Related papers: Bridging the Training-Inference Gap in LLMs by Lev…

Large language models (LMs) such as GPT-3 have the surprising ability to do in-context learning, where the model learns to do a downstream task simply by conditioning on a prompt consisting of input-output examples. The LM learns from these…

Large Language Models (LLMs) are increasingly applied to complex tasks that require extended reasoning. In such settings, models often benefit from diverse chains-of-thought to arrive at multiple candidate solutions. This requires two…

Neural text generation models conditioning on given input (e.g. machine translation and image captioning) are usually trained by maximum likelihood estimation of target text. However, the trained models suffer from various types of errors…

Language models (LMs) are pretrained to imitate internet text, including content that would violate human preferences if generated by an LM: falsehoods, offensive comments, personally identifiable information, low-quality or buggy code, and…

Although large language models (LLMs) have transformed AI, they still make mistakes and can explore unproductive reasoning paths. Self-correction capability is essential for deploying LLMs in safety-critical applications. We uncover a…

Large language models such as GPT and Llama are trained with a next-token prediction loss. In this work, we suggest that training language models to predict multiple future tokens at once results in higher sample efficiency. More…

Alignment training has tradeoffs: it helps language models (LMs) gain in reasoning and instruction following but might lose out on skills such as creativity and calibration, where unaligned base models are better at. We aim to make the best…

Large language models (LLMs) have demonstrated remarkable capabilities in generating high-quality texts across diverse domains. However, the potential misuse of LLMs has raised significant concerns, underscoring the urgent need for reliable…

This paper introduces a simple and scalable approach to improve the data efficiency of large language model (LLM) training by augmenting existing text data with thinking trajectories. The compute for pre-training LLMs has been growing at an…

Recent years have seen significant advancements in foundation models through generative pre-training, yet algorithmic innovation in this space has largely stagnated around autoregressive models for discrete signals and diffusion models for…

Given a question, a language model (LM) implicitly encodes a distribution over possible answers. In practice, post-training procedures for LMs often collapse this distribution onto a single dominant mode. While this is generally not a…

Scheduled sampling is widely used to mitigate the exposure bias problem for neural machine translation. Its core motivation is to simulate the inference scene during training by replacing ground-truth tokens with predicted tokens, thus…

Standard language models generate text by selecting tokens from a fixed, finite, and standalone vocabulary. We introduce a novel method that selects context-aware phrases from a collection of supporting documents. One of the most…

Self-rationalizing models that also generate a free-text explanation for their predicted labels are an important tool to build trustworthy AI applications. Since generating explanations for annotated labels is a laborious and costly pro…

In the context of knowledge-driven seq-to-seq generation tasks, such as document-based question answering and document summarization systems, two fundamental knowledge sources play crucial roles: the inherent knowledge embedded within model…

In the absence of abundant reliable annotations for challenging tasks and contexts, how can we expand the frontier of LLM capabilities with potentially wrong answers? We focus on two research questions: (1) Can LLMs generate reliable…

While large language models (LLMs) such as ChatGPT and PaLM have demonstrated remarkable performance in various language understanding and generation tasks, their capabilities in complex reasoning and intricate knowledge utilization still…

Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by…

Chain-of-thought (CoT) rationale enables language models to use additional task-related text for problem-solving, benefiting not only from detailed reasoning steps but also from the expanded computational space of longer inputs. Prior work…

Integrating free-text explanations to in-context learning of large language models (LLM) is shown to elicit strong reasoning capabilities along with reasonable explanations. In this paper, we consider the problem of leveraging the…