eliminates deep-layer unlearned data traces and then enforces convergence stability to prevent models from "pseudo-convergence", where their outputs are similar to retrained models but still preserve unlearned residuals. Our framework works for both classification and generation tasks. Experimental evaluations confirm that our approach maintains high unlearning efficacy, while reducing the adaptive privacy attack accuracy to nearly random guess, at the computational cost of 2-12% of full retraining from scratch.

cs.LG↗

arXiv2025

dots.llm1 Technical Report

Mixture of Experts (MoE) models have emerged as a promising paradigm for scaling language models efficiently by activating only a subset of parameters for each input token. In this report, we present dots.llm1, a large-scale MoE model that activates 14B parameters out of a total of 142B parameters, delivering performance on par with state-of-the-art models while reducing training and inference costs. Leveraging our meticulously crafted and efficient data processing pipeline, dots.llm1 achieves performance comparable to Qwen2.5-72B after pretraining on 11.2T high-quality tokens and post-training to fully unlock its capabilities. Notably, no synthetic data is used during pretraining. To foster further research, we open-source intermediate training checkpoints at every one trillion tokens, providing valuable insights into the learning dynamics of large language models.

cs.CL↗

arXiv2025

BinMetric: A Comprehensive Binary Analysis Benchmark for Large Language Models

Binary analysis remains pivotal in software security, offering insights into compiled programs without source code access. As large language models (LLMs) continue to excel in diverse language understanding and generation tasks, their potential in decoding complex binary data structures becomes evident. However, the lack of standardized benchmarks in this domain limits the assessment and comparison of LLM's capabilities in binary analysis and hinders the progress of research and practical applications. To bridge this gap, we introduce BinMetric, a comprehensive benchmark designed specifically to evaluate the performance of large language models on binary analysis tasks. BinMetric comprises 1,000 questions derived from 20 real-world open-source projects across 6 practical binary analysis tasks, including decompilation, code summarization, assembly instruction generation, etc., which reflect actual reverse engineering scenarios. Our empirical study on this benchmark investigates the binary analysis capabilities of various state-of-the-art LLMs, revealing their strengths and limitations in this field. The findings indicate that while LLMs show strong potential, challenges still exist, particularly in the areas of precise binary lifting and assembly synthesis. In summary, BinMetric makes a significant step forward in measuring the binary analysis capabilities of LLMs, establishing a new benchmark leaderboard, and our study provides valuable insights for the future development of these LLMs in software security.

cs.SE↗

arXiv2025

CompileAgent: Automated Real-World Repo-Level Compilation with Tool-Integrated LLM-based Agent System

With open-source projects growing in size and complexity, manual compilation becomes tedious and error-prone, highlighting the need for automation to improve efficiency and accuracy. However, the complexity of compilation instruction search and error resolution makes automatic compilation challenging. Inspired by the success of LLM-based agents in various fields, we propose CompileAgent, the first LLM-based agent framework dedicated to repo-level compilation. CompileAgent integrates five tools and a flow-based agent strategy, enabling interaction with software artifacts for compilation instruction search and error resolution. To measure the effectiveness of our method, we design a public repo-level benchmark CompileAgentBench, and we also design two baselines for comparison by combining two compilation-friendly schemes. The performance on this benchmark shows that our method significantly improves the compilation success rate, ranging from 10% to 71%. Meanwhile, we evaluate the performance of CompileAgent under different agent strategies and verify the effectiveness of the flow-based strategy. Additionally, we emphasize the scalability of CompileAgent, further expanding its application prospects.

cs.SE↗

arXiv2025

An Empirical Study on the Effectiveness of Large Language Models for Binary Code Understanding

Binary code analysis plays a pivotal role in the field of software security and is widely used in tasks such as software maintenance, malware detection, software vulnerability discovery, patch analysis, etc. However, unlike source code, reverse engineers face significant challenges in understanding binary code due to the lack of intuitive semantic information. Although traditional reverse tools can convert binary code into C-like pseudo code, the lack of code comments and symbolic information such as function names still makes code understanding difficult. In recent years, two groups of techniques have shown promising prospects: (1) Deep learning-based techniques have demonstrated competitive results in tasks related to binary code understanding, furthermore, (2) Large Language Models (LLMs) have been extensively pre-trained at the source-code level for tasks such as code understanding and generation. This has left participants wondering about the capabilities of LLMs in binary code understanding. To this end, this work proposes a benchmark to evaluate the effectiveness of LLMs in real-world reverse engineering scenarios, which covers two key binary code understanding tasks, i.e., function name recovery and binary code summarization. To more comprehensively evaluate, we include binaries with multiple target architectures as well as different optimization options. We gain valuable insights into the capabilities and limitations through extensive empirical studies of popular LLMs using our benchmark. Our evaluations reveal that existing LLMs can understand binary code to a certain extent, thereby improving the efficiency of binary code analysis. Our results highlight the great potential of the LLMs in advancing the field of binary code understanding, and provide new directions for binary code analysis techniques.

