The process of instruction tuning aligns pre-trained large language models (LLMs) with open-domain instructions and human-preferred responses. While several studies have explored autonomous approaches to distilling and annotating instructions from more powerful proprietary LLMs, such as ChatGPT, they often neglect the impact of task distributions and the varying difficulty of instructions of the training sets. This oversight can lead to imbalanced knowledge capabilities and poor generalization powers of small student LLMs. To address this challenge, we introduce Task-Aware Curriculum Planning for Instruction Refinement (TAPIR), a multi-round distillation framework with balanced task distributions and dynamic difficulty adjustment. This approach utilizes an oracle LLM to select instructions that are difficult for a student LLM to follow and distill instructions with balanced task distributions. By incorporating curriculum planning, our approach systematically escalates the difficulty levels, progressively enhancing the student LLM's capabilities. We rigorously evaluate TAPIR using two widely recognized benchmarks, including AlpacaEval 2.0 and MT-Bench. The empirical results demonstrate that the student LLMs, trained with our method and less training data, outperform larger instruction-tuned models and strong distillation baselines. The improvement is particularly notable in complex tasks, such as logical reasoning and code generation.

In IoT environments, large amounts of procedural data are generated from IoT devices, information systems, and other software applications. The use of this data can foster the development of innovative applications in process control [63, 75, 56, 54, 35, 52, 42, 68], process conformance checking [23, 81, 83, 28], and process enhancement [67, 59], among others. Particularly, the use of process mining techniques to analyze not only process data but also IoT-collected data could provide important insights into processes and interactions as shown in different applications in the manufacturing domain, such as [58, 75, 56, 59, 67]. In these applications, IoT actuators are used to realize and execute process activities, while IoT sensors and smart tags are used to closely monitor the execution environment and involved resources [79, 75, 26, 37, 54]. IoT technology can therefore capture the context in which certain process tasks are performed, allowing process mining techniques to better understand and analyze the processes [7, 76, 12]. As such, besides the procedural data generated from the process execution systems, the data captured by IoT should also be considered an integral part of the process execution in the form of IoT-enriched event logs [57, 53]. Both the procedural nature of sensor logs, and the tight integration of these with the process executions and the executing resources [24] makes sensor data an integral part of process-based application scenarios in IoT [76, 75, 7]. However, the integration of IoT data and process data to be used for process mining is still often done ex-post in a manual fashion during a separate pre-processing phase [95, 73, 53]. In these cases, the data from the IoT environment is still collected and stored separately, and only later it is explicitly connected to the notion of a process, which is non-trivial as pointed out in the challenge "Bridging the Gap Between Event-based and Process-based Systems" in the BPM-IoT manifesto [37].

Multimodal LLMs are the natural evolution of LLMs, and enlarge their capabilities so as to work beyond the pure textual modality. As research is being carried out to design novel architectures and vision-and-language adapters, in this paper we concentrate on endowing such models with the capability of answering questions that require external knowledge. Our approach, termed Wiki-LLaVA, aims at integrating an external knowledge source of multimodal documents, which is accessed through a hierarchical retrieval pipeline. Relevant passages, using this approach, are retrieved from the external knowledge source and employed as additional context for the LLM, augmenting the effectiveness and precision of generated dialogues. We conduct extensive experiments on datasets tailored for visual question answering with external data and demonstrate the appropriateness of our approach.

This work introduces a novel approach to achieving architecture-agnostic equivariance in deep learning, particularly addressing the limitations of traditional equivariant architectures and the inefficiencies of the existing architecture-agnostic methods. Building equivariant models using traditional methods requires designing equivariant versions of existing models and training them from scratch, a process that is both impractical and resource-intensive. Canonicalization has emerged as a promising alternative for inducing equivariance without altering model architecture, but it suffers from the need for highly expressive and expensive equivariant networks to learn canonical orientations accurately. We propose a new method that employs any non-equivariant network for canonicalization. Our method uses contrastive learning to efficiently learn a unique canonical orientation and offers more flexibility for the choice of canonicalization network. We empirically demonstrate that this approach outperforms existing methods in achieving equivariance for large pretrained models and significantly speeds up the canonicalization process, making it up to 2 times faster.

We introduce the Riskman ontology & shapes for representing and analysing information about risk management for medical devices. Risk management is concerned with taking necessary precautions so a medical device does not cause harms for users or the environment. To date, risk management documentation is submitted to notified bodies (for certification) in the form of semi-structured natural language text. We propose to use classes from the Riskman ontology to logically model risk management documentation, and to use the included SHACL constraints to check for syntactic completeness and conformity to relevant standards. In particular, the ontology is modelled after ISO 14971 and the recently published VDE Spec 90025. Our proposed methodology has the potential to save many person-hours for both manufacturers (when creating risk management documentation) as well as notified bodies (when assessing submitted applications for certification), and thus offers considerable benefits for healthcare and, by extension, society as a whole.

