Major cloud providers have employed advanced AI-based solutions like large language models to aid humans in identifying the root causes of cloud incidents. Despite the growing prevalence of AI-driven assistants in the root cause analysis process, their effectiveness in assisting on-call engineers is constrained by low accuracy due to the intrinsic difficulty of the task, a propensity for LLM-based approaches to hallucinate, and difficulties in distinguishing these well-disguised hallucinations. To address this challenge, we propose to perform confidence estimation for the predictions to help on-call engineers make decisions on whether to adopt the model prediction. Considering the black-box nature of many LLM-based root cause predictors, fine-tuning or temperature-scaling-based approaches are inapplicable. We therefore design an innovative confidence estimation framework based on prompting retrieval-augmented large language models (LLMs) that demand a minimal amount of information from the root cause predictor. This approach consists of two scoring phases: the LLM-based confidence estimator first evaluates its confidence in making judgments in the face of the current incident that reflects its ``grounded-ness" level in reference data, then rates the root cause prediction based on historical references. An optimization step combines these two scores for a final confidence assignment. We show that our method is able to produce calibrated confidence estimates for predicted root causes, validate the usefulness of retrieved historical data and the prompting strategy as well as the generalizability across different root cause prediction models. Our study takes an important move towards reliably and effectively embedding LLMs into cloud incident management systems.

Modern neural collaborative filtering techniques are critical to the success of e-commerce, social media, and content-sharing platforms. However, despite technical advances -- for every new application domain, we need to train an NCF model from scratch. In contrast, pre-trained vision and language models are routinely applied to diverse applications directly (zero-shot) or with limited fine-tuning. Inspired by the impact of pre-trained models, we explore the possibility of pre-trained recommender models that support building recommender systems in new domains, with minimal or no retraining, without the use of any auxiliary user or item information. Zero-shot recommendation without auxiliary information is challenging because we cannot form associations between users and items across datasets when there are no overlapping users or items. Our fundamental insight is that the statistical characteristics of the user-item interaction matrix are universally available across different domains and datasets. Thus, we use the statistical characteristics of the user-item interaction matrix to identify dataset-independent representations for users and items. We show how to learn universal (i.e., supporting zero-shot adaptation without user or item auxiliary information) representations for nodes and edges from the bipartite user-item interaction graph. We learn representations by exploiting the statistical properties of the interaction data, including user and item marginals, and the size and density distributions of their clusters.

We introduce Jais and Jais-chat, new state-of-the-art Arabic-centric foundation and instruction-tuned open generative large language models (LLMs). The models are based on the GPT-3 decoder-only architecture and are pretrained on a mixture of Arabic and English texts, including source code in various programming languages. With 13 billion parameters, they demonstrate better knowledge and reasoning capabilities in Arabic than any existing open Arabic and multilingual models by a sizable margin, based on extensive evaluation. Moreover, the models are competitive in English compared to English-centric open models of similar size, despite being trained on much less English data. We provide a detailed description of the training, the tuning, the safety alignment, and the evaluation of the models. We release two open versions of the model -- the foundation Jais model, and an instruction-tuned Jais-chat variant -- with the aim of promoting research on Arabic LLMs. Available at https://huggingface.co/inception-mbzuai/jais-13b-chat

When training powerful AI systems to perform complex tasks, it may be challenging to provide training signals which are robust to optimization. One concern is \textit{measurement tampering}, where the AI system manipulates multiple measurements to create the illusion of good results instead of achieving the desired outcome. In this work, we build four new text-based datasets to evaluate measurement tampering detection techniques on large language models. Concretely, given sets of text inputs and measurements aimed at determining if some outcome occurred, as well as a base model able to accurately predict measurements, the goal is to determine if examples where all measurements indicate the outcome occurred actually had the outcome occur, or if this was caused by measurement tampering. We demonstrate techniques that outperform simple baselines on most datasets, but don't achieve maximum performance. We believe there is significant room for improvement for both techniques and datasets, and we are excited for future work tackling measurement tampering.

With easier access to powerful compute resources, there is a growing trend in the field of AI for software development to develop larger and larger language models (LLMs) to address a variety of programming tasks. Even LLMs applied to tasks from the high-performance computing (HPC) domain are huge in size (e.g., billions of parameters) and demand expensive compute resources for training. We found this design choice confusing - why do we need large LLMs trained on natural languages and programming languages unrelated to HPC for HPC-specific tasks? In this line of work, we aim to question design choices made by existing LLMs by developing smaller LLMs for specific domains - we call them domain-specific LLMs. Specifically, we start off with HPC as a domain and propose a novel tokenizer named Tokompiler, designed specifically for preprocessing code in HPC and compilation-centric tasks. Tokompiler leverages knowledge of language primitives to generate language-oriented tokens, providing a context-aware understanding of code structure while avoiding human semantics attributed to code structures completely. We applied Tokompiler to pre-train two state-of-the-art models, SPT-Code and Polycoder, for a Fortran code corpus mined from GitHub. We evaluate the performance of these models against the conventional LLMs. Results demonstrate that Tokompiler significantly enhances code completion accuracy and semantic understanding compared to traditional tokenizers in normalized-perplexity tests, down to ~1 perplexity score. This research opens avenues for further advancements in domain-specific LLMs, catering to the unique demands of HPC and compilation tasks.

