Numerous capability and safety techniques of Large Language Models (LLMs), including RLHF, automated red-teaming, prompt engineering, and infilling, can be cast as sampling from an unnormalized target distribution defined by a given reward or potential function over the full sequence. In this work, we leverage the rich toolkit of Sequential Monte Carlo (SMC) for these probabilistic inference problems. In particular, we use learned twist functions to estimate the expected future value of the potential at each timestep, which enables us to focus inference-time computation on promising partial sequences. We propose a novel contrastive method for learning the twist functions, and establish connections with the rich literature of soft reinforcement learning. As a complementary application of our twisted SMC framework, we present methods for evaluating the accuracy of language model inference techniques using novel bidirectional SMC bounds on the log partition function. These bounds can be used to estimate the KL divergence between the inference and target distributions in both directions. We apply our inference evaluation techniques to show that twisted SMC is effective for sampling undesirable outputs from a pretrained model (a useful component of harmlessness training and automated red-teaming), generating reviews with varied sentiment, and performing infilling tasks.
In this paper, we present PRISM, a Promptable and Robust Interactive Segmentation Model, aiming for precise segmentation of 3D medical images. PRISM accepts various visual inputs, including points, boxes, and scribbles as sparse prompts, as well as masks as dense prompts. Specifically, PRISM is designed with four principles to achieve robustness: (1) Iterative learning. The model produces segmentations by using visual prompts from previous iterations to achieve progressive improvement. (2) Confidence learning. PRISM employs multiple segmentation heads per input image, each generating a continuous map and a confidence score to optimize predictions. (3) Corrective learning. Following each segmentation iteration, PRISM employs a shallow corrective refinement network to reassign mislabeled voxels. (4) Hybrid design. PRISM integrates hybrid encoders to better capture both the local and global information. Comprehensive validation of PRISM is conducted using four public datasets for tumor segmentation in the colon, pancreas, liver, and kidney, highlighting challenges caused by anatomical variations and ambiguous boundaries in accurate tumor identification. Compared to state-of-the-art methods, both with and without prompt engineering, PRISM significantly improves performance, achieving results that are close to human levels. The code is publicly available at https://github.com/MedICL-VU/PRISM.
This paper presents a study on the integration of domain-specific knowledge in prompt engineering to enhance the performance of large language models (LLMs) in scientific domains. A benchmark dataset is curated to encapsulate the intricate physical-chemical properties of small molecules, their drugability for pharmacology, alongside the functional attributes of enzymes and crystal materials, underscoring the relevance and applicability across biological and chemical domains.The proposed domain-knowledge embedded prompt engineering method outperforms traditional prompt engineering strategies on various metrics, including capability, accuracy, F1 score, and hallucination drop. The effectiveness of the method is demonstrated through case studies on complex materials including the MacMillan catalyst, paclitaxel, and lithium cobalt oxide. The results suggest that domain-knowledge prompts can guide LLMs to generate more accurate and relevant responses, highlighting the potential of LLMs as powerful tools for scientific discovery and innovation when equipped with domain-specific prompts. The study also discusses limitations and future directions for domain-specific prompt engineering development.
In the burgeoning field of Large Language Models (LLMs) like ChatGPT and LLaMA, Prompt Engineering (PE) is renowned for boosting zero-shot or in-context learning (ICL) through prompt modifications. Yet, the realm of the sample design for downstream fine-tuning, crucial for task-specific LLM adaptation, is largely unexplored. This paper introduces Sample Design Engineering (SDE), a methodical approach to enhancing LLMs' post-tuning performance by refining input, output, and reasoning designs. We conduct a series of in-domain (ID) and out-of-domain (OOD) experiments to assess the impact of various design options on LLMs' downstream performance, revealing several intriguing patterns that hold consistently across different LLMs. Based on these insights, we propose an integrated SDE strategy, combining the most effective options, and validate its consistent superiority over heuristic sample designs in complex downstream tasks like multi-aspect sentiment analysis, event extraction, and nested entity recognition. Additionally, analyses of LLMs' inherent prompt/output perplexity, zero-shot, and ICL abilities illustrate that good PE strategies may not always translate to good SDE strategies. Code available at https://github.com/beyondguo/LLM-Tuning.
