arXiv subjects

Yifei Zhang

Publications and source records attributed to Yifei Zhang.

At least 145 records · Page 8Linked to original sources

arXiv2024

RoCar: A Relationship Network-based Evaluation Method for Large Language Models

Large language models (LLMs) have received increasing attention. However, due to the complexity of its capabilities, how to rationally evaluate the capabilities of LLMs is still a task to be solved. We propose the RoCar method, which utilizes the defined basic schemas to randomly construct a task graph and generates natural language evaluation tasks based on the task graph to evaluate the reasoning and memory abilities of LLMs respectively. Due to the very large randomness of the task construction process, it is possible to ensure that none of the LLMs to be tested has directly learned the evaluation tasks, guaranteeing the fairness of the evaluation method.

cs.CL↗

arXiv2024

Less is More: Extreme Gradient Boost Rank-1 Adaption for Efficient Finetuning of LLMs

Fine-tuning Large Language Models (LLMs) has become a crucial technique for adapting pre-trained models to downstream tasks. However, the enormous size of LLMs poses significant challenges in terms of computational complexity and resource requirements. Low-Rank Adaptation (LoRA) has emerged as a promising solution. However, there exists a gap between the practical performance of low-rank adaptations and its theoretical optimum. In this work, we propose eXtreme Gradient Boosting LoRA (XGBLoRA), a novel framework that bridges this gap by leveraging the power of ensemble learning. Inspired by gradient boosting, XGBLoRA iteratively learns and merges a sequence of LoRA adaptations to refine model predictions. It achieves better performance than the standard LoRA, while enjoying the computational efficiency of rank-1 adaptations. We provide theoretical analysis to show the convergence and optimality of our approach, and conduct extensive experiments on a range of natural language processing tasks. The results demonstrate that XGBLoRA consistently outperforms standard LoRA and achieves performance comparable to full fine-tuning with significantly fewer trainable parameters. This work advances parameter-efficient fine-tuning for LLMs, and offers a promising solution for adapting LLMs to downstream tasks while optimizing performance and efficiency.

cs.CL↗

arXiv2024

Sparse Diffusion Policy: A Sparse, Reusable, and Flexible Policy for Robot Learning

The increasing complexity of tasks in robotics demands efficient strategies for multitask and continual learning. Traditional models typically rely on a universal policy for all tasks, facing challenges such as high computational costs and catastrophic forgetting when learning new tasks. To address these issues, we introduce a sparse, reusable, and flexible policy, Sparse Diffusion Policy (SDP). By adopting Mixture of Experts (MoE) within a transformer-based diffusion policy, SDP selectively activates experts and skills, enabling efficient and task-specific learning without retraining the entire model. SDP not only reduces the burden of active parameters but also facilitates the seamless integration and reuse of experts across various tasks. Extensive experiments on diverse tasks in both simulations and real world show that SDP 1) excels in multitask scenarios with negligible increases in active parameters, 2) prevents forgetting in continual learning of new tasks, and 3) enables efficient task transfer, offering a promising solution for advanced robotic applications. Demos and codes can be found in https://forrest-110.github.io/sparse_diffusion_policy/.

cs.RO↗

arXiv2024

Can MLLMs Understand the Deep Implication Behind Chinese Images?

