arXiv subjects

Daoyuan Wu

Publications and source records attributed to Daoyuan Wu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Differentiation-Based Extraction of Proprietary Data from Fine-Tuned LLMs

The increasing demand for domain-specific and human-aligned Large Language Models (LLMs) has led to the widespread adoption of Supervised Fine-Tuning (SFT) techniques. SFT datasets often comprise valuable instruction-response pairs, making them highly valuable targets for potential extraction. This paper studies this critical research problem for the first time. We start by formally defining and formulating the problem, then explore various attack goals, types, and variants based on the unique properties of SFT data in real-world scenarios. Based on our analysis of extraction behaviors of direct extraction, we develop a novel extraction method specifically designed for SFT models, called Differentiated Data Extraction (DDE), which exploits the confidence levels of fine-tuned models and their behavioral differences from pre-trained base models. Through extensive experiments across multiple domains and scenarios, we demonstrate the feasibility of SFT data extraction using DDE. Our results show that DDE consistently outperforms existing extraction baselines in all attack settings. To counter this new attack, we propose a defense mechanism that mitigates DDE attacks with minimal impact on model performance. Overall, our research reveals hidden data leak risks in fine-tuned LLMs and provides insights for developing more secure models.

cs.CR↗

arXiv2025

IP Leakage Attacks Targeting LLM-Based Multi-Agent Systems

The rapid advancement of Large Language Models (LLMs) has led to the emergence of Multi-Agent Systems (MAS) to perform complex tasks through collaboration. However, the intricate nature of MAS, including their architecture and agent interactions, raises significant concerns regarding intellectual property (IP) protection. In this paper, we introduce MASLEAK, a novel attack framework designed to extract sensitive information from MAS applications. MASLEAK targets a practical, black-box setting, where the adversary has no prior knowledge of the MAS architecture or agent configurations. The adversary can only interact with the MAS through its public API, submitting attack query $q$ and observing outputs from the final agent. Inspired by how computer worms propagate and infect vulnerable network hosts, MASLEAK carefully crafts adversarial query $q$ to elicit, propagate, and retain responses from each MAS agent that reveal a full set of proprietary components, including the number of agents, system topology, system prompts, task instructions, and tool usages. We construct the first synthetic dataset of MAS applications with 810 applications and also evaluate MASLEAK against real-world MAS applications, including Coze and CrewAI. MASLEAK achieves high accuracy in extracting MAS IP, with an average attack success rate of 87% for system prompts and task instructions, and 92% for system architecture in most cases. We conclude by discussing the implications of our findings and the potential defenses.

cs.CR↗

arXiv2025

LLM4Vuln: A Unified Evaluation Framework for Decoupling and Enhancing LLMs' Vulnerability Reasoning

Large language models (LLMs) have demonstrated significant potential in various tasks, including those requiring human-level intelligence, such as vulnerability detection. However, recent efforts to use LLMs for vulnerability detection remain preliminary, as they lack a deep understanding of whether a subject LLM's vulnerability reasoning capability stems from the model itself or from external aids such as knowledge retrieval and tooling support. In this paper, we aim to decouple LLMs' vulnerability reasoning from other capabilities, such as vulnerability knowledge adoption, context information retrieval, and advanced prompt schemes. We introduce LLM4Vuln, a unified evaluation framework that separates and assesses LLMs' vulnerability reasoning capabilities and examines improvements when combined with other enhancements. To support this evaluation, we construct UniVul, the first benchmark that provides retrievable knowledge and context-supplementable code across three representative programming languages: Solidity, Java, and C/C++. Using LLM4Vuln and UniVul, we test six representative LLMs (GPT-4.1, Phi-3, Llama-3, o4-mini, DeepSeek-R1, and QwQ-32B) for 147 ground-truth vulnerabilities and 147 non-vulnerable cases in 3,528 controlled scenarios. Our findings reveal the varying impacts of knowledge enhancement, context supplementation, and prompt schemes. We also identify 14 zero-day vulnerabilities in four pilot bug bounty programs, resulting in $3,576 in bounties.

cs.CR↗

arXiv2025

BadMoE: Backdooring Mixture-of-Experts LLMs via Optimizing Routing Triggers and Infecting Dormant Experts