In this paper, we focus on the challenging task of reliably estimating factual knowledge that is embedded inside large language models (LLMs). To avoid reliability concerns with prior approaches, we propose to eliminate prompt engineering when probing LLMs for factual knowledge. Our approach, called Zero-Prompt Latent Knowledge Estimator (ZP-LKE), leverages the in-context learning ability of LLMs to communicate both the factual knowledge question as well as the expected answer format. Our knowledge estimator is both conceptually simpler (i.e., doesn't depend on meta-linguistic judgments of LLMs) and easier to apply (i.e., is not LLM-specific), and we demonstrate that it can surface more of the latent knowledge embedded in LLMs. We also investigate how different design choices affect the performance of ZP-LKE. Using the proposed estimator, we perform a large-scale evaluation of the factual knowledge of a variety of open-source LLMs, like OPT, Pythia, Llama(2), Mistral, Gemma, etc. over a large set of relations and facts from the Wikidata knowledge base. We observe differences in the factual knowledge between different model families and models of different sizes, that some relations are consistently better known than others but that models differ in the precise facts they know, and differences in the knowledge of base models and their finetuned counterparts. Code available at: https://github.com/QinyuanWu0710/ZeroPrompt_LKE
In this paper, we aim to generate text classification data given arbitrary class definitions (i.e., user instruction), so one can train a small text classifier without any human annotation or raw corpus. Compared with pioneer attempts, our proposed Incubator is the first framework that can handle complicated and even mutually dependent classes (e.g., "TED Talk given by Educator" and "Other"). Specifically, Incubator is an LLM firstly tuned on the instruction-to-data mappings that we obtained from classification datasets and descriptions on HuggingFace together with in-context augmentation by GPT-4. We then refine Incubator by learning on the cluster centers of semantic textual embeddings to emphasize the uniformity and semantic diversity in generations. We compare Incubator on various classification tasks with strong baselines such as direct LLM-based inference and training data generation by prompt engineering. Experiments show Incubator is able to (1) perform well on traditional benchmarks, (2) take label dependency and user preference into consideration, and (3) enable logical text mining by incubating multiple classifiers.
Event Coreference Resolution (ECR) as a pairwise mention classification task is expensive both for automated systems and manual annotations. The task's quadratic difficulty is exacerbated when using Large Language Models (LLMs), making prompt engineering for ECR prohibitively costly. In this work, we propose a graphical representation of events, X-AMR, anchored around individual mentions using a \textbf{cross}-document version of \textbf{A}bstract \textbf{M}eaning \textbf{R}epresentation. We then linearize the ECR with a novel multi-hop coreference algorithm over the event graphs. The event graphs simplify ECR, making it a) LLM cost-effective, b) compositional and interpretable, and c) easily annotated. For a fair assessment, we first enrich an existing ECR benchmark dataset with these event graphs using an annotator-friendly tool we introduce. Then, we employ GPT-4, the newest LLM by OpenAI, for these annotations. Finally, using the ECR algorithm, we assess GPT-4 against humans and analyze its limitations. Through this research, we aim to advance the state-of-the-art for efficient ECR and shed light on the potential shortcomings of current LLMs at this task. Code and annotations: \url{https://github.com/ahmeshaf/gpt_coref}
The Segment Anything Model (SAM) and CLIP are remarkable vision foundation models (VFMs). SAM, a prompt driven segmentation model, excels in segmentation tasks across diverse domains, while CLIP is renowned for its zero shot recognition capabilities. However, their unified potential has not yet been explored in medical image segmentation. To adapt SAM to medical imaging, existing methods primarily rely on tuning strategies that require extensive data or prior prompts tailored to the specific task, making it particularly challenging when only a limited number of data samples are available. This work presents an in depth exploration of integrating SAM and CLIP into a unified framework for medical image segmentation. Specifically, we propose a simple unified framework, SaLIP, for organ segmentation. Initially, SAM is used for part based segmentation within the image, followed by CLIP to retrieve the mask corresponding to the region of interest (ROI) from the pool of SAM generated masks. Finally, SAM is prompted by the retrieved ROI to segment a specific organ. Thus, SaLIP is training and fine tuning free and does not rely on domain expertise or labeled data for prompt engineering. Our method shows substantial enhancements in zero shot segmentation, showcasing notable improvements in DICE scores across diverse segmentation tasks like brain (63.46%), lung (50.11%), and fetal head (30.82%), when compared to un prompted SAM. Code and text prompts are available at: https://github.com/aleemsidra/SaLIP.
