Prompt engineering has made significant contributions to the era of large language models, yet its effectiveness depends on the skills of a prompt author. Automatic prompt optimization can support the prompt development process, but requires annotated data. This paper introduces $\textit{iPrOp}$, a novel Interactive Prompt Optimization system, to bridge manual prompt engineering and automatic prompt optimization. With human intervention in the optimization loop, $\textit{iPrOp}$ offers users the flexibility to assess evolving prompts. We present users with prompt variations, selected instances, large language model predictions accompanied by corresponding explanations, and performance metrics derived from a subset of the training data. This approach empowers users to choose and further refine the provided prompts based on their individual preferences and needs. This system not only assists non-technical domain experts in generating optimal prompts tailored to their specific tasks or domains, but also enables to study the intrinsic parameters that influence the performance of prompt optimization. Our evaluation shows that our system has the capability to generate improved prompts, leading to enhanced task performance.
Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language processing tasks, but their vulnerability to jailbreak attacks poses significant security risks. This survey paper presents a comprehensive analysis of recent advancements in attack strategies and defense mechanisms within the field of Large Language Model (LLM) red-teaming. We analyze various attack methods, including gradient-based optimization, reinforcement learning, and prompt engineering approaches. We discuss the implications of these attacks on LLM safety and the need for improved defense mechanisms. This work aims to provide a thorough understanding of the current landscape of red-teaming attacks and defenses on LLMs, enabling the development of more secure and reliable language models.
This paper proposes a novel interdisciplinary framework for the critical evaluation of text-to-image models, addressing the limitations of current technical metrics and bias studies. By integrating art historical analysis, artistic exploration, and critical prompt engineering, the framework offers a more nuanced understanding of these models' capabilities and societal implications. Art historical analysis provides a structured approach to examine visual and symbolic elements, revealing potential biases and misrepresentations. Artistic exploration, through creative experimentation, uncovers hidden potentials and limitations, prompting critical reflection on the algorithms' assumptions. Critical prompt engineering actively challenges the model's assumptions, exposing embedded biases. Case studies demonstrate the framework's practical application, showcasing how it can reveal biases related to gender, race, and cultural representation. This comprehensive approach not only enhances the evaluation of text-to-image models but also contributes to the development of more equitable, responsible, and culturally aware AI systems.
We present RAG Playground, an open-source framework for systematic evaluation of Retrieval-Augmented Generation (RAG) systems. The framework implements and compares three retrieval approaches: naive vector search, reranking, and hybrid vector-keyword search, combined with ReAct agents using different prompting strategies. We introduce a comprehensive evaluation framework with novel metrics and provide empirical results comparing different language models (Llama 3.1 and Qwen 2.5) across various retrieval configurations. Our experiments demonstrate significant performance improvements through hybrid search methods and structured self-evaluation prompting, achieving up to 72.7% pass rate on our multi-metric evaluation framework. The results also highlight the importance of prompt engineering in RAG systems, with our custom-prompted agents showing consistent improvements in retrieval accuracy and response quality.
We conducted three experiments to investigate how large language models (LLMs) evaluate posterior probabilities. Our results reveal the coexistence of two modes in posterior judgment among state-of-the-art models: a normative mode, which adheres to Bayes' rule, and a representative-based mode, which relies on similarity -- paralleling human System 1 and System 2 thinking. Additionally, we observed that LLMs struggle to recall base rate information from their memory, and developing prompt engineering strategies to mitigate representative-based judgment may be challenging. We further conjecture that the dual modes of judgment may be a result of the contrastive loss function employed in reinforcement learning from human feedback. Our findings underscore the potential direction for reducing cognitive biases in LLMs and the necessity for cautious deployment of LLMs in critical areas.
