Multiobjective evolutionary algorithms (MOEAs) are major methods for solving multiobjective optimization problems (MOPs). Many MOEAs have been proposed in the past decades, of which the search operators need a carefully handcrafted design with domain knowledge. Recently, some attempts have been made to replace the manually designed operators in MOEAs with learning-based operators (e.g., neural network models). However, much effort is still required for designing and training such models, and the learned operators might not generalize well on new problems. To tackle the above challenges, this work investigates a novel approach that leverages the powerful large language model (LLM) to design MOEA operators. With proper prompt engineering, we successfully let a general LLM serve as a black-box search operator for decomposition-based MOEA (MOEA/D) in a zero-shot manner. In addition, by learning from the LLM behavior, we further design an explicit white-box operator with randomness and propose a new version of decomposition-based MOEA, termed MOEA/D-LO. Experimental studies on different test benchmarks show that our proposed method can achieve competitive performance with widely used MOEAs. It is also promising to see the operator only learned from a few instances can have robust generalization performance on unseen problems with quite different patterns and settings. The results reveal the potential benefits of using pre-trained LLMs in the design of MOEAs.To foster reproducibility and accessibility, the source code is https://github.com/FeiLiu36/LLM4MOEA.
Despite remarkable advancements in mitigating hallucinations in large language models (LLMs) by retrieval augmentation, it remains challenging to measure the reliability of LLMs using static question-answering (QA) data. Specifically, given the potential of data contamination (e.g., leading to memorization), good static benchmark performance does not ensure that model can reliably use the provided evidence for responding, which is essential to avoid hallucination when the required knowledge is new or private. Inspired by adversarial machine learning, we investigate the feasibility of automatically perturbing existing static one for dynamic evaluation. Specifically, this paper presents ReEval, an LLM-based framework using prompt chaining to perturb the original evidence for generating new test cases for evaluating the LLMs' reliability in using new evidence for answering. We implement ReEval using ChatGPT and evaluate the resulting variants of two popular open-domain QA datasets on a collection of LLMs under various prompting settings. Our generated data is human-readable and useful to trigger hallucination in LLM. Accurate models on static data are observed to produce unsupported answers from the perturbed evidence, with pronounced accuracy drops across LLMs including GPT-4. We find that our adversarial examples are transferable across all considered LLMs. The examples generated by a small model can be used to evaluate a much larger model, making our approach cost-effective.
In this paper, we investigate the usage of large language models (LLMs) to improve the performance of competitive speech recognition systems. Different from previous LLM-based ASR error correction methods, we propose a novel multi-stage approach that utilizes uncertainty estimation of ASR outputs and reasoning capability of LLMs. Specifically, the proposed approach has two stages: the first stage is about ASR uncertainty estimation and exploits N-best list hypotheses to identify less reliable transcriptions; The second stage works on these identified transcriptions and performs LLM-based corrections. This correction task is formulated as a multi-step rule-based LLM reasoning process, which uses explicitly written rules in prompts to decompose the task into concrete reasoning steps. Our experimental results demonstrate the effectiveness of the proposed method by showing 10% ~ 20% relative improvement in WER over competitive ASR systems -- across multiple test domains and in zero-shot settings.
This report focuses on enhancing a binary code comment quality classification model by integrating generated code and comment pairs, to improve model accuracy. The dataset comprises 9048 pairs of code and comments written in the C programming language, each annotated as "Useful" or "Not Useful." Additionally, code and comment pairs are generated using a Large Language Model Architecture, and these generated pairs are labeled to indicate their utility. The outcome of this effort consists of two classification models: one utilizing the original dataset and another incorporating the augmented dataset with the newly generated code comment pairs and labels.
While Reinforcement Learning from Human Feedback (RLHF) aligns Large Language Models (LLMs) with general, aggregate human preferences, it is suboptimal for learning diverse, individual perspectives. In this work, we study Reinforcement Learning from Personalized Human Feedback (RLPHF) problem, wherein LLMs are aligned to multiple (sometimes conflicting) preferences by modeling alignment as a Multi-Objective Reinforcement Learning (MORL) problem. Compared to strong single-objective baselines, we show that we can achieve personalized alignment by decomposing preferences into multiple dimensions. These dimensions are defined based on personalizations that are declared as desirable by the user. In this work, we show that they can be efficiently trained independently in a distributed manner and combined effectively post-hoc through parameter merging. The code is available at https://github.com/joeljang/RLPHF.
