arXiv subjects

Zhiyuan Peng

Publications and source records attributed to Zhiyuan Peng.

At least 37 records · Page 2Linked to original sources

arXiv2025

Efficiency-Effectiveness Reranking FLOPs for LLM-based Rerankers

Large Language Models (LLMs) have recently been applied to reranking tasks in information retrieval, achieving strong performance. However, their high computational demands often hinder practical deployment. Existing studies evaluate the efficiency of LLM-based rerankers using proxy metrics such as latency, the number of forward passes, input tokens, and output tokens. However, these metrics depend on hardware and running-time choices (\eg parallel or not, batch size, etc), and often fail to account for model size, making it difficult to interpret and obscuring the evaluation of the efficiency-effectiveness tradeoff. To address this issue, we propose \ours\footnote{https://github.com/zhiyuanpeng/EER-FLOPs.} for LLM-based rerankers: RPP (ranking metrics per PetaFLOP), measuring how much ranking quality (e.g., NDCG or MRR) a method achieves per PetaFLOP, and QPP (queries per PetaFLOP), measuring how many queries can be processed per PetaFLOP. Accompanied by the new metrics, an interpretable FLOPs estimator is developed to estimate the FLOPs of an LLM-based reranker even without running any experiments. Based on the proposed metrics, we conduct comprehensive experiments to evaluate a wide range of LLM-based rerankers with different architectures, studying the efficiency-effectiveness trade-off and bringing this issue to the attention of the research community.

cs.CL↗

arXiv2025

A Preference-Driven Methodology for High-Quality Solidity Code Generation

While Large Language Models (LLMs) have demonstrated remarkable progress in generating functionally correct Solidity code, they continue to face critical challenges in producing gas-efficient and secure code, which are critical requirements for real-world smart contract deployment. Although recent advances leverage Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) for code preference alignment, existing approaches treat functional correctness, gas optimization, and security as independent objectives, resulting in contracts that may achieve operational soundness but suffer from prohibitive execution costs or dangerous vulnerabilities. To address these limitations, we propose PrefGen, a novel framework that extends standard DPO beyond human preferences to incorporate quantifiable blockchain-specific metrics, enabling holistic multi-objective optimization specifically tailored for smart contract generation. Our framework introduces a comprehensive evaluation methodology with four complementary metrics: Pass@k (functional correctness), Compile@k (syntactic correctness), Gas@k (gas efficiency), and Secure@k (security assessment), providing rigorous multi-dimensional contract evaluation. Through extensive experimentation, we demonstrate that PrefGen significantly outperforms existing approaches across all critical dimensions, achieving 66.7% Pass@5, 58.9% Gas@5, and 62.5% Secure@5, while generating production-ready smart contracts that are functionally correct, cost-efficient, and secure.

cs.SE↗

arXiv2025

Digital Network Twins for Next-generation Wireless: Creation, Optimization, and Challenges

Digital network twins (DNTs), by representing a physical network using a virtual model, offer significant benefits such as streamlined network development, enhanced productivity, and cost reduction for next-generation (nextG) communication infrastructure. Existing works mainly describe the deployment of DNT technologies in various service sections.The full life cycle of DNTs for telecommunication has not yet been comprehensively studied, particularly in the aspects of fine-grained creation, real-time adaptation, resource-efficient deployment, and security protection. This article presents an in-depth overview of DNTs, exploring their concrete integration into networks and communication, covering the fundamental designs, the emergent applications, and critical challenges in multiple dimensions. We also include two detailed case studies to illustrate how DNTs can be applied in real-world scenarios such as wireless traffic forecasting and edge caching. Additionally, a forward-looking vision of the research opportunities in tackling the challenges of DNTs is provided, aiming to fully maximize the benefits of DNTs in nextG networks.

cs.NI↗

arXiv2025

SolEval: Benchmarking Large Language Models for Repository-level Solidity Code Generation