cs.SE↗

arXiv2025

Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance

Recent character image animation methods based on diffusion models, such as Animate Anyone, have made significant progress in generating consistent and generalizable character animations. However, these approaches fail to produce reasonable associations between characters and their environments. To address this limitation, we introduce Animate Anyone 2, aiming to animate characters with environment affordance. Beyond extracting motion signals from source video, we additionally capture environmental representations as conditional inputs. The environment is formulated as the region with the exclusion of characters and our model generates characters to populate these regions while maintaining coherence with the environmental context. We propose a shape-agnostic mask strategy that more effectively characterizes the relationship between character and environment. Furthermore, to enhance the fidelity of object interactions, we leverage an object guider to extract features of interacting objects and employ spatial blending for feature injection. We also introduce a pose modulation strategy that enables the model to handle more diverse motion patterns. Experimental results demonstrate the superior performance of the proposed method.

cs.CV↗

arXiv2025

Distinctive GWBs from eccentric inspiraling SMBH binaries with a DM spike

Recent detections of a low-frequency gravitational wave background (GWB) from various pulsar-timing-array (PTA) observations have renewed the interest in the inspiraling supermassive black hole binaries (SMBHBs), whose population is believed to be the most promising candidate with possible generalizations from including either orbital eccentricity or dark matter (DM) spike. In this paper, we show that the inclusion of both can further display distinctive features detectable in future PTA observations. With a typical initial eccentricity $e_0\sim\mathcal{O}(0.1)$ for the inspiraling SMBHBs, even a shallow DM spike can easily drive the orbital eccentricity close to $1$, leaving behind a large turnover eccentricity when GWs begin to dominate the orbital circularization. In particular, the DM spike index $γ_\mathrm{sp}$ universally manifests itself in the characteristic strain by $h_c\sim f^{7/6-γ_\mathrm{sp}/3}$ in the far infrared and features a novel oscillation structure at low frequencies. Future PTA detection of such characteristics would be the smoking gun for the DM spike and even reveal the nature of DM.

gr-qc↗

arXiv2024

How Far Have We Gone in Binary Code Understanding Using Large Language Models

Binary code analysis plays a pivotal role in various software security applications, such as software maintenance, malware detection, software vulnerability discovery, patch analysis, etc. However, unlike source code, understanding binary code is challenging for reverse engineers due to the absence of semantic information. Therefore, automated tools are needed to assist human players in interpreting binary code. In recent years, two groups of technologies have shown promising prospects: (1) Deep learning-based technologies have demonstrated competitive results in tasks related to binary code understanding, furthermore, (2) Large Language Models (LLMs) have been extensively pre-trained at the source-code level for tasks such as code understanding and generation. This makes participants wonder about the ability of LLMs in binary code understanding. In this work, we propose a benchmark to evaluate the effectiveness of LLMs in real-world reverse engineering scenarios. The benchmark covers two key binary code understanding tasks, including function name recovery and binary code summarization. We gain valuable insights into their capabilities and limitations through extensive evaluations of popular LLMs using our benchmark. Our evaluations reveal that existing LLMs can understand binary code to a certain extent, thereby improving the efficiency of binary code analysis. Our results highlight the great potential of the LLMs in advancing the field of binary code understanding.

cs.SE↗

arXiv2024

Binary Code Similarity Detection via Graph Contrastive Learning on Intermediate Representations

Binary Code Similarity Detection (BCSD) plays a crucial role in numerous fields, including vulnerability detection, malware analysis, and code reuse identification. As IoT devices proliferate and rapidly evolve, their highly heterogeneous hardware architectures and complex compilation settings, coupled with the demand for large-scale function retrieval in practical applications, put forward higher requirements for BCSD methods. In this paper, we propose IRBinDiff, which mitigates compilation differences by leveraging LLVM-IR with higher-level semantic abstraction, and integrates a pre-trained language model with a graph neural network to capture both semantic and structural information from different perspectives. By introducing momentum contrastive learning, it effectively enhances retrieval capabilities in large-scale candidate function sets, distinguishing between subtle function similarities and differences. Our extensive experiments, conducted under varied compilation settings, demonstrate that IRBinDiff outperforms other leading BCSD methods in both One-to-one comparison and One-to-many search scenarios.