Zero-shot learning (ZSL) aims to classify objects that are not observed or seen during training. It relies on class semantic description to transfer knowledge from the seen classes to the unseen classes. Existing methods of obtaining class semantics include manual attributes or automatic word vectors from language models (like word2vec). We know attribute annotation is costly, whereas automatic word-vectors are relatively noisy. To address this problem, we explore how ChatGPT, a large language model, can enhance class semantics for ZSL tasks. ChatGPT can be a helpful source to obtain text descriptions for each class containing related attributes and semantics. We use the word2vec model to get a word vector using the texts from ChatGPT. Then, we enrich word vectors by combining the word embeddings from class names and descriptions generated by ChatGPT. More specifically, we leverage ChatGPT to provide extra supervision for the class description, eventually benefiting ZSL models. We evaluate our approach on various 2D image (CUB and AwA) and 3D point cloud (ModelNet10, ModelNet40, and ScanObjectNN) datasets and show that it improves ZSL performance. Our work contributes to the ZSL literature by applying ChatGPT for class semantics enhancement and proposing a novel word vector fusion method.
Abuse of large language models reveals high risks as large language models are being deployed at an astonishing speed. It is important to protect the model weights to avoid malicious usage that violates licenses of open-source large language models. This paper proposes a novel watermarking strategy that plants watermarks in the quantization process of large language models without pre-defined triggers during inference. The watermark works when the model is used in the fp32 mode and remains hidden when the model is quantized to int8, in this way, the users can only inference the model without further supervised fine-tuning of the model. We successfully plant the watermark into open-source large language model weights including GPT-Neo and LLaMA. We hope our proposed method can provide a potential direction for protecting model weights in the era of large language model applications.
The vision and language generative models have been overgrown in recent years. For video generation, various open-sourced models and public-available services have been developed to generate high-quality videos. However, these methods often use a few metrics, e.g., FVD or IS, to evaluate the performance. We argue that it is hard to judge the large conditional generative models from the simple metrics since these models are often trained on very large datasets with multi-aspect abilities. Thus, we propose a novel framework and pipeline for exhaustively evaluating the performance of the generated videos. Our approach involves generating a diverse and comprehensive list of 700 prompts for text-to-video generation, which is based on an analysis of real-world user data and generated with the assistance of a large language model. Then, we evaluate the state-of-the-art video generative models on our carefully designed benchmark, in terms of visual qualities, content qualities, motion qualities, and text-video alignment with 17 well-selected objective metrics. To obtain the final leaderboard of the models, we further fit a series of coefficients to align the objective metrics to the users' opinions. Based on the proposed human alignment method, our final score shows a higher correlation than simply averaging the metrics, showing the effectiveness of the proposed evaluation method.
The ability to invent novel and interesting problems is a remarkable feature of human intelligence that drives innovation, art, and science. We propose a method that aims to automate this process by harnessing the power of state-of-the-art generative models to produce a diversity of challenging yet solvable problems, here in the context of Python programming puzzles. Inspired by the intrinsically motivated literature, Autotelic CodE Search (ACES) jointly optimizes for the diversity and difficulty of generated problems. We represent problems in a space of LLM-generated semantic descriptors describing the programming skills required to solve them (e.g. string manipulation, dynamic programming, etc.) and measure their difficulty empirically as a linearly decreasing function of the success rate of Llama-3-70B, a state-of-the-art LLM problem solver. ACES iteratively prompts a large language model to generate difficult problems achieving a diversity of target semantic descriptors (goal-directed exploration) using previously generated problems as in-context examples. ACES generates problems that are more diverse and more challenging than problems produced by baseline methods and three times more challenging than problems found in existing Python programming benchmarks on average across 11 state-of-the-art code LLMs.
We study how to perform unlearning, i.e. forgetting undesirable misbehaviors, on large language models (LLMs). We show at least three scenarios of aligning LLMs with human preferences can benefit from unlearning: (1) removing harmful responses, (2) erasing copyright-protected content as requested, and (3) reducing hallucinations. Unlearning, as an alignment technique, has three advantages. (1) It only requires negative (e.g. harmful) examples, which are much easier and cheaper to collect (e.g. via red teaming or user reporting) than positive (e.g. helpful and often human-written) examples required in RLHF (RL from human feedback). (2) It is computationally efficient. (3) It is especially effective when we know which training samples cause the misbehavior. To the best of our knowledge, our work is among the first to explore LLM unlearning. We are also among the first to formulate the settings, goals, and evaluations in LLM unlearning. We show that if practitioners only have limited resources, and therefore the priority is to stop generating undesirable outputs rather than to try to generate desirable outputs, unlearning is particularly appealing. Despite only having negative samples, our ablation study shows that unlearning can still achieve better alignment performance than RLHF with just 2% of its computational time.