The vision-language model has brought great improvement to few-shot industrial anomaly detection, which usually needs to design of hundreds of prompts through prompt engineering. For automated scenarios, we first use conventional prompt learning with many-class paradigm as the baseline to automatically learn prompts but found that it can not work well in one-class anomaly detection. To address the above problem, this paper proposes a one-class prompt learning method for few-shot anomaly detection, termed PromptAD. First, we propose semantic concatenation which can transpose normal prompts into anomaly prompts by concatenating normal prompts with anomaly suffixes, thus constructing a large number of negative samples used to guide prompt learning in one-class setting. Furthermore, to mitigate the training challenge caused by the absence of anomaly images, we introduce the concept of explicit anomaly margin, which is used to explicitly control the margin between normal prompt features and anomaly prompt features through a hyper-parameter. For image-level/pixel-level anomaly detection, PromptAD achieves first place in 11/12 few-shot settings on MVTec and VisA.
Sentence Embedding stands as a fundamental task within the realm of Natural Language Processing, finding extensive application in search engines, expert systems, and question-and-answer platforms. With the continuous evolution of large language models such as LLaMA and Mistral, research on sentence embedding has recently achieved notable breakthroughs. However, these advancements mainly pertain to fine-tuning scenarios, leaving explorations into computationally efficient direct inference methods for sentence representation in a nascent stage. This paper endeavors to bridge this research gap. Through comprehensive experimentation, we challenge the widely held belief in the necessity of an Explicit One-word Limitation for deriving sentence embeddings from Pre-trained Language Models (PLMs). We demonstrate that this approach, while beneficial for generative models under direct inference scenario, is not imperative for discriminative models or the fine-tuning of generative PLMs. This discovery sheds new light on the design of manual templates in future studies. Building upon this insight, we propose two innovative prompt engineering techniques capable of further enhancing the expressive power of PLMs' raw embeddings: Pretended Chain of Thought and Knowledge Enhancement. We confirm their effectiveness across various PLM types and provide a detailed exploration of the underlying factors contributing to their success.
Inspired by human cognition, Jiang et al.(2023c) create a benchmark for assessing LLMs' lateral thinking-thinking outside the box. Building upon this benchmark, we investigate how different prompting methods enhance LLMs' performance on this task to reveal their inherent power for outside-the-box thinking ability. Through participating in SemEval-2024, task 9, Sentence Puzzle sub-task, we explore prompt engineering methods: chain of thoughts (CoT) and direct prompting, enhancing with informative descriptions, and employing contextualizing prompts using a retrieval augmented generation (RAG) pipeline. Our experiments involve three LLMs including GPT-3.5, GPT-4, and Zephyr-7B-beta. We generate a dataset of thinking paths between riddles and options using GPT-4, validated by humans for quality. Findings indicate that compressed informative prompts enhance performance. Dynamic in-context learning enhances model performance significantly. Furthermore, fine-tuning Zephyr on our dataset enhances performance across other commonsense datasets, underscoring the value of innovative thinking.
