Large language models (LLMs), such as ChatGPT, are able to generate human-like, fluent responses for many downstream tasks, e.g., task-oriented dialog and question answering. However, applying LLMs to real-world, mission-critical applications remains challenging mainly due to their tendency to generate hallucinations and their inability to use external knowledge. This paper proposes a LLM-Augmenter system, which augments a black-box LLM with a set of plug-and-play modules. Our system makes the LLM generate responses grounded in external knowledge, e.g., stored in task-specific databases. It also iteratively revises LLM prompts to improve model responses using feedback generated by utility functions, e.g., the factuality score of a LLM-generated response. The effectiveness of LLM-Augmenter is empirically validated on two types of scenarios, task-oriented dialog and open-domain question answering. LLM-Augmenter significantly reduces ChatGPT's hallucinations without sacrificing the fluency and informativeness of its responses. We make the source code and models publicly available.

In this paper, we introduce a novel dIffusion language modEl pre-training framework for text generation, which we call GENIE. GENIE is a large-scale pretrained diffusion language model that consists of an encoder and a diffusion-based decoder, which can generate text by gradually transforming a random noise sequence into a coherent text sequence. To pre-train GENIE on a large-scale language corpus, we design a new continuous paragraph denoise objective, which encourages the diffusion-decoder to reconstruct a clean text paragraph from a corrupted version, while preserving the semantic and syntactic coherence. We evaluate GENIE on four downstream text generation benchmarks, namely XSum, CNN/DailyMail, Gigaword, and CommonGen. Our experimental results show that GENIE achieves comparable performance with the state-of-the-art autoregressive models on these benchmarks, and generates more diverse text samples. The code and models of GENIE are available at https://github.com/microsoft/ProphetNet/tree/master/GENIE.

Large language models can perform various reasoning tasks by using chain-of-thought prompting, which guides them to find answers through step-by-step demonstrations. However, the quality of the prompts depends on the demonstrations given to the models, and creating many of them by hand is costly. We introduce Synthetic prompting, a method that leverages a few handcrafted examples to prompt the model to generate more examples by itself, and selects effective demonstrations to elicit better reasoning. Our method alternates between a backward and forward process to generate new examples. The backward process generates a question that match a sampled reasoning chain, so that the question is solvable and clear. The forward process produces a more detailed reasoning chain for the question, improving the quality of the example. We evaluate our method on numerical, symbolic, and algorithmic reasoning tasks, and show that it outperforms existing prompting techniques.

Pre-trained language models have achieved promising success in code retrieval tasks, where a natural language documentation query is given to find the most relevant existing code snippet. However, existing models focus only on optimizing the documentation code pairs by embedding them into latent space, without the association of external knowledge. In this paper, we propose a generation-augmented query expansion framework. Inspired by the human retrieval process - sketching an answer before searching, in this work, we utilize the powerful code generation model to benefit the code retrieval task. Specifically, we demonstrate that rather than merely retrieving the target code snippet according to the documentation query, it would be helpful to augment the documentation query with its generation counterpart - generated code snippets from the code generation model. To the best of our knowledge, this is the first attempt that leverages the code generation model to enhance the code retrieval task. We achieve new state-of-the-art results on the CodeSearchNet benchmark and surpass the baselines significantly.

The task of generating code solutions for a given programming problem can benefit from the use of pre-trained language models such as Codex, which can produce multiple diverse samples. However, a major challenge for this task is to select the most appropriate solution from the multiple samples generated by the pretrained language models. A natural way to evaluate the quality and correctness of a code solution is to run it against a set of test cases, but the manual creation of such test cases is often costly and time-consuming. T, that leverages the same pre-trained language models to automatically generate test cases for the code samples, thus reducing the human effort and increasing the coverage of the test scenarios. T then executes the code samples using the generated test cases and performs a dual execution agreement, which considers both the consistency of the outputs against the generated test cases and the agreement of the outputs with other code samples. We conduct comprehensive experiments on four benchmarks, HumanEval, MBPP, APPS, and CodeContests, using five different pre-trained language models with varying sizes and capabilities. T can significantly improve the performance of code solution selection over previous methods, achieving remarkable and consistent gains across different models and benchmarks. T improves the pass@1 metric on HumanEval to 65.8%, which represents an absolute improvement of 18.8% over the code-davinci-002 model, and an absolute improvement of more than 20% over the previous state-of-the-art results. Despite the remarkable progress in pre-training techniques for code generation, selecting a single correct solution from multiple candidates generated by large language models remains a hard problem. For instance, Codex (Chen et al., 2021), a state-of-the-art pre-trained language model for code generation, can achieve a pass@100 (pass if one or more among 100 generated solutions for a given problem can pass the corresponding test cases) of 77.4%, but a pass@1 (correct rate of a single solution) of only 33.5% on the HumanEval benchmark (Chen et al., 2021) A straightforward way to verify the correctness of a solution is to execute it and check if it passes all corresponding test cases. This execution-guided approach has been widely adopted in various code-related tasks, such as code generation (Chen et al., 2021; Li et al., 2022b; Shi et al., 2022), code translation (Roziere et al., 2021), and program synthesis (Chen et al., 2018; Ellis et al., 2019). However, this approach relies heavily on the quality and quantity of test cases, which are often costly and time-consuming to create and maintain. Therefore, we propose to automatically generate test cases for arbitrary programming problems and use them to quickly verify any solution. The first three authors contributed equally. We report the results on the HumanEval benchmark with the Codex model code-cushman-001.