In support of open and reproducible research, there has been a rapidly increasing number of datasets made available for research. As the availability of datasets increases, it becomes more important to have quality metadata for discovering and reusing them. Yet, it is a common issue that datasets often lack quality metadata due to limited resources for data curation. Meanwhile, technologies such as artificial intelligence and large language models (LLMs) are progressing rapidly. Recently, systems based on these technologies, such as ChatGPT, have demonstrated promising capabilities for certain data curation tasks. This paper proposes to leverage LLMs for cost-effective annotation of subject metadata through the LLM-based in-context learning. Our method employs GPT-3.5 with prompts designed for annotating subject metadata, demonstrating promising performance in automatic metadata annotation. However, models based on in-context learning cannot acquire discipline-specific rules, resulting in lower performance in several categories. This limitation arises from the limited contextual information available for subject inference. To the best of our knowledge, we are introducing, for the first time, an in-context learning method that harnesses large language models for automated subject metadata annotation.
In response to the pressing need for advanced clinical problem-solving tools in healthcare, we introduce BooksMed, a novel framework based on a Large Language Model (LLM). BooksMed uniquely emulates human cognitive processes to deliver evidence-based and reliable responses, utilizing the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) framework to effectively quantify evidence strength. For clinical decision-making to be appropriately assessed, an evaluation metric that is clinically aligned and validated is required. As a solution, we present ExpertMedQA, a multispecialty clinical benchmark comprised of open-ended, expert-level clinical questions, and validated by a diverse group of medical professionals. By demanding an in-depth understanding and critical appraisal of up-to-date clinical literature, ExpertMedQA rigorously evaluates LLM performance. BooksMed outperforms existing state-of-the-art models Med-PaLM 2, Almanac, and ChatGPT in a variety of medical scenarios. Therefore, a framework that mimics human cognitive stages could be a useful tool for providing reliable and evidence-based responses to clinical inquiries.
In the wake of relentless digital transformation, data-driven solutions are emerging as powerful tools to address multifarious industrial tasks such as forecasting, anomaly detection, planning, and even complex decision-making. Although data-centric R&D has been pivotal in harnessing these solutions, it often comes with significant costs in terms of human, computational, and time resources. This paper delves into the potential of large language models (LLMs) to expedite the evolution cycle of data-centric R&D. Assessing the foundational elements of data-centric R&D, including heterogeneous task-related data, multi-facet domain knowledge, and diverse computing-functional tools, we explore how well LLMs can understand domain-specific requirements, generate professional ideas, utilize domain-specific tools to conduct experiments, interpret results, and incorporate knowledge from past endeavors to tackle new challenges. We take quantitative investment research as a typical example of industrial data-centric R&D scenario and verified our proposed framework upon our full-stack open-sourced quantitative research platform Qlib and obtained promising results which shed light on our vision of automatic evolving of industrial data-centric R&D cycle.
The "Information Retrieval in Software Engineering (IRSE)" at FIRE 2023 shared task introduces code comment classification, a challenging task that pairs a code snippet with a comment that should be evaluated as either useful or not useful to the understanding of the relevant code. We answer the code comment classification shared task challenge by providing a two-fold evaluation: from an algorithmic perspective, we compare the performance of classical machine learning systems and complement our evaluations from a data-driven perspective by generating additional data with the help of large language model (LLM) prompting to measure the potential increase in performance. Our best model, which took second place in the shared task, is a Neural Network with a Macro-F1 score of 88.401% on the provided seed data and a 1.5% overall increase in performance on the data generated by the LLM.
This paper proposes an interactive navigation framework by using large language and vision-language models, allowing robots to navigate in environments with traversable obstacles. We utilize the large language model (GPT-3.5) and the open-set Vision-language Model (Grounding DINO) to create an action-aware costmap to perform effective path planning without fine-tuning. With the large models, we can achieve an end-to-end system from textual instructions like "Can you pass through the curtains to deliver medicines to me?", to bounding boxes (e.g., curtains) with action-aware attributes. They can be used to segment LiDAR point clouds into two parts: traversable and untraversable parts, and then an action-aware costmap is constructed for generating a feasible path. The pre-trained large models have great generalization ability and do not require additional annotated data for training, allowing fast deployment in the interactive navigation tasks. We choose to use multiple traversable objects such as curtains and grasses for verification by instructing the robot to traverse them. Besides, traversing curtains in a medical scenario was tested. All experimental results demonstrated the proposed framework's effectiveness and adaptability to diverse environments.