Large language models (LLMs) have transformed code generation. However, most existing approaches focus on mainstream languages such as Python and Java, neglecting the Solidity language, the predominant programming language for Ethereum smart contracts. Due to the lack of adequate benchmarks for Solidity, LLMs' ability to generate secure, cost-effective smart contracts remains unexplored. To fill this gap, we construct SolEval, the first repository-level benchmark designed for Solidity smart contract generation, to evaluate the performance of LLMs on Solidity. SolEval consists of 1,507 samples from 28 different repositories, covering 6 popular domains, providing LLMs with a comprehensive evaluation benchmark. Unlike the existing Solidity benchmark, SolEval not only includes complex function calls but also reflects the real-world complexity of the Ethereum ecosystem by incorporating Gas@k and Vul@k. We evaluate 16 LLMs on SolEval, and our results show that the best-performing LLM achieves only 26.29% Pass@10, highlighting substantial room for improvement in Solidity code generation by LLMs. Additionally, we conduct supervised fine-tuning (SFT) on Qwen-7B using SolEval, resulting in a significant performance improvement, with Pass@5 increasing from 16.67% to 58.33%, demonstrating the effectiveness of fine-tuning LLMs on our benchmark. We release our data and code at https://github.com/pzy2000/SolEval.

cs.SE↗

arXiv2025

RepoTransAgent: Multi-Agent LLM Framework for Repository-Aware Code Translation

Repository-aware code translation is critical for modernizing legacy systems, enhancing maintainability, and enabling interoperability across diverse programming languages. While recent advances in large language models (LLMs) have improved code translation quality, existing approaches face significant challenges in practical scenarios: insufficient contextual understanding, inflexible prompt designs, and inadequate error correction mechanisms. These limitations severely hinder accurate and efficient translation of complex, real-world code repositories. To address these challenges, we propose RepoTransAgent, a novel multi-agent LLM framework for repository-aware code translation. RepoTransAgent systematically decomposes the translation process into specialized subtasks-context retrieval, dynamic prompt construction, and iterative code refinement-each handled by dedicated agents. Our approach leverages retrieval-augmented generation (RAG) for contextual information gathering, employs adaptive prompts tailored to varying repository scenarios, and introduces a reflection-based mechanism for systematic error correction. We evaluate RepoTransAgent on hundreds of Java-C# translation pairs from six popular open-source projects. Experimental results demonstrate that RepoTransAgent significantly outperforms state-of-the-art baselines in both compile and pass rates. Specifically, RepoTransAgent achieves up to 55.34% compile rate and 45.84% pass rate. Comprehensive analysis confirms the robustness and generalizability of RepoTransAgent across different LLMs, establishing its effectiveness for real-world repository-aware code translation.

cs.SE↗

arXiv2025

Deep Learning Framework Testing via Model Mutation: How Far Are We?

Deep Learning (DL) frameworks are a fundamental component of DL development. Therefore, the detection of DL framework defects is important and challenging. As one of the most widely adopted DL testing techniques, model mutation has recently gained significant attention. In this study, we revisit the defect detection ability of existing mutation-based testing methods and investigate the factors that influence their effectiveness. To begin with, we reviewed existing methods and observed that many of them mutate DL models (e.g., changing their parameters) without any customization, ignoring the unique challenges in framework testing. Another issue with these methods is their limited effectiveness, characterized by a high rate of false positives caused by illegal mutations arising from the use of generic, non-customized mutation operators. Moreover, we tracked the defects identified by these methods and discovered that most of them were ignored by developers. Motivated by these observations, we investigate the effectiveness of existing mutation-based testing methods in detecting important defects that have been authenticated by framework developers. We begin by collecting defect reports from three popular frameworks and classifying them based on framework developers' ratings to build a comprehensive dataset. We then perform an in-depth analysis to uncover valuable insights. Based on our findings, we propose optimization strategies to address the shortcomings of existing approaches. Following these optimizations, we identified seven new defects, four of which were confirmed by developers as high-priority issues, with three resolved. In summary, we identified 39 unique defects across just 23 models, of which 31 were confirmed by developers, and eight have been fixed.

cs.SE↗

arXiv2025

ELOQ: Resources for Enhancing LLM Detection of Out-of-Scope Questions

Retrieval-augmented generation (RAG) has become integral to large language models (LLMs), particularly for conversational AI systems where user questions may reference knowledge beyond the LLMs' training cutoff. However, many natural user questions lack well-defined answers, either due to limited domain knowledge or because the retrieval system returns documents that are relevant in appearance but uninformative in content. In such cases, LLMs often produce hallucinated answers without flagging them. While recent work has largely focused on questions with false premises, we study out-of-scope questions, where the retrieved document appears semantically similar to the question but lacks the necessary information to answer it. In this paper, we propose a guided hallucination-based approach ELOQ to automatically generate a diverse set of out-of-scope questions from post-cutoff documents, followed by human verification to ensure quality. We use this dataset to evaluate several LLMs on their ability to detect out-of-scope questions and generate appropriate responses. Finally, we introduce an improved detection method that enhances the reliability of LLM-based question-answering systems in handling out-of-scope questions.