cs.SE↗

arXiv2024

Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation

Character Animation aims to generating character videos from still images through driving signals. Currently, diffusion models have become the mainstream in visual generation research, owing to their robust generative capabilities. However, challenges persist in the realm of image-to-video, especially in character animation, where temporally maintaining consistency with detailed information from character remains a formidable problem. In this paper, we leverage the power of diffusion models and propose a novel framework tailored for character animation. To preserve consistency of intricate appearance features from reference image, we design ReferenceNet to merge detail features via spatial attention. To ensure controllability and continuity, we introduce an efficient pose guider to direct character's movements and employ an effective temporal modeling approach to ensure smooth inter-frame transitions between video frames. By expanding the training data, our approach can animate arbitrary characters, yielding superior results in character animation compared to other image-to-video methods. Furthermore, we evaluate our method on benchmarks for fashion video and human dance synthesis, achieving state-of-the-art results.

cs.CV↗

arXiv2023

DanceMeld: Unraveling Dance Phrases with Hierarchical Latent Codes for Music-to-Dance Synthesis

In the realm of 3D digital human applications, music-to-dance presents a challenging task. Given the one-to-many relationship between music and dance, previous methods have been limited in their approach, relying solely on matching and generating corresponding dance movements based on music rhythm. In the professional field of choreography, a dance phrase consists of several dance poses and dance movements. Dance poses composed of a series of basic meaningful body postures, while dance movements can reflect dynamic changes such as the rhythm, melody, and style of dance. Taking inspiration from these concepts, we introduce an innovative dance generation pipeline called DanceMeld, which comprising two stages, i.e., the dance decouple stage and the dance generation stage. In the decouple stage, a hierarchical VQ-VAE is used to disentangle dance poses and dance movements in different feature space levels, where the bottom code represents dance poses, and the top code represents dance movements. In the generation stage, we utilize a diffusion model as a prior to model the distribution and generate latent codes conditioned on music features. We have experimentally demonstrated the representational capabilities of top code and bottom code, enabling the explicit decoupling expression of dance poses and dance movements. This disentanglement not only provides control over motion details, styles, and rhythm but also facilitates applications such as dance style transfer and dance unit editing. Our approach has undergone qualitative and quantitative experiments on the AIST++ dataset, demonstrating its superiority over other methods.

cs.OH↗

arXiv2022

Recurrent Dynamic Embedding for Video Object Segmentation

Space-time memory (STM) based video object segmentation (VOS) networks usually keep increasing memory bank every several frames, which shows excellent performance. However, 1) the hardware cannot withstand the ever-increasing memory requirements as the video length increases. 2) Storing lots of information inevitably introduces lots of noise, which is not conducive to reading the most important information from the memory bank. In this paper, we propose a Recurrent Dynamic Embedding (RDE) to build a memory bank of constant size. Specifically, we explicitly generate and update RDE by the proposed Spatio-temporal Aggregation Module (SAM), which exploits the cue of historical information. To avoid error accumulation owing to the recurrent usage of SAM, we propose an unbiased guidance loss during the training stage, which makes SAM more robust in long videos. Moreover, the predicted masks in the memory bank are inaccurate due to the inaccurate network inference, which affects the segmentation of the query frame. To address this problem, we design a novel self-correction strategy so that the network can repair the embeddings of masks with different qualities in the memory bank. Extensive experiments show our method achieves the best tradeoff between performance and speed. Code is available at https://github.com/Limingxing00/RDE-VOS-CVPR2022.

cs.CV↗

arXiv2022

Reduce the rank calculation of a high-dimensional sparse matrix based on network controllability theory

Numerical computing of the rank of a matrix is a fundamental problem in scientific computation. The datasets generated by the internet often correspond to the analysis of high-dimensional sparse matrices. Notwithstanding recent advances in the promotion of traditional singular value decomposition (SVD), an efficient estimation algorithm for the rank of a high-dimensional sparse matrix is still lacking. Inspired by the controllability theory of complex networks, we converted the rank of a matrix into maximum matching computing. Then, we established a fast rank estimation algorithm by using the cavity method, a powerful approximate technique for computing the maximum matching, to estimate the rank of a sparse matrix. In the merit of the natural low complexity of the cavity method, we showed that the rank of a high-dimensional sparse matrix can be estimated in a much faster way than SVD with high accuracy. Our method offers an efficient pathway to quickly estimate the rank of the high-dimensional sparse matrix when the time cost of computing the rank by SVD is unacceptable.

