Connected and autonomous driving is developing rapidly in recent years. However, current autonomous driving systems, which are primarily based on data-driven approaches, exhibit deficiencies in interpretability, generalization, and continuing learning capabilities. In addition, the single-vehicle autonomous driving systems lack of the ability of collaboration and negotiation with other vehicles, which is crucial for the safety and efficiency of autonomous driving systems. In order to address these issues, we leverage large language models (LLMs) to develop a novel framework, AgentsCoDriver, to enable multiple vehicles to conduct collaborative driving. AgentsCoDriver consists of five modules: observation module, reasoning engine, cognitive memory module, reinforcement reflection module, and communication module. It can accumulate knowledge, lessons, and experiences over time by continuously interacting with the environment, thereby making itself capable of lifelong learning. In addition, by leveraging the communication module, different agents can exchange information and realize negotiation and collaboration in complex traffic environments. Extensive experiments are conducted and show the superiority of AgentsCoDriver.

Reinforcement Learning from Human Feedback (RLHF) is currently the most widely used method to align large language models (LLMs) with human preferences. Existing RLHF methods can be roughly categorized as either reward-based or reward-free. Novel applications such as ChatGPT and Claude leverage reward-based methods that first learn a reward model and apply actor-critic algorithms, such as Proximal Policy Optimization (PPO). However, in academic benchmarks, state-of-the-art results are often achieved via reward-free methods, such as Direct Preference Optimization (DPO). Is DPO truly superior to PPO? Why does PPO perform poorly on these benchmarks? In this paper, we first conduct both theoretical and empirical studies on the algorithmic properties of DPO and show that DPO may have fundamental limitations. Moreover, we also comprehensively examine PPO and reveal the key factors for the best performances of PPO in fine-tuning LLMs. Finally, we benchmark DPO and PPO across a collection of RLHF testbeds, ranging from dialogue to code generation. Experiment results demonstrate that PPO is able to surpass other alignment methods in all cases and achieve state-of-the-art results in challenging code competitions.

The Connections puzzle published each day by the New York Times tasks players with dividing a bank of sixteen words into four groups of four words that each relate to a common theme. Solving the puzzle requires both common linguistic knowledge (i.e. definitions and typical usage) as well as, in many cases, lateral or abstract thinking. This is because the four categories ascend in complexity, with the most challenging category often requiring thinking about words in uncommon ways or as parts of larger phrases. We investigate the capacity for automated AI systems to play Connections and explore the game's potential as an automated benchmark for abstract reasoning and a way to measure the semantic information encoded by data-driven linguistic systems. In particular, we study both a sentence-embedding baseline and modern large language models (LLMs). We report their accuracy on the task, measure the impacts of chain-of-thought prompting, and discuss their failure modes. Overall, we find that the Connections task is challenging yet feasible, and a strong test-bed for future work.

In an era where language models are increasingly integrated into decision-making and communication, understanding the biases within Large Language Models (LLMs) becomes imperative, especially when these models are applied in the economic and political domains. This work investigates the impact of fine-tuning and data selection on economic and political biases in LLM. We explore the methodological aspects of biasing LLMs towards specific ideologies, mindful of the biases that arise from their extensive training on diverse datasets. Our approach, distinct from earlier efforts that either focus on smaller models or entail resource-intensive pre-training, employs Parameter-Efficient Fine-Tuning (PEFT) techniques. These techniques allow for the alignment of LLMs with targeted ideologies by modifying a small subset of parameters. We introduce a systematic method for dataset selection, annotation, and instruction tuning, and we assess its effectiveness through both quantitative and qualitative evaluations. Our work analyzes the potential of embedding specific biases into LLMs and contributes to the dialogue on the ethical application of AI, highlighting the importance of deploying AI in a manner that aligns with societal values.

"Monopoly," said Peter Thiel, Silicon Valley's answer to Darth Vader, "is the condition of every successful business." This aspiration is widely shared by Gamman, the new acronynm for the Valley's giants – Google, Apple, Microsoft, Meta, Amazon and Nvidia. And the arrival of AI has sharpened the appetite of each for attaining that blessed state before the others get there. One symptom of their anxiety is the way they have been throwing unconscionable amounts of money at the 70-odd generative AI startups that have mushroomed since it became clear that AI was going to be the new new thing. Microsoft reportedly put 13bn (about 10.4bn) into OpenAI, for example, but it was also the lead investor in a 1.3bn funding round for Inflection, Deepmind co-founder Mustafa Suleyman's startup.