cs.CL↗

arXiv2025

W-RAG: Weakly Supervised Dense Retrieval in RAG for Open-domain Question Answering

In knowledge-intensive tasks such as open-domain question answering (OpenQA), large language models (LLMs) often struggle to generate factual answers, relying solely on their internal (parametric) knowledge. To address this limitation, Retrieval-Augmented Generation (RAG) systems enhance LLMs by retrieving relevant information from external sources, thereby positioning the retriever as a pivotal component. Although dense retrieval demonstrates state-of-the-art performance, its training poses challenges due to the scarcity of ground-truth evidence, largely attributed to the high costs of human annotation. In this paper, we propose W-RAG, a method that draws weak training signals from the downstream task (such as OpenQA) of an LLM, and fine-tunes the retriever to prioritize passages that most benefit the task. Specifically, we rerank the top-$k$ passages retrieved via BM25 by assessing the probability that the LLM will generate the correct answer for a question given each passage. The highest-ranking passages are then used as positive fine-tuning examples for dense retrieval. We conduct comprehensive experiments across four publicly available OpenQA datasets to demonstrate that our approach enhances both retrieval and OpenQA performance compared to baseline models, achieving results comparable to models fine-tuned with human-labeled data.

cs.CL↗

arXiv2025

Rank-Based Modeling for Universal Packets Compression in Multi-Modal Communications

The rapid increase in networked systems and data transmission requires advanced data compression solutions to optimize bandwidth utilization and enhance network performance. This study introduces a novel byte-level predictive model using Transformer architecture, capable of handling the redundancy and diversity of data types in network traffic as byte sequences. Unlike traditional methods that require separate compressors for different data types, this unified approach sets new benchmarks and simplifies predictive modeling across various data modalities such as video, audio, images, and text, by processing them at the byte level. This is achieved by predicting subsequent byte probability distributions, encoding them into a sparse rank sequence using lossless entropy coding, and significantly reducing both data size and entropy. Experimental results show that our model achieves compression ratios below 50%, while offering models of various sizes tailored for different communication devices. Additionally, we successfully deploy these models on a range of edge devices and servers, demonstrating their practical applicability and effectiveness in real-world network scenarios. This approach significantly enhances data throughput and reduces bandwidth demands, making it particularly valuable in resource-constrained environments like the Internet of Things sensor networks.

cs.NI↗

arXiv2025

CLAPSep: Leveraging Contrastive Pre-trained Model for Multi-Modal Query-Conditioned Target Sound Extraction

Universal sound separation (USS) aims to extract arbitrary types of sounds from real-world recordings. This can be achieved by language-queried target sound extraction (TSE), which typically consists of two components: a query network that converts user queries into conditional embeddings, and a separation network that extracts the target sound accordingly. Existing methods commonly train models from scratch. As a consequence, substantial data and computational resources are required to make the randomly initialized model comprehend sound events and perform separation accordingly. In this paper, we propose to integrate pre-trained models into TSE models to address the above issue. To be specific, we tailor and adapt the powerful contrastive language-audio pre-trained model (CLAP) for USS, denoted as CLAPSep. CLAPSep also accepts flexible user inputs, taking both positive and negative user prompts of uni- and/or multi-modalities for target sound extraction. These key features of CLAPSep can not only enhance the extraction performance but also improve the versatility of its application. We provide extensive experiments on 5 diverse datasets to demonstrate the superior performance and zero- and few-shot generalizability of our proposed CLAPSep with fast training convergence, surpassing previous methods by a significant margin. Full codes and some audio examples are released for reproduction and evaluation.

eess.AS↗

arXiv2025

Language-Queried Target Sound Extraction Without Parallel Training Data

Language-queried target sound extraction (TSE) aims to extract specific sounds from mixtures based on language queries. Traditional fully-supervised training schemes require extensively annotated parallel audio-text data, which are labor-intensive. We introduce a parallel-data-free training scheme, requiring only unlabelled audio clips for TSE model training by utilizing the contrastive language-audio pre-trained model (CLAP). In a vanilla parallel-data-free training stage, target audio is encoded using the pre-trained CLAP audio encoder to form a condition embedding, while during testing, user language queries are encoded by CLAP text encoder as the condition embedding. This vanilla approach assumes perfect alignment between text and audio embeddings, which is unrealistic. Two major challenges arise from training-testing mismatch: the persistent modality gap between text and audio and the risk of overfitting due to the exposure of rich acoustic details in target audio embedding during training. To address this, we propose a retrieval-augmented strategy. Specifically, we create an embedding cache using audio captions generated by a large language model (LLM). During training, target audio embeddings retrieve text embeddings from this cache to use as condition embeddings, ensuring consistent modalities between training and testing and eliminating information leakage. Extensive experiment results show that our retrieval-augmented approach achieves consistent and notable performance improvements over existing state-of-the-art with better generalizability.