Mixture-of-Experts (MoE) have emerged as a powerful architecture for large language models (LLMs), enabling efficient scaling of model capacity while maintaining manageable computational costs. The key advantage lies in their ability to route different tokens to different ``expert'' networks within the model, enabling specialization and efficient handling of diverse input. However, the vulnerabilities of MoE-based LLMs still have barely been studied, and the potential for backdoor attacks in this context remains largely unexplored. This paper presents the first backdoor attack against MoE-based LLMs where the attackers poison ``dormant experts'' (i.e., underutilized experts) and activate them by optimizing routing triggers, thereby gaining control over the model's output. We first rigorously prove the existence of a few ``dominating experts'' in MoE models, whose outputs can determine the overall MoE's output. We also show that dormant experts can serve as dominating experts to manipulate model predictions. Accordingly, our attack, namely BadMoE, exploits the unique architecture of MoE models by 1) identifying dormant experts unrelated to the target task, 2) constructing a routing-aware loss to optimize the activation triggers of these experts, and 3) promoting dormant experts to dominating roles via poisoned training data. Extensive experiments show that BadMoE successfully enforces malicious prediction on attackers' target tasks while preserving overall model utility, making it a more potent and stealthy attack than existing methods.

cs.CR↗

arXiv2025

STShield: Single-Token Sentinel for Real-Time Jailbreak Detection in Large Language Models

Large Language Models (LLMs) have become increasingly vulnerable to jailbreak attacks that circumvent their safety mechanisms. While existing defense methods either suffer from adaptive attacks or require computationally expensive auxiliary models, we present STShield, a lightweight framework for real-time jailbroken judgement. STShield introduces a novel single-token sentinel mechanism that appends a binary safety indicator to the model's response sequence, leveraging the LLM's own alignment capabilities for detection. Our framework combines supervised fine-tuning on normal prompts with adversarial training using embedding-space perturbations, achieving robust detection while preserving model utility. Extensive experiments demonstrate that STShield successfully defends against various jailbreak attacks, while maintaining the model's performance on legitimate queries. Compared to existing approaches, STShield achieves superior defense performance with minimal computational overhead, making it a practical solution for real-world LLM deployment.

cs.CL↗

arXiv2025

SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner

Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs) and has evolved into multiple categories: human-based, optimization-based, generation-based, and the recent indirect and multilingual jailbreaks. However, delivering a practical jailbreak defense is challenging because it needs to not only handle all the above jailbreak attacks but also incur negligible delays to user prompts, as well as be compatible with both open-source and closed-source LLMs. Inspired by how the traditional security concept of shadow stacks defends against memory overflow attacks, this paper introduces a generic LLM jailbreak defense framework called SelfDefend, which establishes a shadow LLM as a defense instance (in detection state) to concurrently protect the target LLM instance (in normal answering state) in the normal stack and collaborate with it for checkpoint-based access control. The effectiveness of SelfDefend builds upon our observation that existing LLMs can identify harmful prompts or intentions in user queries, which we empirically validate using mainstream GPT-3.5/4 models against major jailbreak attacks. To further improve the defense's robustness and minimize costs, we employ a data distillation approach to tune dedicated open-source defense models. When deployed to protect GPT-3.5/4, Claude, Llama-2-7b/13b, and Mistral, these models outperform seven state-of-the-art defenses and match the performance of GPT-4-based SelfDefend, with significantly lower extra delays. Further experiments show that the tuned models are robust to adaptive jailbreaks and prompt injections.

cs.CR↗

arXiv2025

Low-Cost and Comprehensive Non-textual Input Fuzzing with LLM-Synthesized Input Generators

Modern software often accepts inputs with highly complex grammars. Recent advances in large language models (LLMs) have shown that they can be used to synthesize high-quality natural language text and code that conforms to the grammar of a given input format. Nevertheless, LLMs are often incapable or too costly to generate non-textual outputs, such as images, videos, and PDF files. This limitation hinders the application of LLMs in grammar-aware fuzzing. We present a novel approach to enabling grammar-aware fuzzing over non-textual inputs. We employ LLMs to synthesize and also mutate input generators, in the form of Python scripts, that generate data conforming to the grammar of a given input format. Then, non-textual data yielded by the input generators are further mutated by traditional fuzzers (AFL++) to explore the software input space effectively. Our approach, namely G2FUZZ, features a hybrid strategy that combines a holistic search driven by LLMs and a local search driven by industrial quality fuzzers. Two key advantages are: (1) LLMs are good at synthesizing and mutating input generators and enabling jumping out of local optima, thus achieving a synergistic effect when combined with mutation-based fuzzers; (2) LLMs are less frequently invoked unless really needed, thus significantly reducing the cost of LLM usage. We have evaluated G2FUZZ on a variety of input formats, including TIFF images, MP4 audios, and PDF files. The results show that G2FUZZ outperforms SOTA tools such as AFL++, Fuzztruction, and FormatFuzzer in terms of code coverage and bug finding across most programs tested on three platforms: UNIFUZZ, FuzzBench, and MAGMA.