As the capabilities of Multimodal Large Language Models (MLLMs) continue to improve, the need for higher-order capability evaluation of MLLMs is increasing. However, there is a lack of work evaluating MLLM for higher-order perception and understanding of Chinese visual content. To fill the gap, we introduce the **C**hinese **I**mage **I**mplication understanding **Bench**mark, **CII-Bench**, which aims to assess the higher-order perception and understanding capabilities of MLLMs for Chinese images. CII-Bench stands out in several ways compared to existing benchmarks. Firstly, to ensure the authenticity of the Chinese context, images in CII-Bench are sourced from the Chinese Internet and manually reviewed, with corresponding answers also manually crafted. Additionally, CII-Bench incorporates images that represent Chinese traditional culture, such as famous Chinese traditional paintings, which can deeply reflect the model's understanding of Chinese traditional culture. Through extensive experiments on CII-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on CII-Bench. The highest accuracy of MLLMs attains 64.4%, where as human accuracy averages 78.2%, peaking at an impressive 81.0%. Subsequently, MLLMs perform worse on Chinese traditional culture images, suggesting limitations in their ability to understand high-level semantics and lack a deep knowledge base of Chinese traditional culture. Finally, it is observed that most models exhibit enhanced accuracy when image emotion hints are incorporated into the prompts. We believe that CII-Bench will enable MLLMs to gain a better understanding of Chinese semantics and Chinese-specific images, advancing the journey towards expert artificial general intelligence (AGI). Our project is publicly available at https://cii-bench.github.io/.

cs.CL↗

arXiv2024

Dual-frame Fluid Motion Estimation with Test-time Optimization and Zero-divergence Loss

3D particle tracking velocimetry (PTV) is a key technique for analyzing turbulent flow, one of the most challenging computational problems of our century. At the core of 3D PTV is the dual-frame fluid motion estimation algorithm, which tracks particles across two consecutive frames. Recently, deep learning-based methods have achieved impressive accuracy in dual-frame fluid motion estimation; however, they heavily depend on large volumes of labeled data. In this paper, we introduce a new method that is completely self-supervised and notably outperforms its fully-supervised counterparts while requiring only 1% of the training samples (without labels) used by previous methods. Our method features a novel zero-divergence loss that is specific to the domain of turbulent flow. Inspired by the success of splat operation in high-dimensional filtering and random fields, we propose a splat-based implementation for this loss which is both efficient and effective. The self-supervised nature of our method naturally supports test-time optimization, leading to the development of a tailored Dynamic Velocimetry Enhancer (DVE) module. We demonstrate that strong cross-domain robustness is achieved through test-time optimization on unseen leave-one-out synthetic domains and real physical/biological domains. Code, data and models are available at https://github.com/Forrest-110/FluidMotionNet.

cs.CV↗

arXiv2024

Thermal conductivity of REBCO tapes with different stabilizers from 4.2 to 200 K

REBCO coated conductor is a high temperature superconductor that has a wide range of applications, one of which is the current leads of magnet systems. In the design of current leads, it is crucial to minimize their thermal conduction while maintain stable electrical conduction. Therefore, thermal conductivity of various REBCO tapes need to be characterized and analyzed. In this research, we measured thermal conductivity of REBCO tapes in the longitudinal direction from 4.2 to 200 K. Samples with Cu, Ag and Ag-3at%Au stabilizers of various thicknesses were measured. The electrical conductivity of these stabilizers was also characterized by residual resistance ratio (RRR) measurements and correlated with the thermal conductivity results. We showed that in samples with 10 micron or less Cu stabilizer, thermal conduction is dominated by that of the Cu which has much higher thermal conductivity than the Hastelloy substrate and the superconductor layer. In addition, the sample with 3 micron Ag-3at%Au stabilizer has significantly lower thermal conductivity than that with 3 micron silver stabilizer. It is concluded that REBCO with Ag-3at%Au stabilizer is promising for current lead applications.

cond-mat.supr-con↗

arXiv2024

Coalescence formation of muonic atoms at RHIC

The discovery of exotic mounic atoms, including antimatter hydrogen muonic atoms and kaon mounic atoms, constitutes a milestone in our ability to make and study new forms of matter. Relativistic heavy-ion collisions provide the only likely condition for production and detection of these exotic atoms. Taking a Coulomb correlations into account from the time of the fireball freeze-out until the formation of a stable atom has dramatic consequence on the expected yields of these atoms. When the coalescence model with the assumption of quantum wave function localization is applied to the formation of muonic atoms, we find that the atom yields are about two orders of magnitude higher than previously predicted.