Few-shot segmentation is the problem of learning to identify specific types of objects (e.g., airplanes) in images from a small set of labeled reference images. The current state of the art is driven by resource-intensive construction of models for every new domain-specific application. Such models must be trained on enormous labeled datasets of unrelated objects (e.g., cars, trains, animals) so that their ``knowledge'' can be transferred to new types of objects. In this paper, we show how to leverage existing vision foundation models (VFMs) to reduce the incremental cost of creating few-shot segmentation models for new domains. Specifically, we introduce SAMIC, a small network that learns how to prompt VFMs in order to segment new types of objects in domain-specific applications. SAMIC enables any task to be approached as a few-shot learning problem. At 2.6 million parameters, it is 94% smaller than the leading models (e.g., having ResNet 101 backbone with 45+ million parameters). Even using 1/5th of the training data provided by one-shot benchmarks, SAMIC is competitive with, or sets the state of the art, on a variety of few-shot and semantic segmentation datasets including COCO-$20^i$, Pascal-$5^i$, PerSeg, FSS-1000, and NWPU VHR-10.
Caregivers (i.e., parents and members of a child's caring community) are underappreciated stakeholders in learning analytics. Although caregiver involvement can enhance student academic outcomes, many obstacles hinder involvement, most notably knowledge gaps with respect to modern school curricula. An emerging topic of interest in learning analytics is hybrid tutoring, which includes instructional and motivational support. Caregivers assert similar roles in homework, yet it is unknown how learning analytics can support them. Our past work with caregivers suggested that conversational support is a promising method of providing caregivers with the guidance needed to effectively support student learning. We developed a system that provides instructional support to caregivers through conversational recommendations generated by a Large Language Model (LLM). Addressing known instructional limitations of LLMs, we use instructional intelligence from tutoring systems while conducting prompt engineering experiments with the open-source Llama 3 LLM. This LLM generated message recommendations for caregivers supporting their child's math practice via chat. Few-shot prompting and combining real-time problem-solving context from tutoring systems with examples of tutoring practices yielded desirable message recommendations. These recommendations were evaluated with ten middle school caregivers, who valued recommendations facilitating content-level support and student metacognition through self-explanation. We contribute insights into how tutoring systems can best be merged with LLMs to support hybrid tutoring settings through conversational assistance, facilitating effective caregiver involvement in tutoring systems.
Extracting sentence embeddings from large language models (LLMs) is a promising direction, as LLMs have demonstrated stronger semantic understanding capabilities. Previous studies typically focus on prompt engineering to elicit sentence embeddings from LLMs by prompting the model to encode sentence information into the embedding of the last token. However, LLMs are mostly decoder-only models with causal attention and the earlier tokens in the sentence cannot attend to the latter tokens, resulting in biased encoding of sentence information and cascading effects on the final decoded token. To this end, we propose a novel Token Prepending (TP) technique that prepends each layer's decoded sentence embedding to the beginning of the sentence in the next layer's input, allowing earlier tokens to attend to the complete sentence information under the causal attention mechanism. The proposed TP technique is a plug-and-play and training-free technique, which means it can be seamlessly integrated with various prompt-based sentence embedding methods and autoregressive LLMs. Extensive experiments on various Semantic Textual Similarity (STS) tasks and downstream classification tasks demonstrate that our proposed TP technique can significantly improve the performance of existing prompt-based sentence embedding methods across different LLMs, while incurring negligible additional inference cost.
The capacity of LLMs to carry out automated qualitative analysis has been questioned by corpus linguists, and it has been argued that corpus-based discourse analysis incorporating LLMs is hindered by issues of unsatisfying performance, hallucination, and irreproducibility. Our proposed method, TACOMORE, aims to address these concerns by serving as an effective prompting framework in this domain. The framework consists of four principles, i.e., Task, Context, Model and Reproducibility, and specifies five fundamental elements of a good prompt, i.e., Role Description, Task Definition, Task Procedures, Contextual Information and Output Format. We conduct experiments on three LLMs, i.e., GPT-4o, Gemini-1.5-Pro and Gemini-1.5.Flash, and find that TACOMORE helps improve LLM performance in three representative discourse analysis tasks, i.e., the analysis of keywords, collocates and concordances, based on an open corpus of COVID-19 research articles. Our findings show the efficacy of the proposed prompting framework TACOMORE in corpus-based discourse analysis in terms of Accuracy, Ethicality, Reasoning, and Reproducibility, and provide novel insights into the application and evaluation of LLMs in automated qualitative studies.