cs.SE↗

arXiv2024

How Multi-Modal LLMs Reshape Visual Deep Learning Testing? A Comprehensive Study Through the Lens of Image Mutation

Visual deep learning (VDL) systems have shown significant success in real-world applications like image recognition, object detection, and autonomous driving. To evaluate the reliability of VDL, a mainstream approach is software testing, which requires diverse mutations over image semantics. The rapid development of multi-modal large language models (MLLMs) has introduced revolutionary image mutation potentials through instruction-driven methods. Users can now freely describe desired mutations and let MLLMs generate the mutated images. Hence, parallel to large language models' (LLMs) recent success in traditional software fuzzing, one may also expect MLLMs to be promising for VDL testing in terms of offering unified, diverse, and complex image mutations. However, the quality and applicability of MLLM-based mutations in VDL testing remain largely unexplored. We present the first study, aiming to assess MLLMs' adequacy from 1) the semantic validity of MLLM mutated images, 2) the alignment of MLLM mutated images with their text instructions (prompts), and 3) the faithfulness of how different mutations preserve semantics that are ought to remain unchanged. With large-scale human studies and quantitative evaluations, we identify MLLM's promising potentials in expanding the covered semantics of image mutations. Notably, while SoTA MLLMs (e.g., GPT-4V) fail to support or perform worse in editing existing semantics in images (as in traditional mutations like rotation), they generate high-quality test inputs using "semantic-replacement" mutations (e.g., "dress a dog with clothes"), which bring extra semantics to images; these were infeasible for past approaches. Hence, we view MLLM-based mutations as a vital complement to traditional mutations, and advocate future VDL testing tasks to combine MLLM-based methods and traditional image mutations for comprehensive and reliable testing.

cs.SE↗

arXiv2024

Split and Merge: Aligning Position Biases in LLM-based Evaluators

Large language models (LLMs) have shown promise as automated evaluators for assessing the quality of answers generated by AI systems. However, these LLM-based evaluators exhibit position bias, or inconsistency, when used to evaluate candidate answers in pairwise comparisons, favoring either the first or second answer regardless of content. To address this limitation, we propose PORTIA, an alignment-based system designed to mimic human comparison strategies to calibrate position bias in a lightweight yet effective manner. Specifically, PORTIA splits the answers into multiple segments, aligns similar content across candidate answers, and then merges them back into a single prompt for evaluation by LLMs. We conducted extensive experiments with six diverse LLMs to evaluate 11,520 answer pairs. Our results show that PORTIA markedly enhances the consistency rates for all the models and comparison forms tested, achieving an average relative improvement of 47.46%. Remarkably, PORTIA enables less advanced GPT models to achieve 88% agreement with the state-of-the-art GPT-4 model at just 10% of the cost. Furthermore, it rectifies around 80% of the position bias instances within the GPT-4 model, elevating its consistency rate up to 98%. Subsequent human evaluations indicate that the PORTIA-enhanced GPT-3.5 model can even surpass the standalone GPT-4 in terms of alignment with human evaluators. These findings highlight PORTIA's ability to correct position bias, improve LLM consistency, and boost performance while keeping cost-efficiency. This represents a valuable step toward a more reliable and scalable use of LLMs for automated evaluations across diverse applications.

cs.CL↗

arXiv2024

PropertyGPT: LLM-driven Formal Verification of Smart Contracts through Retrieval-Augmented Property Generation