eess.AS↗

arXiv2025

MulChain: Enabling Advanced Cross-Modal Queries in Hybrid-Storage Blockchains

With its decentralization and immutability, blockchain has emerged as a trusted foundation for data management and querying. Because blockchain storage space is limited, large multimodal data files, such as videos, are often stored offline, leaving only lightweight metadata on the chain. While this hybrid storage approach enhances storage efficiency, it introduces significant challenges for executing advanced queries on multimodal data. The metadata stored on-chain is often minimal and may not include all the attributes necessary for queries like time range or fuzzy queries. In addition, existing blockchains do not provide native support for multimodal data querying. Achieving this capability would necessitate extensive modifications to the underlying blockchain framework, even reconstructing its core architecture. Consequently, enabling blockchains with multimodal query capabilities remains a significant problem, which necessitates overcoming the following three key challenges: (1) Designing efficient indexing methods to adapt to varying workloads that involve frequent insertions and query operations; (2) Achieving seamless integration with existing blockchains without altering the underlying infrastructure; (3) Ensuring high query performance while minimizing gas consumption. To address these challenges, we propose MulChain, a novel middleware architecture to enable smooth integration with existing blockchains. At the core of MulChain is the BHashTree, a flexible data structure that dynamically switches between tree and hash nodes based on workload characteristics, ensuring efficient insertion and query operations. Furthermore, the middleware provides standardized interfaces for blockchain systems, unifying query methods across different platforms.

cs.DB↗

arXiv2025

Denoising Diffusion Probabilistic Model for Radio Map Estimation in Generative Wireless Networks

The increasing demand for high-speed and reliable wireless networks has driven advancements in technologies such as millimeter-wave and 5G radios, which requires efficient planning and timely deployment of wireless access points. A critical tool in this process is the radio map, a graphical representation of radio-frequency signal strengths that plays a vital role in optimizing overall network performance. However, existing methods for estimating radio maps face challenges due to the need for extensive real-world data collection or computationally intensive ray-tracing analyses, which is costly and time-consuming. Inspired by the success of generative AI techniques in large language models and image generation, we explore their potential applications in the realm of wireless networks. In this work, we propose RM-Gen, a novel generative framework leveraging conditional denoising diffusion probabilistic models to synthesize radio maps using minimal and readily collected data. We then introduce an environment-aware method for selecting critical data pieces, enhancing the generative model's applicability and usability. Comprehensive evaluations demonstrate that RM-Gen achieves over 95% accuracy in generating radio maps for networks that operate at 60 GHz and sub-6GHz frequency bands, outperforming the baseline GAN and pix2pix models. This approach offers a cost-effective, adaptable solution for various downstream network optimization tasks.

cs.NI↗

arXiv2024

Evaluating and Enhancing Large Language Models for Novelty Assessment in Scholarly Publications

Recent studies have evaluated the creativity/novelty of large language models (LLMs) primarily from a semantic perspective, using benchmarks from cognitive science. However, accessing the novelty in scholarly publications is a largely unexplored area in evaluating LLMs. In this paper, we introduce a scholarly novelty benchmark (SchNovel) to evaluate LLMs' ability to assess novelty in scholarly papers. SchNovel consists of 15000 pairs of papers across six fields sampled from the arXiv dataset with publication dates spanning 2 to 10 years apart. In each pair, the more recently published paper is assumed to be more novel. Additionally, we propose RAG-Novelty, which simulates the review process taken by human reviewers by leveraging the retrieval of similar papers to assess novelty. Extensive experiments provide insights into the capabilities of different LLMs to assess novelty and demonstrate that RAG-Novelty outperforms recent baseline models.

cs.CL↗

arXiv2024

Digital Twin-Assisted Data-Driven Optimization for Reliable Edge Caching in Wireless Networks