Large language models (LLMs) have demonstrated remarkable performance across a wide range of tasks. Advances in prompt engineering and fine-tuning techniques have further enhanced their ability to address complex reasoning challenges. However, these advanced capabilities are often exclusive to models exceeding 100 billion parameters. Although Chain-of-Thought (CoT) fine-tuning methods have been explored for smaller models (under 10 billion parameters), they typically depend on extensive CoT training data, which can introduce inconsistencies and limit effectiveness in low-data settings. To overcome these limitations, this paper introduce a new reasoning strategy Solution Guidance (SG) and a plug-and-play training paradigm Solution-Guidance Fine-Tuning (SGFT) for enhancing the reasoning capabilities of small language models. SG focuses on problem understanding and decomposition at the semantic and logical levels, rather than specific computations, which can effectively improve the SLMs' generalization and reasoning abilities. With only a small amount of SG training data, SGFT can fine-tune a SLM to produce accurate problem-solving guidances, which can then be flexibly fed to any SLM as prompts, enabling it to generate correct answers directly. Experimental results demonstrate that our method significantly improves the performance of SLMs on various reasoning tasks, enhancing both their practicality and efficiency within resource-constrained environments.
The effectiveness of large language models (LLMs) is closely tied to the design of prompts, making prompt optimization essential for enhancing their performance across a wide range of tasks. Many existing approaches to automating prompt engineering rely exclusively on textual feedback, refining prompts based solely on inference errors identified by large, computationally expensive LLMs. Unfortunately, smaller models struggle to generate high-quality feedback, resulting in complete dependence on large LLM judgment. Moreover, these methods fail to leverage more direct and finer-grained information, such as gradients, due to operating purely in text space. To this end, we introduce GReaTer, a novel prompt optimization technique that directly incorporates gradient information over task-specific reasoning. By utilizing task loss gradients, GReaTer enables self-optimization of prompts for open-source, lightweight language models without the need for costly closed-source LLMs. This allows high-performance prompt optimization without dependence on massive LLMs, closing the gap between smaller models and the sophisticated reasoning often needed for prompt refinement. Extensive evaluations across diverse reasoning tasks including BBH, GSM8k, and FOLIO demonstrate that GReaTer consistently outperforms previous state-of-the-art prompt optimization methods, even those reliant on powerful LLMs. Additionally, GReaTer-optimized prompts frequently exhibit better transferability and, in some cases, boost task performance to levels comparable to or surpassing those achieved by larger language models, highlighting the effectiveness of prompt optimization guided by gradients over reasoning. Code of GReaTer is available at https://github.com/psunlpgroup/GreaTer.
Understanding and extracting the grammar of a domain-specific language (DSL) is crucial for various software engineering tasks; however, manually creating these grammars is time-intensive and error-prone. This paper presents Kajal, a novel approach that automatically infers grammar from DSL code snippets by leveraging Large Language Models (LLMs) through prompt engineering and few-shot learning. Kajal dynamically constructs input prompts, using contextual information to guide the LLM in generating the corresponding grammars, which are iteratively refined through a feedback-driven approach. Our experiments show that Kajal achieves 60% accuracy with few-shot learning and 45% without it, demonstrating the significant impact of few-shot learning on the tool's effectiveness. This approach offers a promising solution for automating DSL grammar extraction, and future work will explore using smaller, open-source LLMs and testing on larger datasets to further validate Kajal's performance.
Ensuring that Software Requirements Specifications (SRS) align with higher-level organizational or national requirements is vital, particularly in regulated environments such as finance and aerospace. In these domains, maintaining consistency, adhering to regulatory frameworks, minimizing errors, and meeting critical expectations are essential for the reliable functioning of systems. The widespread adoption of large language models (LLMs) highlights their immense potential, yet there remains considerable scope for improvement in retrieving relevant information and enhancing reasoning capabilities. This study demonstrates that integrating a robust Graph-RAG framework with advanced prompt engineering techniques, such as Chain of Thought and Tree of Thought, can significantly enhance performance. Compared to baseline RAG methods and simple prompting strategies, this approach delivers more accurate and context-aware results. While this method demonstrates significant improvements in performance, it comes with challenges. It is both costly and more complex to implement across diverse contexts, requiring careful adaptation to specific scenarios. Additionally, its effectiveness heavily relies on having complete and accurate input data, which may not always be readily available, posing further limitations to its scalability and practicality.