Fine-tuning large language models for different tasks can be costly and inefficient, and even methods that reduce the number of tuned parameters still require full gradient-based optimization. We propose HyperTuning, a novel approach to model adaptation that uses a hypermodel to generate task-specific parameters for a fixed downstream model. We demonstrate a simple setup for hypertuning with HyperT5, a T5-based hypermodel that produces soft prefixes or LoRA parameters for a frozen T5 model from few-shot examples. We train HyperT5 in two stages: first, hyperpretraining with a modified conditional language modeling objective that trains a hypermodel to generate parameters; second, multi-task fine-tuning (MTF) on a large number of diverse language tasks. We evaluate HyperT5 on P3, MetaICL and Super-NaturalInstructions datasets, and show that it can effectively generate parameters for unseen tasks. Moreover, we show that using hypermodel-generated parameters as initializations for further parameter-efficient fine-tuning improves performance. HyperTuning can thus be a flexible and efficient way to leverage large language models for diverse downstream applications.

We introduce GENIUS: a conditional text generation model using sketches as input, which can fill in the missing contexts for a given sketch (key information consisting of textual spans, phrases, or words, concatenated by mask tokens). GENIUS is pre-trained on a large-scale textual corpus with a novel reconstruction from sketch objective using an extreme and selective masking strategy, enabling it to generate diverse and high-quality texts given sketches. Comparison with other competitive conditional language models (CLMs) reveals the superiority of GENIUS's text generation quality. We further show that GENIUS can be used as a strong and ready-to-use data augmentation tool for various natural language processing (NLP) tasks. Most existing textual data augmentation methods are either too conservative, by making small changes to the original text, or too aggressive, by creating entirely new samples. With GENIUS, we propose GeniusAug, which first extracts the target-aware sketches from the original training set and then generates new samples based on the sketches. Empirical experiments on 6 text classification datasets show that GeniusAug significantly improves the models' performance in both in-distribution (ID) and out-of-distribution (OOD) settings. We also demonstrate the effectiveness of GeniusAug on named entity recognition (NER) and machine reading comprehension (MRC) tasks. (Code and models are publicly available at https://github.com/microsoft/SCGLab and https://github.com/beyondguo/genius)

Active learning, which effectively collects informative unlabeled data for annotation, reduces the demand for labeled data. In this work, we propose to retrieve unlabeled samples with a local sensitivity and hardness-aware acquisition function. The proposed method generates data copies through local perturbations and selects data points whose predictive likelihoods diverge the most from their copies. We further empower our acquisition function by injecting the select-worst case perturbation. Our method achieves consistent gains over the commonly used active learning strategies in various classification tasks. Furthermore, we observe consistent improvements over the baselines on the study of prompt selection in prompt-based few-shot learning. These experiments demonstrate that our acquisition guided by local sensitivity and hardness can be effective and beneficial for many NLP tasks.

Reasoning over natural language is a long-standing goal for the research community. However, studies have shown that existing language models are inadequate in reasoning. To address the issue, we present POET, a new pre-training paradigm. Through pre-training language models with programs and their execution results, POET empowers language models to harvest the reasoning knowledge possessed in program executors via a data-driven approach. POET is conceptually simple and can be instantiated by different kinds of programs. In this paper, we show three empirically powerful instances, i.e., POET-Math, POET-Logic, and POET-SQL. Experimental results on six benchmarks demonstrate that POET can significantly boost model performance on natural language reasoning, such as numerical reasoning, logical reasoning, and multi-hop reasoning. Taking the DROP benchmark as a representative example, POET improves the F1 metric of BART from 69.2% to 80.6%. Furthermore, POET shines in giant language models, pushing the F1 metric of T5-11B to 87.6% and achieving a new state-of-the-art performance on DROP. POET opens a new gate on reasoning-enhancement pre-training and we hope our analysis would shed light on the future research of reasoning like program executors.

The dominant paradigm of natural language processing consists of large-scale pre-training on general domain data and adaptation to particular tasks or domains. As we pre-train larger models, conventional fine-tuning, which retrains all model parameters, becomes less feasible. Using GPT-3 175B as an example, deploying many independent instances of fine-tuned models, each with 175B parameters, is extremely expensive. We propose Low-Rank Adaptation, or LoRA, which freezes the pre-trained model weights and injects trainable rank decomposition matrices into each layer of the Transformer architecture, greatly reducing the number of trainable parameters for downstream tasks. For GPT-3, LoRA can reduce the number of trainable parameters by 10,000 times and the computation hardware requirement by 3 times compared to full fine-tuning. LoRA performs on-par or better than fine-tuning in model quality on both GPT-3 and GPT-2, despite having fewer trainable parameters, a higher training throughput, and no additional inference latency. We also provide an empirical investigation into rank-deficiency in language model adaptations, which sheds light on the efficacy of LoRA. We release our implementation in GPT-2 at https://github.com/microsoft/LoRA .