Pre-trained vision-language models have shown impressive success on various computer vision tasks with their zero-shot generalizability. Recently, prompt learning approaches have been explored to efficiently and effectively adapt the vision-language models to a variety of downstream tasks. However, most existing prompt learning methods suffer from task overfitting since the general knowledge of the pre-trained vision language models is forgotten while the prompts are finetuned on a small data set from a specific target task. To address this issue, we propose a Prompt Meta-Regularization (ProMetaR) to improve the generalizability of prompt learning for vision-language models. Specifically, ProMetaR meta-learns both the regularizer and the soft prompts to harness the task-specific knowledge from the downstream tasks and task-agnostic general knowledge from the vision-language models. Further, ProMetaR augments the task to generate multiple virtual tasks to alleviate the meta-overfitting. In addition, we provide the analysis to comprehend how ProMetaR improves the generalizability of prompt tuning in the perspective of the gradient alignment. Our extensive experiments demonstrate that our ProMetaR improves the generalizability of conventional prompt learning methods under base-to-base/base-to-new and domain generalization settings. The code of ProMetaR is available at https://github.com/mlvlab/ProMetaR.
Recent trends are emerging in the use of Large Language Models (LLMs) as autonomous agents that take actions based on the content of the user text prompts. We intend to apply these concepts to the field of Guidance, Navigation, and Control in space, enabling LLMs to have a significant role in the decision-making process for autonomous satellite operations. As a first step towards this goal, we have developed a pure LLM-based solution for the Kerbal Space Program Differential Games (KSPDG) challenge, a public software design competition where participants create autonomous agents for maneuvering satellites involved in non-cooperative space operations, running on the KSP game engine. Our approach leverages prompt engineering, few-shot prompting, and fine-tuning techniques to create an effective LLM-based agent that ranked 2nd in the competition. To the best of our knowledge, this work pioneers the integration of LLM agents into space research. Code is available at https://github.com/ARCLab-MIT/kspdg.
Code translation between programming languages is a long-existing and critical task in software engineering, facilitating the modernization of legacy systems, ensuring cross-platform compatibility, and enhancing software performance. With the recent advances in large language models (LLMs) and their applications to code translation, there is an increasing need for comprehensive evaluation of these models. In this study, we empirically analyze the generated outputs of eleven popular instruct-tuned LLMs with parameters ranging from 1B up to 46.7B on 3,820 translation pairs across five languages, including C, C++, Go, Java, and Python. Our analysis found that between 26.4% and 73.7% of code translations produced by our evaluated LLMs necessitate post-processing, as these translations often include a mix of code, quotes, and text rather than being purely source code. Overlooking the output format of these models can inadvertently lead to underestimation of their actual performance. This is particularly evident when evaluating them with execution-based metrics such as Computational Accuracy (CA). Our results demonstrate that a strategic combination of prompt engineering and regular expression can effectively extract the source code from the model generation output. In particular, our method can help eleven selected models achieve an average Code Extraction Success Rate (CSR) of 92.73%. Our findings shed light on and motivate future research to conduct more reliable benchmarks of LLMs for code translation.
Crafting effective topic models for brief texts, like tweets and news headlines, is essential for capturing the swift shifts in social dynamics. Traditional topic models, however, often fall short in accurately representing the semantic intricacies of short texts due to their brevity and lack of contextual data. In our study, we harness the advanced capabilities of Large Language Models (LLMs) to introduce a novel approach termed "Topic Refinement". This approach does not directly involve itself in the initial modeling of topics but focuses on improving topics after they have been mined. By employing prompt engineering, we direct LLMs to eliminate off-topic words within a given topic, ensuring that only contextually relevant words are preserved or substituted with ones that fit better semantically. This method emulates human-like scrutiny and improvement of topics, thereby elevating the semantic quality of the topics generated by various models. Our comprehensive evaluation across three unique datasets has shown that our topic refinement approach significantly enhances the semantic coherence of topics.