Ontologies of research topics are crucial for structuring scientific knowledge, enabling scientists to navigate vast amounts of research, and forming the backbone of intelligent systems such as search engines and recommendation systems. However, manual creation of these ontologies is expensive, slow, and often results in outdated and overly general representations. As a solution, researchers have been investigating ways to automate or semi-automate the process of generating these ontologies. This paper offers a comprehensive analysis of the ability of large language models (LLMs) to identify semantic relationships between different research topics, which is a critical step in the development of such ontologies. To this end, we developed a gold standard based on the IEEE Thesaurus to evaluate the task of identifying four types of relationships between pairs of topics: broader, narrower, same-as, and other. Our study evaluates the performance of seventeen LLMs, which differ in scale, accessibility (open vs. proprietary), and model type (full vs. quantised), while also assessing four zero-shot reasoning strategies. Several models have achieved outstanding results, including Mixtral-8x7B, Dolphin-Mistral-7B, and Claude 3 Sonnet, with F1-scores of 0.847, 0.920, and 0.967, respectively. Furthermore, our findings demonstrate that smaller, quantised models, when optimised through prompt engineering, can deliver performance comparable to much larger proprietary models, while requiring significantly fewer computational resources.
Due to the remarkable generative potential of diffusion-based models, numerous researches have investigated jailbreak attacks targeting these frameworks. A particularly concerning threat within image models is the generation of Not-Safe-for-Work (NSFW) content. Despite the implementation of security filters, numerous efforts continue to explore ways to circumvent these safeguards. Current attack methodologies primarily encompass adversarial prompt engineering or concept obfuscation, yet they frequently suffer from slow search efficiency, conspicuous attack characteristics and poor alignment with targets. To overcome these challenges, we propose Antelope, a more robust and covert jailbreak attack strategy designed to expose security vulnerabilities inherent in generative models. Specifically, Antelope leverages the confusion of sensitive concepts with similar ones, facilitates searches in the semantically adjacent space of these related concepts and aligns them with the target imagery, thereby generating sensitive images that are consistent with the target and capable of evading detection. Besides, we successfully exploit the transferability of model-based attacks to penetrate online black-box services. Experimental evaluations demonstrate that Antelope outperforms existing baselines across multiple defensive mechanisms, underscoring its efficacy and versatility.
Physical adversarial attacks in driving scenarios can expose critical vulnerabilities in visual perception models. However, developing such attacks remains challenging due to diverse real-world backgrounds and the requirement for maintaining visual naturality. Building upon this challenge, we reformulate physical adversarial attacks as a one-shot patch-generation problem. Our approach generates adversarial patches through a deep generative model that considers the specific scene context, enabling direct physical deployment in matching environments. The primary challenge lies in simultaneously achieving two objectives: generating adversarial patches that effectively mislead object detection systems while determining contextually appropriate placement within the scene. We propose MAGIC (Mastering Physical Adversarial Generation In Context), a novel framework powered by multi-modal LLM agents to address these challenges. MAGIC automatically understands scene context and orchestrates adversarial patch generation through the synergistic interaction of language and vision capabilities. MAGIC orchestrates three specialized LLM agents: The adv-patch generation agent (GAgent) masters the creation of deceptive patches through strategic prompt engineering for text-to-image models. The adv-patch deployment agent (DAgent) ensures contextual coherence by determining optimal placement strategies based on scene understanding. The self-examination agent (EAgent) completes this trilogy by providing critical oversight and iterative refinement of both processes. We validate our method on both digital and physical level, \ie, nuImage and manually captured real scenes, where both statistical and visual results prove that our MAGIC is powerful and effectively for attacking wide-used object detection systems.