With recent advances in large language models (LLMs), this paper explores the potential of leveraging state-of-the-art LLMs,such as GPT-4, to transfer existing human-written properties (e.g.,those from Certora auditing reports) and automatically generate customized properties for unknown code. To this end, we embed existing properties into a vector database and retrieve a reference property for LLM-based in-context learning to generate a new property for a given code. While this basic process is relatively straightforward, ensuring that the generated properties are (i) compilable, (ii) appropriate, and (iii) verifiable presents challenges. To address (i), we use the compilation and static analysis feedback as an external oracle to guide LLMs in iteratively revising the generated properties. For (ii), we consider multiple dimensions of similarity to rank the properties and employ a weighted algorithm to identify the top-K properties as the final result. For (iii), we design a dedicated prover to formally verify the correctness of the generated properties. We have implemented these strategies into a novel LLM-based property generation tool called PropertyGPT. Our experiments show that PropertyGPT can generate comprehensive and high-quality properties, achieving an 80% recall compared to the ground truth. It successfully detected 26 CVEs/attack incidents out of 37 tested and also uncovered 12 zero-day vulnerabilities, leading to $8,256 in bug bounty rewards.

cs.SE↗

arXiv2024

Machine Learning for Actionable Warning Identification: A Comprehensive Survey

Actionable Warning Identification (AWI) plays a crucial role in improving the usability of static code analyzers. With recent advances in Machine Learning (ML), various approaches have been proposed to incorporate ML techniques into AWI. These ML-based AWI approaches, benefiting from ML's strong ability to learn subtle and previously unseen patterns from historical data, have demonstrated superior performance. However, a comprehensive overview of these approaches is missing, which could hinder researchers/practitioners from understanding the current process and discovering potential for future improvement in the ML-based AWI community. In this paper, we systematically review the state-of-the-art ML-based AWI approaches. First, we employ a meticulous survey methodology and gather 51 primary studies from 2000/01/01 to 2023/09/01. Then, we outline the typical ML-based AWI workflow, including warning dataset preparation, preprocessing, AWI model construction, and evaluation stages. In such a workflow, we categorize ML-based AWI approaches based on the warning output format. Besides, we analyze the techniques used in each stage, along with their strengths, weaknesses, and distribution. Finally, we provide practical research directions for future ML-based AWI approaches, focusing on aspects like data improvement (e.g., enhancing the warning labeling strategy) and model exploration (e.g., exploring large language models for AWI).

cs.SE↗

arXiv2024

Combining Fine-Tuning and LLM-based Agents for Intuitive Smart Contract Auditing with Justifications

Smart contracts are decentralized applications built atop blockchains like Ethereum. Recent research has shown that large language models (LLMs) have potential in auditing smart contracts, but the state-of-the-art indicates that even GPT-4 can achieve only 30% precision (when both decision and justification are correct). This is likely because off-the-shelf LLMs were primarily pre-trained on a general text/code corpus and not fine-tuned on the specific domain of Solidity smart contract auditing. In this paper, we propose iAudit, a general framework that combines fine-tuning and LLM-based agents for intuitive smart contract auditing with justifications. Specifically, iAudit is inspired by the observation that expert human auditors first perceive what could be wrong and then perform a detailed analysis of the code to identify the cause. As such, iAudit employs a two-stage fine-tuning approach: it first tunes a Detector model to make decisions and then tunes a Reasoner model to generate causes of vulnerabilities. However, fine-tuning alone faces challenges in accurately identifying the optimal cause of a vulnerability. Therefore, we introduce two LLM-based agents, the Ranker and Critic, to iteratively select and debate the most suitable cause of vulnerability based on the output of the fine-tuned Reasoner model. To evaluate iAudit, we collected a balanced dataset with 1,734 positive and 1,810 negative samples to fine-tune iAudit. We then compared it with traditional fine-tuned models (CodeBERT, GraphCodeBERT, CodeT5, and UnixCoder) as well as prompt learning-based LLMs (GPT4, GPT-3.5, and CodeLlama-13b/34b). On a dataset of 263 real smart contract vulnerabilities, iAudit achieves an F1 score of 91.21% and an accuracy of 91.11%. The causes generated by iAudit achieved a consistency of about 38% compared to the ground truth causes.