This study evaluates the performance of ChatGPT variants, GPT-3.5 and GPT-4, both with and without prompt engineering, against solely student work and a mixed category containing both student and GPT-4 contributions in university-level physics coding assignments using the Python language. Comparing 50 student submissions to 50 AI-generated submissions across different categories, and marked blindly by three independent markers, we amassed $n = 300$ data points. Students averaged 91.9% (SE:0.4), surpassing the highest performing AI submission category, GPT-4 with prompt engineering, which scored 81.1% (SE:0.8) - a statistically significant difference (p = $2.482 \times 10^{-10}$). Prompt engineering significantly improved scores for both GPT-4 (p = $1.661 \times 10^{-4}$) and GPT-3.5 (p = $4.967 \times 10^{-9}$). Additionally, the blinded markers were tasked with guessing the authorship of the submissions on a four-point Likert scale from `Definitely AI' to `Definitely Human'. They accurately identified the authorship, with 92.1% of the work categorized as 'Definitely Human' being human-authored. Simplifying this to a binary `AI' or `Human' categorization resulted in an average accuracy rate of 85.3%. These findings suggest that while AI-generated work closely approaches the quality of university students' work, it often remains detectable by human evaluators.
In this paper, we borrow the large language model (LLM) ChatGPT-3.5 to automatically and quickly design a new metaheuristic algorithm (MA) with only a small amount of input. The novel animal-inspired MA named zoological search optimization (ZSO) draws inspiration from the collective behaviors of animals for solving continuous optimization problems. Specifically, the basic ZSO algorithm involves two search operators: the prey-predator interaction operator and the social flocking operator to balance exploration and exploitation well. Besides, the standard prompt engineering framework CRISPE (i.e., Capacity and Role, Insight, Statement, Personality, and Experiment) is responsible for the specific prompt design. Furthermore, we designed four variants of the ZSO algorithm with slight human-interacted adjustment. In numerical experiments, we comprehensively investigate the performance of ZSO-derived algorithms on CEC2014 benchmark functions, CEC2022 benchmark functions, and six engineering optimization problems. 20 popular and state-of-the-art MAs are employed as competitors. The experimental results and statistical analysis confirm the efficiency and effectiveness of ZSO-derived algorithms. At the end of this paper, we explore the prospects for the development of the metaheuristics community under the LLM era.
This study constructs the LanguAge Model with Prompt EngineeRing (LAMPER) framework, designed to systematically evaluate the adaptability of pre-trained language models (PLMs) in accommodating diverse prompts and their integration in zero-shot time series (TS) classification. We deploy LAMPER in experimental assessments using 128 univariate TS datasets sourced from the UCR archive. Our findings indicate that the feature representation capacity of LAMPER is influenced by the maximum input token threshold imposed by PLMs.
Large Language Models (LLMs) raise concerns about lowering the cost of generating texts that could be used for unethical or illegal purposes, especially on social media. This paper investigates the promise of such models to help enforce legal requirements related to the disclosure of sponsored content online. We investigate the use of LLMs for generating synthetic Instagram captions with two objectives: The first objective (fidelity) is to produce realistic synthetic datasets. For this, we implement content-level and network-level metrics to assess whether synthetic captions are realistic. The second objective (utility) is to create synthetic data that is useful for sponsored content detection. For this, we evaluate the effectiveness of the generated synthetic data for training classifiers to identify undisclosed advertisements on Instagram. Our investigations show that the objectives of fidelity and utility may conflict and that prompt engineering is a useful but insufficient strategy. Additionally, we find that while individual synthetic posts may appear realistic, collectively they lack diversity, topic connectivity, and realistic user interaction patterns.
Behavior-driven development (BDD) is an Agile testing methodology fostering collaboration among developers, QA analysts, and stakeholders. In this manuscript, we propose a novel approach to enhance BDD practices using large language models (LLMs) to automate acceptance test generation. Our study uses zero and few-shot prompts to evaluate LLMs such as GPT-3.5, GPT-4, Llama-2-13B, and PaLM-2. The paper presents a detailed methodology that includes the dataset, prompt techniques, LLMs, and the evaluation process. The results demonstrate that GPT-3.5 and GPT-4 generate error-free BDD acceptance tests with better performance. The few-shot prompt technique highlights its ability to provide higher accuracy by incorporating examples for in-context learning. Furthermore, the study examines syntax errors, validation accuracy, and comparative analysis of LLMs, revealing their effectiveness in enhancing BDD practices. However, our study acknowledges that there are limitations to the proposed approach. We emphasize that this approach can support collaborative BDD processes and create opportunities for future research into automated BDD acceptance test generation using LLMs.