The recent success of LLMs has been predominantly driven by curating the training dataset composition, scaling of model architectures and dataset sizes and advancements in pretraining objectives, leaving tokenizer influence as a blind spot. Shedding light on this underexplored area, we conduct a comprehensive study on the influence of tokenizer choice on LLM downstream performance by training 24 mono- and multilingual LLMs at a 2.6B parameter scale, ablating different tokenizer algorithms and parameterizations. Our studies highlight that the tokenizer choice can significantly impact the model's downstream performance, training and inference costs. In particular, we find that the common tokenizer evaluation metrics fertility and parity are not always predictive of model downstream performance, rendering these metrics a questionable proxy for the model's downstream performance. Furthermore, we show that multilingual tokenizers trained on the five most frequent European languages require vocabulary size increases of factor three in comparison to English. While English-only tokenizers have been applied to the training of multi-lingual LLMs, we find that this approach results in a severe downstream performance degradation and additional training costs of up to 68%, due to an inefficient tokenization vocabulary.

Large Language Models (LLMs) have not only exhibited exceptional performance across various tasks, but also demonstrated sparks of intelligence. Recent studies have focused on assessing their capabilities on human exams and revealed their impressive competence in different domains. However, cognitive research on the overall knowledge structure of LLMs is still lacking. In this paper, based on educational diagnostic assessment method, we conduct an evaluation using MoocRadar, a meticulously annotated human test dataset based on Bloom Taxonomy. We aim to reveal the knowledge structures of LLMs and gain insights of their cognitive capabilities. This research emphasizes the significance of investigating LLMs' knowledge and understanding the disparate cognitive patterns of LLMs. By shedding light on models' knowledge, researchers can advance development and utilization of LLMs in a more informed and effective manner.

Do large language models (LLMs) genuinely understand the semantics of the language, or just memorize the training data? The recent concern on potential data contamination of LLMs has raised awareness of the community to conduct research on LLMs evaluation. In this paper, we propose MSTemp, an approach that creates meta semantic templates to evaluate the semantic understanding ability of LLMs. The core of MSTemp is not to perform evaluation directly on existing benchmark datasets, but to generate new out-of-distribution (OOD) evaluation sets using existing datasets as seeds. Specifically, for a given sentence, MSTemp leverages another language model to generate new samples while preserving its semantics. The new samples are called semantic templates to the original sentence. Then, MSTemp generates evaluation samples via sentence parsing and random word replacement on the semantic templates. MSTemp is highly flexible, dynamic, and cost-effective. Our initial experiments show that MSTemp-generated samples can significantly reduce the performance of LLMs using existing datasets as seeds. We hope this initial work can shed light on future research of LLMs evaluation.

Recent advancements in Large Language Models (LLMs) have heightened concerns about their potential misalignment with human values. However, evaluating their grasp of these values is complex due to their intricate and adaptable nature. We argue that truly understanding values in LLMs requires considering both "know what" and "know why". To this end, we present the Value Understanding Measurement (VUM) framework that quantitatively assesses both "know what" and "know why" by measuring the discriminator-critique gap related to human values. Using the Schwartz Value Survey, we specify our evaluation values and develop a thousand-level dialogue dataset with GPT-4. Our assessment looks at both the value alignment of LLM's outputs compared to baseline answers and how LLM responses align with reasons for value recognition versus GPT-4's annotations. We evaluate five representative LLMs and provide strong evidence that the scaling law significantly impacts "know what" but not much on "know why", which has consistently maintained a high level. This may further suggest that LLMs might craft plausible explanations based on the provided context without truly understanding their inherent value, indicating potential risks.

Business optimisation refers to the process of finding and implementing efficient and cost-effective means of operation to bring a competitive advantage for businesses. Synthesizing problem formulations is an integral part of business optimisation, which relies on human expertise to construct problem formulations using optimisation languages. Interestingly, with advancements in Large Language Models (LLMs), the human expertise needed in problem formulation can be minimized. However, developing an LLM for problem formulation is challenging, due to training data, token limitations, and lack of appropriate performance metrics. For the requirement of training data, recent attention has been directed towards fine-tuning pre-trained LLMs for downstream tasks rather than training an LLM from scratch for a specific task. In this paper, we adopt an LLM fine-tuning approach and propose an AI-Copilot for business optimisation problem formulation. For token limitations, we introduce modularization and prompt engineering techniques to synthesize complex problem formulations as modules that fit into the token limits of LLMs. Additionally, we design performance evaluation metrics that are better suited for assessing the accuracy and quality of problem formulations. The experiment results demonstrate that with this approach we can synthesize complex and large problem formulations for a typical business optimisation problem in production scheduling.