cs.SE↗

arXiv2024

API-guided Dataset Synthesis to Finetune Large Code Models

Large code models (LCMs), pre-trained on vast code corpora, have demonstrated remarkable performance across a wide array of code-related tasks. Supervised fine-tuning (SFT) plays a vital role in aligning these models with specific requirements and enhancing their performance in particular domains. However, synthesizing high-quality SFT datasets poses a significant challenge due to the uneven quality of datasets and the scarcity of domain-specific datasets. Inspired by APIs as high-level abstractions of code that encapsulate rich semantic information in a concise structure, we propose DataScope, an API-guided dataset synthesis framework designed to enhance the SFT process for LCMs in both general and domain-specific scenarios. DataScope comprises two main components: Dsel and Dgen. On one hand, Dsel employs API coverage as a core metric, enabling efficient dataset synthesis in general scenarios by selecting subsets of existing (uneven-quality) datasets with higher API coverage. On the other hand, Dgen recasts domain dataset synthesis as a process of using API-specified high-level functionality and deliberately-constituted code skeletons to synthesize concrete code. Extensive experiments demonstrate DataScope's effectiveness, with models fine-tuned on its synthesized datasets outperforming those tuned on unoptimized datasets five times larger. Furthermore, a series of analyses on model internals, relevant hyperparameters, and case studies provide additional evidence for the efficacy of our proposed methods. These findings underscore the significance of dataset quality in SFT and advance the field of LCMs by providing an efficient, cost-effective framework for constructing high-quality datasets. This contribution enhances performance across both general and domain-specific scenarios, paving the way for more powerful and tailored LCMs.

cs.SE↗

arXiv2024

GPTScan: Detecting Logic Vulnerabilities in Smart Contracts by Combining GPT with Program Analysis

Smart contracts are prone to various vulnerabilities, leading to substantial financial losses over time. Current analysis tools mainly target vulnerabilities with fixed control or data-flow patterns, such as re-entrancy and integer overflow. However, a recent study on Web3 security bugs revealed that about 80% of these bugs cannot be audited by existing tools due to the lack of domain-specific property description and checking. Given recent advances in Large Language Models (LLMs), it is worth exploring how Generative Pre-training Transformer (GPT) could aid in detecting logicc vulnerabilities. In this paper, we propose GPTScan, the first tool combining GPT with static analysis for smart contract logic vulnerability detection. Instead of relying solely on GPT to identify vulnerabilities, which can lead to high false positives and is limited by GPT's pre-trained knowledge, we utilize GPT as a versatile code understanding tool. By breaking down each logic vulnerability type into scenarios and properties, GPTScan matches candidate vulnerabilities with GPT. To enhance accuracy, GPTScan further instructs GPT to intelligently recognize key variables and statements, which are then validated by static confirmation. Evaluation on diverse datasets with around 400 contract projects and 3K Solidity files shows that GPTScan achieves high precision (over 90%) for token contracts and acceptable precision (57.14%) for large projects like Web3Bugs. It effectively detects ground-truth logic vulnerabilities with a recall of over 70%, including 9 new vulnerabilities missed by human auditors. GPTScan is fast and cost-effective, taking an average of 14.39 seconds and 0.01 USD to scan per thousand lines of Solidity code. Moreover, static confirmation helps GPTScan reduce two-thirds of false positives.

cs.CR↗

arXiv2024

Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs

Agents based on large language models (LLMs) have demonstrated effectiveness in solving a wide range of tasks by integrating LLMs with key modules such as planning, memory, and tool usage. Increasingly, customers are adopting LLM agents across a variety of commercial applications critical to reliability, including support for mental well-being, chemical synthesis, and software development. Nevertheless, our observations and daily use of LLM agents indicate that they are prone to making erroneous plans, especially when the tasks are complex and require long-term planning. In this paper, we propose PDoctor, a novel and automated approach to testing LLM agents and understanding their erroneous planning. As the first work in this direction, we formulate the detection of erroneous planning as a constraint satisfiability problem: an LLM agent's plan is considered erroneous if its execution violates the constraints derived from the user inputs. To this end, PDoctor first defines a domain-specific language (DSL) for user queries and synthesizes varying inputs with the assistance of the Z3 constraint solver. These synthesized inputs are natural language paragraphs that specify the requirements for completing a series of tasks. Then, PDoctor derives constraints from these requirements to form a testing oracle. We evaluate PDoctor with three mainstream agent frameworks and two powerful LLMs (GPT-3.5 and GPT-4). The results show that PDoctor can effectively detect diverse errors in agent planning and provide insights and error characteristics that are valuable to both agent developers and users. We conclude by discussing potential alternative designs and directions to extend PDoctor.