Large language models have shown their remarkable capabilities as a general interface for various language-related applications. Motivated by this, we target to build a unified interface for completing many vision-language tasks including image description, visual question answering, and visual grounding, among others. The challenge is to use a single model for performing diverse vision-language tasks effectively with simple multi-modal instructions. Towards this objective, we introduce MiniGPT-v2, a model that can be treated as a unified interface for better handling various vision-language tasks. We propose using unique identifiers for different tasks when training the model. These identifiers enable our model to better distinguish each task instruction effortlessly and also improve the model learning efficiency for each task. After the three-stage training, the experimental results show that MiniGPT-v2 achieves strong performance on many visual question-answering and visual grounding benchmarks compared to other vision-language generalist models. Our model and codes are available at https://minigpt-v2.github.io/
Language models have become the backbone of today's AI systems. However, their predominant left-to-right generation limits the use of bidirectional context, which is essential for tasks that involve filling text in the middle. We propose the Fill-in Language Model (FiLM), a new language modeling approach that allows for flexible generation at any position without adhering to a specific generation order. Its training extends the masked language modeling objective by adopting varying mask probabilities sampled from the Beta distribution to enhance the generative capabilities of FiLM. During inference, FiLM can seamlessly insert missing phrases, sentences, or paragraphs, ensuring that the outputs are fluent and are coherent with the surrounding context. In both automatic and human evaluations, FiLM outperforms existing infilling methods that rely on left-to-right language models trained on rearranged text segments. FiLM is easy to implement and can be either trained from scratch or fine-tuned from a left-to-right language model. Notably, as the model size grows, FiLM's perplexity approaches that of strong left-to-right language models of similar sizes, indicating FiLM's scalability and potential as a large language model.
Large language models (LLMs) have been treated as knowledge bases due to their strong performance in knowledge probing tasks. LLMs are typically evaluated using accuracy, yet this metric does not capture the vulnerability of LLMs to hallucination-inducing factors like prompt and context variability. How do we evaluate the capabilities of LLMs to consistently produce factually correct answers? In this paper, we propose MOdel kNowledge relIabiliTy scORe (MONITOR), a novel metric designed to directly measure LLMs' factual reliability. MONITOR computes the distance between the probability distributions of a valid output and its counterparts produced by the same LLM probing the same fact using different styles of prompts and contexts.Experiments on a comprehensive range of 12 LLMs demonstrate the effectiveness of MONITOR in evaluating the factual reliability of LLMs while maintaining a low computational overhead. In addition, we release the FKTC (Factual Knowledge Test Corpus) test set, containing 210,158 prompts in total to foster research along this line (https://github.com/Vicky-Wil/MONITOR).
Large language models (LLMs) have shown impressive in-context learning (ICL) ability in code generation. LLMs take a prompt consisting of requirement-code examples and a new requirement as input, and output new programs. Existing studies have found that ICL is highly dominated by the examples and thus arises research on example selection. However, existing approaches randomly select examples or only consider the textual similarity of requirements to retrieve, leading to sub-optimal performance. In this paper, we propose a novel learning-based selection approach named LAIL (LLM-Aware In-context Learning) for code generation. Given a candidate example, we exploit LLMs themselves to estimate it by considering the generation probabilities of ground-truth programs given a requirement and the example. We then label candidate examples as positive or negative through the probability feedback. Based on the labeled data, we import a contrastive learning objective to train an effective retriever that acquires the preference of LLMs in code generation. We apply LAIL to three LLMs and evaluate it on three representative datasets (e.g., MBJP, MBPP, and MBCPP). LATA outperforms the state-of-the-art baselines by 11.58%, 6.89%, and 5.07% on CodeGen, and 4.38%, 2.85%, and 2.74% on GPT-3.5 in terms of Pass@1, respectively.
Large language models (LLMs) have showcased remarkable capabilities in understanding and generating language. However, their ability in comprehending ancient languages, particularly ancient Chinese, remains largely unexplored. To bridge this gap, we present ACLUE, an evaluation benchmark designed to assess the capability of language models in comprehending ancient Chinese. ACLUE consists of 15 tasks cover a range of skills, spanning phonetic, lexical, syntactic, semantic, inference and knowledge. Through the evaluation of eight state-of-the-art LLMs, we observed a noticeable disparity in their performance between modern Chinese and ancient Chinese. Among the assessed models, ChatGLM2 demonstrates the most remarkable performance, achieving an average score of 37.4%. We have made our code and data public available.