Generative Artificial Intelligence is set to revolutionize healthcare delivery by transforming traditional patient care into a more personalized, efficient, and proactive process. Chatbots, serving as interactive conversational models, will probably drive this patient-centered transformation in healthcare. Through the provision of various services, including diagnosis, personalized lifestyle recommendations, and mental health support, the objective is to substantially augment patient health outcomes, all the while mitigating the workload burden on healthcare providers. The life-critical nature of healthcare applications necessitates establishing a unified and comprehensive set of evaluation metrics for conversational models. Existing evaluation metrics proposed for various generic large language models (LLMs) demonstrate a lack of comprehension regarding medical and health concepts and their significance in promoting patients' well-being. Moreover, these metrics neglect pivotal user-centered aspects, including trust-building, ethics, personalization, empathy, user comprehension, and emotional support. The purpose of this paper is to explore state-of-the-art LLM-based evaluation metrics that are specifically applicable to the assessment of interactive conversational models in healthcare. Subsequently, we present an comprehensive set of evaluation metrics designed to thoroughly assess the performance of healthcare chatbots from an end-user perspective. These metrics encompass an evaluation of language processing abilities, impact on real-world clinical tasks, and effectiveness in user-interactive conversations. Finally, we engage in a discussion concerning the challenges associated with defining and implementing these metrics, with particular emphasis on confounding factors such as the target audience, evaluation methods, and prompt techniques involved in the evaluation process.

The application of the Large Language Model (LLM) to robot action planning has been actively studied. The instructions given to the LLM by natural language may include ambiguity and lack of information depending on the task context. It is possible to adjust the output of LLM by making the instruction input more detailed; however, the design cost is high. In this paper, we propose the interactive robot action planning method that allows the LLM to analyze and gather missing information by asking questions to humans. The method can minimize the design cost of generating precise robot instructions. We demonstrated the effectiveness of our method through concrete examples in cooking tasks. However, our experiments also revealed challenges in robot action planning with LLM, such as asking unimportant questions and assuming crucial information without asking. Shedding light on these issues provides valuable insights for future research on utilizing LLM for robotics.

Realistic and controllable traffic simulation is a core capability that is necessary to accelerate autonomous vehicle (AV) development. However, current approaches for controlling learning-based traffic models require significant domain expertise and are difficult for practitioners to use. To remedy this, we present CTG++, a scene-level conditional diffusion model that can be guided by language instructions. Developing this requires tackling two challenges: the need for a realistic and controllable traffic model backbone, and an effective method to interface with a traffic model using language. To address these challenges, we first propose a scene-level diffusion model equipped with a spatio-temporal transformer backbone, which generates realistic and controllable traffic. We then harness a large language model (LLM) to convert a user's query into a loss function, guiding the diffusion model towards query-compliant generation. Through comprehensive evaluation, we demonstrate the effectiveness of our proposed method in generating realistic, query-compliant traffic simulations.

The increasing reliance on Large Language Models (LLMs) across academia and industry necessitates a comprehensive understanding of their robustness to prompts. In response to this vital need, we introduce PromptBench, a robustness benchmark designed to measure LLMs' resilience to adversarial prompts. This study uses a plethora of adversarial textual attacks targeting prompts across multiple levels: character, word, sentence, and semantic. The adversarial prompts, crafted to mimic plausible user errors like typos or synonyms, aim to evaluate how slight deviations can affect LLM outcomes while maintaining semantic integrity. These prompts are then employed in diverse tasks, such as sentiment analysis, natural language inference, reading comprehension, machine translation, and math problem-solving. Our study generates 4788 adversarial prompts, meticulously evaluated over 8 tasks and 13 datasets. Our findings demonstrate that contemporary LLMs are not robust to adversarial prompts. Furthermore, we present comprehensive analysis to understand the mystery behind prompt robustness and its transferability. We then offer insightful robustness analysis and pragmatic recommendations for prompt composition, beneficial to both researchers and everyday users. Code is available at: https://github.com/microsoft/promptbench.

Large Language Models (LLMs) have attained the impressive capability to resolve a wide range of NLP tasks by fine-tuning high-quality instruction data. However, collecting human-written data of high quality, especially multi-turn dialogues, is expensive and unattainable for most people. Though previous studies have used powerful LLMs to generate the dialogues automatically, they all suffer from generating untruthful dialogues because of the model hallucination. Therefore, we propose a method called RefGPT to generate enormous truthful and customized dialogues without worrying about factual errors caused by the model hallucination. RefGPT solves the model hallucination in dialogue generation by restricting the LLMs to leverage the given reference instead of reciting their own knowledge to generate dialogues. Additionally, RefGPT adds detailed controls on every utterance to enable high customization capability, which previous studies have ignored. On the basis of RefGPT, we also propose two high-quality dialogue datasets generated by GPT-4, namely RefGPT-Fact and RefGPT-Code. RefGPT-Fact is a dataset with 100k multi-turn dialogues based on factual knowledge and RefGPT-Code has 76k multi-turn dialogues covering a wide range of coding scenarios. Our code and datasets are released in https://github.com/mutonix/RefGPT.