Multimodal foundation models are prone to hallucination, generating outputs that either contradict the input or are not grounded by factual information. Given the diversity in architectures, training data and instruction tuning techniques, there can be large variations in systems' susceptibility to hallucinations. To assess system hallucination robustness, hallucination ranking approaches have been developed for specific tasks such as image captioning, question answering, summarization, or biography generation. However, these approaches typically compare model outputs to gold-standard references or labels, limiting hallucination benchmarking for new domains. This work proposes "CrossCheckGPT", a reference-free universal hallucination ranking for multimodal foundation models. The core idea of CrossCheckGPT is that the same hallucinated content is unlikely to be generated by different independent systems, hence cross-system consistency can provide meaningful and accurate hallucination assessment scores. CrossCheckGPT can be applied to any model or task, provided that the information consistency between outputs can be measured through an appropriate distance metric. Focusing on multimodal large language models that generate text, we explore two information consistency measures: CrossCheck-explicit and CrossCheck-implicit. We showcase the applicability of our method for hallucination ranking across various modalities, namely the text, image, and audio-visual domains. Further, we propose the first audio-visual hallucination benchmark, "AVHalluBench", and illustrate the effectiveness of CrossCheckGPT, achieving correlations of 98% and 89% with human judgements on MHaluBench and AVHalluBench, respectively.

Multimodal Large Language Models (MLLMs) are widely regarded as crucial in the exploration of Artificial General Intelligence (AGI). The core of MLLMs lies in their capability to achieve cross-modal alignment. To attain this goal, current MLLMs typically follow a two-phase training paradigm: the pre-training phase and the instruction-tuning phase. Despite their success, there are shortcomings in the modeling of alignment capabilities within these models. Firstly, during the pre-training phase, the model usually assumes that all image-text pairs are uniformly aligned, but in fact the degree of alignment between different image-text pairs is inconsistent. Secondly, the instructions currently used for finetuning incorporate a variety of tasks, different tasks's instructions usually require different levels of alignment capabilities, but previous MLLMs overlook these differentiated alignment needs. To tackle these issues, we propose a new multimodal large language model AlignGPT. In the pre-training stage, instead of treating all image-text pairs equally, we assign different levels of alignment capabilities to different image-text pairs. Then, in the instruction-tuning phase, we adaptively combine these different levels of alignment capabilities to meet the dynamic alignment needs of different instructions. Extensive experimental results show that our model achieves competitive performance on 12 benchmarks.

We present a new challenge to examine whether large language models understand social norms. In contrast to existing datasets, our dataset requires a fundamental understanding of social norms to solve. Our dataset features the largest set of social norm skills, consisting of 402 skills and 12,383 questions covering a wide set of social norms ranging from opinions and arguments to culture and laws. We design our dataset according to the K-12 curriculum. This enables the direct comparison of the social understanding of large language models to humans, more specifically, elementary students. While prior work generates nearly random accuracy on our benchmark, recent large language models such as GPT3.5-Turbo and LLaMA2-Chat are able to improve the performance significantly, only slightly below human performance. We then propose a multi-agent framework based on large language models to improve the models' ability to understand social norms. This method further improves large language models to be on par with humans. Given the increasing adoption of large language models in real-world applications, our finding is particularly important and presents a unique direction for future improvements.

We introduce a new challenge to test the STEM skills of neural models. The problems in the real world often require solutions, combining knowledge from STEM (science, technology, engineering, and math). Unlike existing datasets, our dataset requires the understanding of multimodal vision-language information of STEM. Our dataset features one of the largest and most comprehensive datasets for the challenge. It includes 448 skills and 1,073,146 questions spanning all STEM subjects. Compared to existing datasets that often focus on examining expert-level ability, our dataset includes fundamental skills and questions designed based on the K-12 curriculum. We also add state-of-the-art foundation models such as CLIP and GPT-3.5-Turbo to our benchmark. Results show that the recent model advances only help master a very limited number of lower grade-level skills (2.5% in the third grade) in our dataset. In fact, these models are still well below (averaging 54.7%) the performance of elementary students, not to mention near expert-level performance. To understand and increase the performance on our dataset, we teach the models on a training split of our dataset. Even though we observe improved performance, the model performance remains relatively low compared to average elementary students. To solve STEM problems, we will need novel algorithmic innovations from the community.

Modern code review is a process in which an incremental code contribution made by a code author is reviewed by one or more peers before it is committed to the version control system. An important element of modern code review is verifying that code contributions adhere to best practices. While some of these best practices can be automatically verified, verifying others is commonly left to human reviewers. This paper reports on the development, deployment, and evaluation of AutoCommenter, a system backed by a large language model that automatically learns and enforces coding best practices. We implemented AutoCommenter for four programming languages (C++, Java, Python, and Go) and evaluated its performance and adoption in a large industrial setting. Our evaluation shows that an end-to-end system for learning and enforcing coding best practices is feasible and has a positive impact on the developer workflow. Additionally, this paper reports on the challenges associated with deploying such a system to tens of thousands of developers and the corresponding lessons learned.