The rapid advancement of Generative AI (Gen AI) technologies, particularly tools like ChatGPT, is significantly impacting the labor market by reshaping job roles and skill requirements. This study examines the demand for ChatGPT-related skills in the U.S. labor market by analyzing job advertisements collected from major job platforms between May and December 2023. Using text mining and topic modeling techniques, we extracted and analyzed the Gen AI-related skills that employers are hiring for. Our analysis identified five distinct ChatGPT-related skill sets: general familiarity, creative content generation, marketing, advanced functionalities (such as prompt engineering), and product development. In addition, the study provides insights into job attributes such as occupation titles, degree requirements, salary ranges, and other relevant job characteristics. These findings highlight the increasing integration of Gen AI across various industries, emphasizing the growing need for both foundational knowledge and advanced technical skills. The study offers valuable insights into the evolving demands of the labor market, as employers seek candidates equipped to leverage generative AI tools to improve productivity, streamline processes, and drive innovation.
Recent advances in decision-making policies have led to significant progress in fields such as autonomous driving and robotics. However, testing these policies remains crucial with the existence of critical scenarios that may threaten their reliability. Despite ongoing research, challenges such as low testing efficiency and limited diversity persist due to the complexity of the decision-making policies and their environments. To address these challenges, this paper proposes an adaptable Large Language Model (LLM)-driven online testing framework to explore critical and diverse testing scenarios for decision-making policies. Specifically, we design a "generate-test-feedback" pipeline with templated prompt engineering to harness the world knowledge and reasoning abilities of LLMs. Additionally, a multi-scale scenario generation strategy is proposed to address the limitations of LLMs in making fine-grained adjustments, further enhancing testing efficiency. Finally, the proposed LLM-driven method is evaluated on five widely recognized benchmarks, and the experimental results demonstrate that our method significantly outperforms baseline methods in uncovering both critical and diverse scenarios. These findings suggest that LLM-driven methods hold significant promise for advancing the testing of decision-making policies.
Understanding the vulnerabilities of Large Vision Language Models (LVLMs) to jailbreak attacks is essential for their responsible real-world deployment. Most previous work requires access to model gradients, or is based on human knowledge (prompt engineering) to complete jailbreak, and they hardly consider the interaction of images and text, resulting in inability to jailbreak in black box scenarios or poor performance. To overcome these limitations, we propose a Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for toxicity maximization, referred to as PBI-Attack. Our method begins by extracting malicious features from a harmful corpus using an alternative LVLM and embedding these features into a benign image as prior information. Subsequently, we enhance these features through bidirectional cross-modal interaction optimization, which iteratively optimizes the bimodal perturbations in an alternating manner through greedy search, aiming to maximize the toxicity of the generated response. The toxicity level is quantified using a well-trained evaluation model. Experiments demonstrate that PBI-Attack outperforms previous state-of-the-art jailbreak methods, achieving an average attack success rate of 92.5% across three open-source LVLMs and around 67.3% on three closed-source LVLMs. Disclaimer: This paper contains potentially disturbing and offensive content.
Large Multimodal Models (LMMs) have demonstrated impressive capabilities in multimodal understanding and generation, pushing forward advancements in text-to-image generation. However, achieving accurate text-image alignment for LMMs, particularly in compositional scenarios, remains challenging. Existing approaches, such as layout planning for multi-step generation and learning from human feedback or AI feedback, depend heavily on prompt engineering, costly human annotations, and continual upgrading, limiting flexibility and scalability. In this work, we introduce a model-agnostic iterative self-improvement framework (SILMM) that can enable LMMs to provide helpful and scalable self-feedback and optimize text-image alignment via Direct Preference Optimization (DPO). DPO can readily applied to LMMs that use discrete visual tokens as intermediate image representations; while it is less suitable for LMMs with continuous visual features, as obtaining generation probabilities is challenging. To adapt SILMM to LMMs with continuous features, we propose a diversity mechanism to obtain diverse representations and a kernel-based continuous DPO for alignment. Extensive experiments on three compositional text-to-image generation benchmarks validate the effectiveness and superiority of SILMM, showing improvements exceeding 30% on T2I-CompBench++ and around 20% on DPG-Bench.