Optimizing edge caching is crucial for the advancement of next-generation (nextG) wireless networks, ensuring high-speed and low-latency services for mobile users. Existing data-driven optimization approaches often lack awareness of the distribution of random data variables and focus solely on optimizing cache hit rates, neglecting potential reliability concerns, such as base station overload and unbalanced cache issues. This oversight can result in system crashes and degraded user experience. To bridge this gap, we introduce a novel digital twin-assisted optimization framework, called D-REC, which integrates reinforcement learning (RL) with diverse intervention modules to ensure reliable caching in nextG wireless networks. We first develop a joint vertical and horizontal twinning approach to efficiently create network digital twins, which are then employed by D-REC as RL optimizers and safeguards, providing ample datasets for training and predictive evaluation of our cache replacement policy. By incorporating reliability modules into a constrained Markov decision process, D-REC can adaptively adjust actions, rewards, and states to comply with advantageous constraints, minimizing the risk of network failures. Theoretical analysis demonstrates comparable convergence rates between D-REC and vanilla data-driven methods without compromising caching performance. Extensive experiments validate that D-REC outperforms conventional approaches in cache hit rate and load balancing while effectively enforcing predetermined reliability intervention modules.

cs.NI↗

arXiv2024

Passage-specific Prompt Tuning for Passage Reranking in Question Answering with Large Language Models

Effective passage retrieval and reranking methods have been widely utilized to identify suitable candidates in open-domain question answering tasks, recent studies have resorted to LLMs for reranking the retrieved passages by the log-likelihood of the question conditioned on each passage. Although these methods have demonstrated promising results, the performance is notably sensitive to the human-written prompt (or hard prompt), and fine-tuning LLMs can be computationally intensive and time-consuming. Furthermore, this approach limits the leverage of question-passage relevance pairs and passage-specific knowledge to enhance the ranking capabilities of LLMs. In this paper, we propose passage-specific prompt tuning for reranking in open-domain question answering (PSPT): a parameter-efficient method that fine-tunes learnable passage-specific soft prompts, incorporating passage-specific knowledge from a limited set of question-passage relevance pairs. The method involves ranking retrieved passages based on the log-likelihood of the model generating the question conditioned on each passage and the learned soft prompt. We conducted extensive experiments utilizing the Llama-2-chat-7B model across three publicly available open-domain question answering datasets and the results demonstrate the effectiveness of the proposed approach.

cs.CL↗

arXiv2024

Soft Prompt Tuning for Augmenting Dense Retrieval with Large Language Models

Dense retrieval (DR) converts queries and documents into dense embeddings and measures the similarity between queries and documents in vector space. One of the challenges in DR is the lack of domain-specific training data. While DR models can learn from large-scale public datasets like MS MARCO through transfer learning, evidence shows that not all DR models and domains can benefit from transfer learning equally. Recently, some researchers have resorted to large language models (LLMs) to improve the zero-shot and few-shot DR models. However, the hard prompts or human-written prompts utilized in these works cannot guarantee the good quality of generated weak queries. To tackle this, we propose soft prompt tuning for augmenting DR (SPTAR): For each task, we leverage soft prompt-tuning to optimize a task-specific soft prompt on limited ground truth data and then prompt the LLMs to tag unlabeled documents with weak queries, yielding enough weak document-query pairs to train task-specific dense retrievers. We design a filter to select high-quality example document-query pairs in the prompt to further improve the quality of weak tagged queries. To the best of our knowledge, there is no prior work utilizing soft prompt tuning to augment DR models. The experiments demonstrate that SPTAR outperforms the unsupervised baselines BM25 and the recently proposed LLMs-based augmentation method for DR.

cs.IR↗

arXiv2024

Q-PEFT: Query-dependent Parameter Efficient Fine-tuning for Text Reranking with Large Language Models

Parameter Efficient Fine-Tuning (PEFT) methods have been extensively utilized in Large Language Models (LLMs) to improve the down-streaming tasks without the cost of fine-tuing the whole LLMs. Recent studies have shown how to effectively use PEFT for fine-tuning LLMs in ranking tasks with convincing performance; there are some limitations, including the learned prompt being fixed for different documents, overfitting to specific tasks, and low adaptation ability. In this paper, we introduce a query-dependent parameter efficient fine-tuning (Q-PEFT) approach for text reranking to leak the information of the true queries to LLMs and then make the generation of true queries from input documents much easier. Specifically, we utilize the query to extract the top-$k$ tokens from concatenated documents, serving as contextual clues. We further augment Q-PEFT by substituting the retrieval mechanism with a multi-head attention layer to achieve end-to-end training and cover all the tokens in the documents, guiding the LLMs to generate more document-specific synthetic queries, thereby further improving the reranking performance. Extensive experiments are conducted on four public datasets, demonstrating the effectiveness of our proposed approach.

cs.CL↗