cs.AI↗

arXiv2024

Pre-trained Model-based Actionable Warning Identification: A Feasibility Study

Actionable Warning Identification (AWI) plays a pivotal role in improving the usability of static code analyzers. Currently, Machine Learning (ML)-based AWI approaches, which mainly learn an AWI classifier from labeled warnings, are notably common. However, these approaches still face the problem of restricted performance due to the direct reliance on a limited number of labeled warnings to develop a classifier. Very recently, Pre-Trained Models (PTMs), which have been trained through billions of text/code tokens and demonstrated substantial success applications on various code-related tasks, could potentially circumvent the above problem. Nevertheless, the performance of PTMs on AWI has not been systematically investigated, leaving a gap in understanding their pros and cons. In this paper, we are the first to explore the feasibility of applying various PTMs for AWI. By conducting the extensive evaluation on 10K+ SpotBugs warnings from 10 large-scale and open-source projects, we observe that all studied PTMs are consistently 9.85%~21.12% better than the state-of-the-art ML-based AWI approaches. Besides, we investigate the impact of three primary aspects (i.e., data preprocessing, model training, and model prediction) in the typical PTM-based AWI workflow. Further, we identify the reasons for current PTMs' underperformance on AWI. Based on our findings, we provide several practical guidelines to enhance PTM-based AWI in future work.

cs.SE↗

arXiv2024

LLMs Can Defend Themselves Against Jailbreaking in a Practical Manner: A Vision Paper

Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs). A considerable amount of research exists proposing more effective jailbreak attacks, including the recent Greedy Coordinate Gradient (GCG) attack, jailbreak template-based attacks such as using "Do-Anything-Now" (DAN), and multilingual jailbreak. In contrast, the defensive side has been relatively less explored. This paper proposes a lightweight yet practical defense called SELFDEFEND, which can defend against all existing jailbreak attacks with minimal delay for jailbreak prompts and negligible delay for normal user prompts. Our key insight is that regardless of the kind of jailbreak strategies employed, they eventually need to include a harmful prompt (e.g., "how to make a bomb") in the prompt sent to LLMs, and we found that existing LLMs can effectively recognize such harmful prompts that violate their safety policies. Based on this insight, we design a shadow stack that concurrently checks whether a harmful prompt exists in the user prompt and triggers a checkpoint in the normal stack once a token of "No" or a harmful prompt is output. The latter could also generate an explainable LLM response to adversarial prompts. We demonstrate our idea of SELFDEFEND works in various jailbreak scenarios through manual analysis in GPT-3.5/4. We also list three future directions to further enhance SELFDEFEND.

cs.CR↗

arXiv2023

VRPTEST: Evaluating Visual Referring Prompting in Large Multimodal Models

With recent advancements in Large Multimodal Models (LMMs) across various domains, a novel prompting method called visual referring prompting has emerged, showing significant potential in enhancing human-computer interaction within multimodal systems. This method offers a more natural and flexible approach to human interaction with these systems compared to traditional text descriptions or coordinates. However, the categorization of visual referring prompting remains undefined, and its impact on the performance of LMMs has yet to be formally examined. In this study, we conduct the first comprehensive analysis of LMMs using a variety of visual referring prompting strategies. We introduce a benchmark dataset called VRPTEST, comprising 3 different visual tasks and 2,275 images, spanning diverse combinations of prompt strategies. Using VRPTEST, we conduct a comprehensive evaluation of eight versions of prominent open-source and proprietary foundation models, including two early versions of GPT-4V. We develop an automated assessment framework based on software metamorphic testing techniques to evaluate the accuracy of LMMs without the need for human intervention or manual labeling. We find that the current proprietary models generally outperform the open-source ones, showing an average accuracy improvement of 22.70%; however, there is still potential for improvement. Moreover, our quantitative analysis shows that the choice of prompt strategy significantly affects the accuracy of LMMs, with variations ranging from -17.5% to +7.3%. Further case studies indicate that an appropriate visual referring prompting strategy can improve LMMs' understanding of context and location information, while an unsuitable one might lead to answer rejection. We also provide insights on minimizing the negative impact of visual referring prompting on LMMs.

cs.CV↗