hep-ph↗

arXiv2024

Minstrel: Structural Prompt Generation with Multi-Agents Coordination for Non-AI Experts

LLMs have demonstrated commendable performance across diverse domains. Nevertheless, formulating high-quality prompts to assist them in their work poses a challenge for non-AI experts. Existing research in prompt engineering suggests somewhat scattered optimization principles and designs empirically dependent prompt optimizers. Unfortunately, these endeavors lack a structural design, incurring high learning costs and it is not conducive to the iterative updating of prompts, especially for non-AI experts. Inspired by structured reusable programming languages, we propose LangGPT, a structural prompt design framework. Furthermore, we introduce Minstrel, a multi-generative agent system with reflection to automate the generation of structural prompts. Experiments and the case study illustrate that structural prompts generated by Minstrel or written manually significantly enhance the performance of LLMs. Furthermore, we analyze the ease of use of structural prompts through a user survey in our online community.

cs.CL↗

arXiv2024

Unravelling and circumventing failure mechanisms in chalcogenide optical phase change materials

Chalcogenide optical phase change materials (PCMs) have garnered significant interest for their growing applications in programmable photonics, optical analog computing, active metasurfaces, and beyond. Limited endurance or cycling lifetime is however increasingly becoming a bottleneck toward their practical deployment for these applications. To address this issue, we performed a systematic study elucidating the cycling failure mechanisms of Ge$_2$Sb$_2$Se$_4$Te (GSST), a common optical PCM tailored for infrared photonic applications, in an electrothermal switching configuration commensurate with their applications in on-chip photonic devices. We further propose a set of design rules building on insights into the failure mechanisms, and successfully implemented them to boost the endurance of the GSST device to over 67,000 cycles.

physics.optics↗

arXiv2024

An Attribute-Enriched Dataset and Auto-Annotated Pipeline for Open Detection

Detecting objects of interest through language often presents challenges, particularly with objects that are uncommon or complex to describe, due to perceptual discrepancies between automated models and human annotators. These challenges highlight the need for comprehensive datasets that go beyond standard object labels by incorporating detailed attribute descriptions. To address this need, we introduce the Objects365-Attr dataset, an extension of the existing Objects365 dataset, distinguished by its attribute annotations. This dataset reduces inconsistencies in object detection by integrating a broad spectrum of attributes, including color, material, state, texture and tone. It contains an extensive collection of 5.6M object-level attribute descriptions, meticulously annotated across 1.4M bounding boxes. Additionally, to validate the dataset's effectiveness, we conduct a rigorous evaluation of YOLO-World at different scales, measuring their detection performance and demonstrating the dataset's contribution to advancing object detection.

cs.CV↗

arXiv2024

Charm Sivers function at EicC

The Electron-Ion Collider in China (EicC) is pivotal in enhancing our knowledge of the internal structure of nucleons and nuclei, particularly through the study of transverse momentum-dependent parton distributions (TMDs). Among the leading-twist TMDs, the Sivers function is of particular interest, as it provides crucial insights into the spin and momentum structure of hadrons and plays a significant role in describing transverse single spin asymmetries (SSAs) in high-energy scatterings. In this study, we focus on the theoretical framework and phenomenological implications of the Sivers function in the context of small-x physics, where it is intricately connected to the spin-dependent QCD odderon, demonstrating that the SSA can be expressed in terms of transverse momentum-dependent factorization within the Color Glass Condensate effective theory. Furthermore, we present simulation results using PythiaeRHIC to assess the feasibility of measuring the charm quark Sivers function at EicC. The simulation outcomes suggest that EicC, with its unique kinematic coverage, offers distinct advantages for probing the Sivers function, which would provide compelling evidence for the existence of the elusive spin-dependent odderon.

hep-ph↗

arXiv2024

Can LLMs Beat Humans in Debating? A Dynamic Multi-agent Framework for Competitive Debate

