Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

By Bevan K. Jones, Mark W. Johnson, Sharon Goldwater. 2018

132 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention (2023, neutral): “PT denotes pre-training on additional datasets [8, 29, 58, 7, 57], FT denotes fine-tuning on COCO Caption.”
Neural Approaches to Conversational AI (2018, neutral): “• Persona-Chat dataset: This crowdsourced dataset (Zhang et al., 2018c) was developed to meet the need for conversational data where dialogues exhibit distinct user personas.”
PaLM 2 Technical Report (2023, neutral): “…and generation capabilities over the last few years, leading to breakthrough performance in reasoning, math, science, and language tasks (Howard & Ruder, 2018; Brown et al., 2020; Du et al., 2022; Chowdhery et al., 2022; Rae et al., 2021; Lewkowycz et al., 2022; Tay et al., 2023;…”
Simplifying Graph Convolutional Networks (2019, positive): “We replace the 4layer, highway-connection GCN with a 3rd degree propagation matrix (K = 3) SGC and use the same set of hyperparameters as Rahimi et al. (2018). All experiments on the GEOTEXT dataset are conducted on a single Nvidia GTX-1080Ti GPU while the ones on the TWITTER-NA and TWITTER-WORLD da”
Quark: Controllable Text Generation with Reinforced Unlearning (2022, neutral): “Additionally, we also conduct out-of-domain evaluation with the WRITINGPROMPTS dataset [15], which is created for creative writing (i.”
Planning with Large Language Models for Code Generation (2023, unknown): “Transformer-based pipelines like AlphaCode follow the tradition of natural language processing and use sampling methods (Fan et al., 2018; Dabre & Fujita, 2020) during the generation process.”
The Effect of Natural Distribution Shift on Question Answering Models (2020, unknown): “Following Rajpurkar et al. (2016), we automatically categorize each answer into numerical and non-numerical answers, named entities, and constituents us-ing spaCy (Honnibal & Montani, 2017) and the constituency parser from Kitaev & Klein (2018).”
Can neural networks acquire a structural bias from raw linguistic data? (2020, unknown): “However, rather than training a low bias model from scratch to perform a particular linguistic task, recent results show that it is far more more effective to pretrain a general purpose model on raw data and subsequently fine-tune it on a downstream task (e.g. Howard & Ruder, 2018).”
Unsupervised Paraphrase Generation using Pre-trained Language Models (2020, unknown): “To generate more than one paraphrase for a given input sentence, we use top-k sampling [24].”
Biases in Large Language Models: Origins, Inventory, and Discussion (2023, unknown): “Indeed, language models are rarely used “as is”; instead, they are adapted to the task of interest by either fine-tuning [65, 109] on smaller, task-specific datasets, or by designing prompts [81], usually in natural language, to work in a zero-shot or few-shot setting.”
Finetuned Language Models Are Zero-Shot Learners (2021, unknown): “Our work is broadly inspired by a long line of prior work on language models for NLP applications (Dai & Le, 2015; Peters et al., 2018; Howard & Ruder, 2018; Radford et al., 2018; 2019, inter alia).”
Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing (2021, unknown): “We summarize representative methods in the “Multi-prompt Learning” section of Figure 1 as well as Figure 2.”
Self-Consistency Improves Chain of Thought Reasoning in Language Models (2022, unknown): “, 1985; Ficler & Goldberg, 2017), top-k sampling (Fan et al., 2018; Holtzman et al., 2018; Radford et al., 2019), nucleus sampling (Holtzman et al.”
Machine Learning in Python: Main developments and technology trends in data science, machine learning, and artificial intelligence (2020, unknown): “While machine learning enables us to push the state-of-the-art in many fields such as natural language processing [117,124,165,166] and computer vision [132,167–169], certain applications involve sensitive data that demands responsible treatment.”
TAPEX: Table Pre-training via Learning a Neural SQL Executor (2021, unknown): “, SQL), they have difficulties in training due to the large search space and the presence of spurious programs (Goldman et al., 2018).”
Recent Advances in Natural Language Processing via Large Pre-trained Language Models: A Survey (2021, unknown): “Pre-training and fine-tuning entire models did not gain popularity in NLP until the advent of ELMo [134] and ULMFiT [66].”