math.NA↗

arXiv2022

Plasmonic and photonic enhancement of chiral near fields

A chiral near field with a highly contorted electromagnetic field builds a bridge to match the chiral molecules and light wavelengths with large size differences. It significantly enhances the circular dichroism of chiral molecules and has great prospects in chirality sensing, detection, trapping, and other chirality-related applications. Surface plasmons feature outstanding light-trapping and electromagnetic-field-concentrating abilities. Plasmonic chiral nanostructures facilitate light manipulation to generate superchiral near fields. Meanwhile, the nanophotonic structures have attracted significant interest to obtain strong chiral fields due to their unique electromagnetic resonant properties. During the interaction of light and chiral materials, the chiral near field not only bridges the light and chiral molecules but is also responsible for the optical activities. This paper reviews state-of-the-art studies on chiral near field enhancement using plasmonic and photonic nanostructures. We review the principle of chiral electromagnetic fields and the development of plasmonic and photonic nanostructures for near field enhancement. The properties and applications of enhanced chiral near fields for chiral molecule detection, spin-orbit angular interaction, and the generation of the chiral optical force are examined. Finally, we discuss current challenges and provide a brief outlook of this field.

physics.optics↗

arXiv2021

Control core of undirected complex networks

With the development of complex networks, many researchers have paid greater attention to studying the control of complex networks over the last decade. Although some theoretical breakthroughs allow us to identify all driver nodes, we still lack an efficient method to identify the driver nodes and understand the roles of individual nodes in contributing to the control of a large complex network. Here, we apply a leaf removal process (LRP) to find a substructure of an undirected network, which is considered as the control core of the original network. Based on a strict mathematical proof, the control core obtained by the LRP has the same controllability as the original network, and it contains at least one set of driver nodes. With this method, we systematically investigate the structural property of the control core with respect to different average degrees of the original networks ($\langle k \rangle$). We denote the node density ($n_\text{core}$) and link density ($l_\text{core}$) to characterize the control core when applying the LRP, and we study the impact of $\langle k \rangle$ on $n_\text{core}$ and $l_\text{core}$ in two artificial networks: undirected Erdös-Rényi (ER) random networks and undirected scale-free (SF) networks. We find that $n_\text{core}$ and $l_\text{core}$ both change nonmonotonously with increasing $\langle k \rangle$ in the two typical undirected networks. With the aid of core percolation theory, we can offer the theoretical predictions for both $n_\text{core}$ and $l_\text{core}$ as a function of $\langle k \rangle$. Then, we recognize that finding the driver nodes in the control core is much more efficient than in the original network by comparing $n_{\text{D}}$, the controllability of the original network, and $n_{\text{core}}$, regardless of how $\langle k \rangle$ increases.

physics.soc-ph↗

arXiv2021

Learning Position and Target Consistency for Memory-based Video Object Segmentation

This paper studies the problem of semi-supervised video object segmentation(VOS). Multiple works have shown that memory-based approaches can be effective for video object segmentation. They are mostly based on pixel-level matching, both spatially and temporally. The main shortcoming of memory-based approaches is that they do not take into account the sequential order among frames and do not exploit object-level knowledge from the target. To address this limitation, we propose to Learn position and target Consistency framework for Memory-based video object segmentation, termed as LCM. It applies the memory mechanism to retrieve pixels globally, and meanwhile learns position consistency for more reliable segmentation. The learned location response promotes a better discrimination between target and distractors. Besides, LCM introduces an object-level relationship from the target to maintain target consistency, making LCM more robust to error drifting. Experiments show that our LCM achieves state-of-the-art performance on both DAVIS and Youtube-VOS benchmark. And we rank the 1st in the DAVIS 2020 challenge semi-supervised VOS task.

cs.CV↗

arXiv2020

Long- and Short-Ranged Chiral Interactions in DNA Assembled Plasmonic Chains

Molecular chirality plays a crucial role in innumerable biological processes. The chirality of a molecule can typically be identified by its characteristic optical response, the circular dichroism (CD). CD signals have thus long been used to identify the state of molecules or to follow dynamic protein configurations. In recent years, the focus has moved towards plasmonic nanostructures, as they show potential for applications ranging from pathogen sensing to novel optical materials. The plasmonic coupling of the individual elements of such chiral metallic structures is a crucial prerequisite to obtain sizeable CD signals. We here identified and implemented various coupling entities - chiral and achiral - to obtain chiral transfer over distances close to 100 nm. The coupling is realized by an achiral nanosphere situated between a pair of gold nanorods that are arranged far apart but in a chiral fashion. We synthesized these structures with nanometer precision using DNA origami and obtained sample homogeneity that allowed us to directly demonstrate efficient chiral energy transfer between the distant nanorods. The transmitter particle causes a strong enhancement in amplitude of the CD response, the emergence of an additional chiral feature at the resonance frequency of the nanosphere, and a redshift of the longitudinal plasmonic resonance frequency of the nanorods. Numerical simulations closely match our experimental observations and give insights in the intricate behavior of chiral optical fields and the transfer of plasmons in complex architectures.

physics.optics↗