Competitive debate is a complex task of computational argumentation. Large Language Models (LLMs) suffer from hallucinations and lack competitiveness in this field. To address these challenges, we introduce Agent for Debate (Agent4Debate), a dynamic multi-agent framework based on LLMs designed to enhance their capabilities in competitive debate. Drawing inspiration from human behavior in debate preparation and execution, Agent4Debate employs a collaborative architecture where four specialized agents, involving Searcher, Analyzer, Writer, and Reviewer, dynamically interact and cooperate. These agents work throughout the debate process, covering multiple stages from initial research and argument formulation to rebuttal and summary. To comprehensively evaluate framework performance, we construct the Competitive Debate Arena, comprising 66 carefully selected Chinese debate motions. We recruit ten experienced human debaters and collect records of 200 debates involving Agent4Debate, baseline models, and humans. The evaluation employs the Debatrix automatic scoring system and professional human reviewers based on the established Debatrix-Elo and Human-Elo ranking. Experimental results indicate that the state-of-the-art Agent4Debate exhibits capabilities comparable to those of humans. Furthermore, ablation studies demonstrate the effectiveness of each component in the agent structure.

cs.CL↗

arXiv2024

Hierarchical Retrieval-Augmented Generation Model with Rethink for Multi-hop Question Answering

Multi-hop Question Answering (QA) necessitates complex reasoning by integrating multiple pieces of information to resolve intricate questions. However, existing QA systems encounter challenges such as outdated information, context window length limitations, and an accuracy-quantity trade-off. To address these issues, we propose a novel framework, the Hierarchical Retrieval-Augmented Generation Model with Rethink (HiRAG), comprising Decomposer, Definer, Retriever, Filter, and Summarizer five key modules. We introduce a new hierarchical retrieval strategy that incorporates both sparse retrieval at the document level and dense retrieval at the chunk level, effectively integrating their strengths. Additionally, we propose a single-candidate retrieval method to mitigate the limitations of multi-candidate retrieval. We also construct two new corpora, Indexed Wikicorpus and Profile Wikicorpus, to address the issues of outdated and insufficient knowledge. Our experimental results on four datasets demonstrate that HiRAG outperforms state-of-the-art models across most metrics, and our Indexed Wikicorpus is effective. The code for HiRAG is available at https://github.com/2282588541a/HiRAG

cs.CL↗

arXiv2024

Enhancing Source Code Classification Effectiveness via Prompt Learning Incorporating Knowledge Features

Researchers have investigated the potential of leveraging pre-trained language models, such as CodeBERT, to enhance source code-related tasks. Previous methodologies have relied on CodeBERT's '[CLS]' token as the embedding representation of input sequences for task performance, necessitating additional neural network layers to enhance feature representation, which in turn increases computational expenses. These approaches have also failed to fully leverage the comprehensive knowledge inherent within the source code and its associated text, potentially limiting classification efficacy. We propose CodeClassPrompt, a text classification technique that harnesses prompt learning to extract rich knowledge associated with input sequences from pre-trained models, thereby eliminating the need for additional layers and lowering computational costs. By applying an attention mechanism, we synthesize multi-layered knowledge into task-specific features, enhancing classification accuracy. Our comprehensive experimentation across four distinct source code-related tasks reveals that CodeClassPrompt achieves competitive performance while significantly reducing computational overhead.