The conversion of natural language queries into SQL queries, known as Text-to-SQL, is a critical yet challenging task. This paper introduces EPI-SQL, a novel methodological framework leveraging Large Language Models (LLMs) to enhance the performance of Text-to-SQL tasks. EPI-SQL operates through a four-step process. Initially, the method involves gathering instances from the Spider dataset on which LLMs are prone to failure. These instances are then utilized to generate general error-prevention instructions (EPIs). Subsequently, LLMs craft contextualized EPIs tailored to the specific context of the current task. Finally, these context-specific EPIs are incorporated into the prompt used for SQL generation. EPI-SQL is distinguished in that it provides task-specific guidance, enabling the model to circumvent potential errors for the task at hand. Notably, the methodology rivals the performance of advanced few-shot methods despite being a zero-shot approach. An empirical assessment using the Spider benchmark reveals that EPI-SQL achieves an execution accuracy of 85.1\%, underscoring its effectiveness in generating accurate SQL queries through LLMs. The findings indicate a promising direction for future research, i.e. enhancing instructions with task-specific and contextualized rules, for boosting LLMs' performance in NLP tasks.

Large language models (LLMs) have garnered considerable attention for their proficiency in tackling intricate tasks, particularly leveraging their capacities for zero-shot and in-context learning. However, their utility has been predominantly restricted to general tasks due to an absence of domain-specific knowledge. This constraint becomes particularly pertinent in the realm of protein engineering, where specialized expertise is required for tasks such as protein function prediction, protein evolution analysis, and protein design, with a level of specialization that existing LLMs cannot furnish. In response to this challenge, we introduce \textsc{ProteinEngine}, a human-centered platform aimed at amplifying the capabilities of LLMs in protein engineering by seamlessly integrating a comprehensive range of relevant tools, packages, and software via API calls. Uniquely, \textsc{ProteinEngine} assigns three distinct roles to LLMs, facilitating efficient task delegation, specialized task resolution, and effective communication of results. This design fosters high extensibility and promotes the smooth incorporation of new algorithms, models, and features for future development. Extensive user studies, involving participants from both the AI and protein engineering communities across academia and industry, consistently validate the superiority of \textsc{ProteinEngine} in augmenting the reliability and precision of deep learning in protein engineering tasks. Consequently, our findings highlight the potential of \textsc{ProteinEngine} to bride the disconnected tools for future research in the protein engineering domain.

The rapid advancements in generative AI and large language models (LLMs) have opened up new avenues for producing synthetic data, particularly in the realm of structured tabular formats, such as product reviews. Despite the potential benefits, concerns regarding privacy leakage have surfaced, especially when personal information is utilized in the training datasets. In addition, there is an absence of a comprehensive evaluation framework capable of quantitatively measuring the quality of the generated synthetic data and their utility for downstream tasks. In response to this gap, we introduce SynEval, an open-source evaluation framework designed to assess the fidelity, utility, and privacy preservation of synthetically generated tabular data via a suite of diverse evaluation metrics. We validate the efficacy of our proposed framework - SynEval - by applying it to synthetic product review data generated by three state-of-the-art LLMs: ChatGPT, Claude, and Llama. Our experimental findings illuminate the trade-offs between various evaluation metrics in the context of synthetic data generation. Furthermore, SynEval stands as a critical instrument for researchers and practitioners engaged with synthetic tabular data,, empowering them to judiciously determine the suitability of the generated data for their specific applications, with an emphasis on upholding user privacy.

This work explores a novel data augmentation method based on Large Language Models (LLMs) for predicting item difficulty and response time of retired USMLE Multiple-Choice Questions (MCQs) in the BEA 2024 Shared Task. Our approach is based on augmenting the dataset with answers from zero-shot LLMs (Falcon, Meditron, Mistral) and employing transformer-based models based on six alternative feature combinations. The results suggest that predicting the difficulty of questions is more challenging. Notably, our top performing methods consistently include the question text, and benefit from the variability of LLM answers, highlighting the potential of LLMs for improving automated assessment in medical licensing exams. We make our code available https://github.com/ana-rogoz/BEA-2024.

Intrusion detection is a long standing and crucial problem in security. A system capable of detecting intrusions automatically is on great demand in enterprise security solutions. Existing solutions rely heavily on hand-crafted rules designed by security operators, which suffer from high false negative rates and poor generalization ability to new, zero-day attacks at scale. AI and machine learning offer promising solutions to address the issues, by inspecting abnormal user behaviors intelligently and automatically from data. However, existing learning-based intrusion detection systems in the literature are mostly designed for small data, and they lack the ability to leverage the power of big data in cloud environments. In this paper, we target at this problem and introduce an intrusion detection system which incorporates large-scale pre-training, so as to train a large language model based on tens of millions of command lines for AI-based intrusion detection. Experiments performed on 30 million training samples and 10 million test samples verify the effectiveness of our solution.