Chain-of-Thought (CoT) prompting in large language models (LLMs) has shown promising performance on mathematical reasoning tasks. Recently, Self-Consistency samples a diverse set of reasoning chains with different answers and chooses the answer by majority voting. Though effective, its performance cannot be further improved by sampling more reasoning chains. To address this problem, we propose to integrate backward reasoning into answer verification. We first mask a number in the question by ${\bf x}$. The LLM is then asked to predict the masked number with a candidate answer $A$ embedded in the template: ``If we know the answer to the above question is $\{A\}$, what is the value of unknown variable ${\bf x}$?'' The LLM is expected to predict the masked number successfully if the provided candidate answer is correct. To further improve performance, we propose FOBAR (FOrward-BAckward Reasoning) to combine forward and backward reasoning for verifying candidate answers. Experiments are performed on six standard mathematical data sets and three LLMs (text-davinci-003, GPT-3.5-Turbo, GPT-4). Results show that FOBAR achieves state-of-the-art performance. In particular, FOBAR outperforms Self-Consistency which uses forward reasoning alone, demonstrating that combining forward and forward reasoning is better. It also outperforms existing verification methods, verifying the effectiveness of using the simple template in backward reasoning and the proposed combination.

ASR error correction is an interesting option for post processing speech recognition system outputs. These error correction models are usually trained in a supervised fashion using the decoding results of a target ASR system. This approach can be computationally intensive and the model is tuned to a specific ASR system. Recently generative large language models (LLMs) have been applied to a wide range of natural language processing tasks, as they can operate in a zero-shot or few shot fashion. In this paper we investigate using ChatGPT, a generative LLM, for ASR error correction. Based on the ASR N-best output, we propose both unconstrained and constrained, where a member of the N-best list is selected, approaches. Additionally, zero and 1-shot settings are evaluated. Experiments show that this generative LLM approach can yield performance gains for two different state-of-the-art ASR architectures, transducer and attention-encoder-decoder based, and multiple test sets.

Despite the promising progress in multi-modal tasks, current large multi-modal models (LMMs) are prone to hallucinating inconsistent descriptions with respect to the associated image and human instructions. This paper addresses this issue by introducing the first large and diverse visual instruction tuning dataset, named Large-scale Robust Visual (LRV)-Instruction. Our dataset comprises 400k visual instructions generated by GPT4, covering 16 vision-and-language tasks with open-ended instructions and answers. Unlike existing studies that primarily focus on positive instruction samples, we design LRV-Instruction to include both positive and negative instructions for more robust visual instruction tuning. Our negative instructions are designed at three semantic levels: (i) Nonexistent Object Manipulation, (ii) Existent Object Manipulation and (iii) Knowledge Manipulation. To efficiently measure the hallucination generated by LMMs, we propose GPT4-Assisted Visual Instruction Evaluation (GAVIE), a stable approach to evaluate visual instruction tuning like human experts. GAVIE does not require human-annotated groundtruth answers and can adapt to diverse instruction formats. We conduct comprehensive experiments to investigate the hallucination of LMMs. Our results demonstrate existing LMMs exhibit significant hallucinations when presented with our negative instructions, particularly Existent Object and Knowledge Manipulation instructions. Moreover, we successfully mitigate hallucination by finetuning MiniGPT4 and mPLUG-Owl on LRV-Instruction while improving performance on several public datasets compared to state-of-the-art methods. Additionally, we observed that a balanced ratio of positive and negative instances in the training data leads to a more robust model.

A core tension in models of concept learning is that the model must carefully balance the tractability of inference against the expressivity of the hypothesis class. Humans, however, can efficiently learn a broad range of concepts. We introduce a model of inductive learning that seeks to be human-like in that sense. It implements a Bayesian reasoning process where a language model first proposes candidate hypotheses expressed in natural language, which are then re-weighed by a prior and a likelihood. By estimating the prior from human data, we can predict human judgments on learning problems involving numbers and sets, spanning concepts that are generative, discriminative, propositional, and higher-order.

In this work, we propose GPT-FL, a generative pre-trained model-assisted federated learning (FL) framework. At its core, GPT-FL leverages generative pretrained models to generate diversified synthetic data. These generated data are used to train a downstream model on the server, which is then fine-tuned with private client data under the standard FL framework. We show that GPT-FL consistently outperforms state-of-the-art FL methods in terms of model test accuracy, communication efficiency, and client sampling efficiency. Through comprehensive ablation analysis, we discover that the downstream model generated by synthetic data plays a crucial role in controlling the direction of gradient diversity during FL training, which enhances convergence speed and contributes to the notable accuracy boost observed with GPT-FL. Also, regardless of whether the target data falls within or outside the domain of the pre-trained generative model, GPT-FL consistently achieves significant performance gains, surpassing the results obtained by models trained solely with FL or synthetic data. Federated learning (FL) is a privacy-preserving machine learning paradigm that allows a collection of clients to collaboratively train a machine learning model without sharing their private data Zhang et al. (2021). Most existing FL studies such as McMahan et al. (2016); Bonawitz et al. (2019) follow the standard FL architecture, where each participating client trains a local model using its own private data and a central server aggregates these locally trained models to update a global model and send it back to the clients for the next round of training. However, although many efforts have been made Sahu et al. (2018); Karimireddy et al. (2019); Reddi et al. (2020), the performance of standard FL is still constrained by client drift caused by the heterogeneity in private data distribution across the clients. To enhance the performance of FL, recent studies propose to incorporate data collected from public spaces such as the internet into the FL process Lin et al. (2020); Li et al. (2021); Itahara et al. (2020); Cho et al. (2022). However, the performance of such public data-based approaches is heavily dependent on the quality of the collected public data. Unfortunately, obtaining the desired public data can be extremely challenging in practice and there is a lack of principled guidance on how to obtain them. To address the issues of public data-based approaches, FL methods based on synthetic data emerge Zhang et al. (2022); Zhu et al. (2021); Pi et al. (2022).