cs.CL↗

arXiv2024

Towards Effective Top-N Hamming Search via Bipartite Graph Contrastive Hashing

Searching on bipartite graphs serves as a fundamental task for various real-world applications, such as recommendation systems, database retrieval, and document querying. Conventional approaches rely on similarity matching in continuous Euclidean space of vectorized node embeddings. To handle intensive similarity computation efficiently, hashing techniques for graph-structured data have emerged as a prominent research direction. However, despite the retrieval efficiency in Hamming space, previous studies have encountered catastrophic performance decay. To address this challenge, we investigate the problem of hashing with Graph Convolutional Network for effective Top-N search. Our findings indicate the learning effectiveness of incorporating hashing techniques within the exploration of bipartite graph reception fields, as opposed to simply treating hashing as post-processing to output embeddings. To further enhance the model performance, we advance upon these findings and propose Bipartite Graph Contrastive Hashing (BGCH+). BGCH+ introduces a novel dual augmentation approach to both intermediate information and hash code outputs in the latent feature spaces, thereby producing more expressive and robust hash codes within a dual self-supervised learning paradigm. Comprehensive empirical analyses on six real-world benchmarks validate the effectiveness of our dual feature contrastive learning in boosting the performance of BGCH+ compared to existing approaches.

cs.IR↗

arXiv2024

ChatZero:Zero-shot Cross-Lingual Dialogue Generation via Pseudo-Target Language

Although large language models(LLMs) show amazing capabilities, among various exciting applications discovered for LLMs fall short in other low-resource languages. Besides, most existing methods depend on large-scale dialogue corpora and thus building systems for dialogue generation in a zero-shot scenario remains a considerable challenge. To address this challenge, we propose a novel end-to-end zero-shot dialogue generation model ChatZero based on cross-lingual code-switching method. First, we construct code-switching language and pseudo-target language with placeholders. Then for cross-lingual semantic transfer, we employ unsupervised contrastive learning to minimize the semantics gap of the source language, code-switching language, and pseudo-target language that are mutually positive examples in the high dimensional semantic space. Experiments on the multilingual DailyDialog and DSTC7-AVSD datasets demonstrate that ChatZero can achieve more than 90\% of the original performance under the zero-shot case compared to supervised learning, and achieve state-of-the-art performance compared with other baselines.

cs.CL↗

arXiv2024

Data-driven analysis of the beauty hadron production in p+p collisions at the LHC with Bayesian unfolding

Heavy flavour production in proton-proton (pp) collisions provides insights into the fundamental properties of Quantum Chromodynamics (QCD). Beauty hadron production measurements are widely performed through indirect approaches based on their inclusive decay modes. A Bayesian unfolding data-driven analysis of the ALICE and LHCb data was performed in this study, which recovers the full kinematic information of the beauty hadrons via different inclusive decay channels. The corresponding beauty hadron production cross sections obtained after the Bayesian unfolding are found to be consistent within their uncertainties. The weighted average open beauty production cross sections are presented as a function of the transverse momentum and rapidity in pp collisions at $\sqrt{s}$ = 5.02 TeV and $\sqrt{s}$ = 13 TeV, respectively. The $p_T$-integrated open beauty production $\mathrm{d}σ/\mathrm{d}y$ and the total $\mathrm{b}\rm\overline{b}$ cross section $σ_{\rm \mathrm{b}\rm\overline{b}}$ are also reported. The precision of these results significantly improves upon worldwide measurements, providing valuable validation and constraints on mechanisms of heavy flavour production in pp collisions at the LHC energies.

nucl-ex↗

arXiv2024

Distilling Large Language Models for Text-Attributed Graph Learning

Text-Attributed Graphs (TAGs) are graphs of connected textual documents. Graph models can efficiently learn TAGs, but their training heavily relies on human-annotated labels, which are scarce or even unavailable in many applications. Large language models (LLMs) have recently demonstrated remarkable capabilities in few-shot and zero-shot TAG learning, but they suffer from scalability, cost, and privacy issues. Therefore, in this work, we focus on synergizing LLMs and graph models with their complementary strengths by distilling the power of LLMs to a local graph model on TAG learning. To address the inherent gaps between LLMs (generative models for texts) and graph models (discriminative models for graphs), we propose first to let LLMs teach an interpreter with rich textual rationale and then let a student model mimic the interpreter's reasoning without LLMs' textual rationale. Extensive experiments validate the efficacy of our proposed framework.

cs.CL↗