Ian Magnusson


2024

pdf bib
Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research
Luca Soldaini | Rodney Kinney | Akshita Bhagia | Dustin Schwenk | David Atkinson | Russell Authur | Ben Bogin | Khyathi Chandu | Jennifer Dumas | Yanai Elazar | Valentin Hofmann | Ananya Jha | Sachin Kumar | Li Lucy | Xinxi Lyu | Nathan Lambert | Ian Magnusson | Jacob Morrison | Niklas Muennighoff | Aakanksha Naik | Crystal Nam | Matthew Peters | Abhilasha Ravichander | Kyle Richardson | Zejiang Shen | Emma Strubell | Nishant Subramani | Oyvind Tafjord | Evan Walsh | Luke Zettlemoyer | Noah Smith | Hannaneh Hajishirzi | Iz Beltagy | Dirk Groeneveld | Jesse Dodge | Kyle Lo
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Information about pretraining corpora used to train the current best-performing language models is seldom discussed: commercial models rarely detail their data, and even open models are often released without accompanying training data or recipes to reproduce them. As a result, it is challenging to conduct and advance scientific research on language modeling, such as understanding how training data impacts model capabilities and limitations. To facilitate scientific research on language model pretraining, we curate and release Dolma, a three-trillion-token English corpus, built from a diverse mixture of web content, scientific papers, code, public-domain books, social media, and encyclopedic materials. We extensively document Dolma, including its design principles, details about its construction, and a summary of its contents. We present analyses and experimental results on intermediate states of Dolma to share what we have learned about important data curation practices. Finally, we open-source our data curation toolkit to enable reproduction of our work as well as support further research in large-scale data curation.

pdf bib
OLMo: Accelerating the Science of Language Models
Dirk Groeneveld | Iz Beltagy | Evan Walsh | Akshita Bhagia | Rodney Kinney | Oyvind Tafjord | Ananya Jha | Hamish Ivison | Ian Magnusson | Yizhong Wang | Shane Arora | David Atkinson | Russell Authur | Khyathi Chandu | Arman Cohan | Jennifer Dumas | Yanai Elazar | Yuling Gu | Jack Hessel | Tushar Khot | William Merrill | Jacob Morrison | Niklas Muennighoff | Aakanksha Naik | Crystal Nam | Matthew Peters | Valentina Pyatkin | Abhilasha Ravichander | Dustin Schwenk | Saurabh Shah | William Smith | Emma Strubell | Nishant Subramani | Mitchell Wortsman | Pradeep Dasigi | Nathan Lambert | Kyle Richardson | Luke Zettlemoyer | Jesse Dodge | Kyle Lo | Luca Soldaini | Noah Smith | Hannaneh Hajishirzi
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Language models (LMs) have become ubiquitous in both NLP research and in commercial product offerings. As their commercial importance has surged, the most powerful models have become closed off, gated behind proprietary interfaces, with important details of their training data, architectures, and development undisclosed. Given the importance of these details in scientifically studying these models, including their biases and potential risks, we believe it is essential for the research community to have access to powerful, truly open LMs. To this end, we have built OLMo, a competitive, truly Open Language Model, to enable the scientific study of language models. Unlike most prior efforts that have only released model weights and inference code, we release OLMo alongside open training data and training and evaluation code. We hope this release will empower the open research community and inspire a new wave of innovation.

2023

pdf bib
Reproducibility in NLP: What Have We Learned from the Checklist?
Ian Magnusson | Noah A. Smith | Jesse Dodge
Findings of the Association for Computational Linguistics: ACL 2023

Scientific progress in NLP rests on the reproducibility of researchers’ claims. The *CL conferences created the NLP Reproducibility Checklist in 2020 to be completed by authors at submission to remind them of key information to include. We provide the first analysis of the Checklist by examining 10,405 anonymous responses to it. First, we find evidence of an increase in reporting of information on efficiency, validation performance, summary statistics, and hyperparameters after the Checklist’s introduction. Further, we show acceptance rate grows for submissions with more Yes responses. We find that the 44% of submissions that gather new data are 5% less likely to be accepted than those that did not; the average reviewer-rated reproducibility of these submissions is also 2% lower relative to the rest. We find that only 46% of submissions claim to open-source their code, though submissions that do have 8% higher reproducibility score relative to those that do not, the most for any item. We discuss what can be inferred about the state of reproducibility in NLP, and provide a set of recommendations for future conferences, including: a) allowing submitting code and appendices one week after the deadline, and b) measuring dataset reproducibility by a checklist of data collection practices.

2022

pdf bib
Towards a Multi-Entity Aspect-Based Sentiment Analysis for Characterizing Directed Social Regard in Online Messaging
Joan Zheng | Scott Friedman | Sonja Schmer-galunder | Ian Magnusson | Ruta Wheelock | Jeremy Gottlieb | Diana Gomez | Christopher Miller
Proceedings of the Sixth Workshop on Online Abuse and Harms (WOAH)

Online messaging is dynamic, influential, and highly contextual, and a single post may contain contrasting sentiments towards multiple entities, such as dehumanizing one actor while empathizing with another in the same message. These complexities are important to capture for understanding the systematic abuse voiced within an online community, or for determining whether individuals are advocating for abuse, opposing abuse, or simply reporting abuse. In this work, we describe a formulation of directed social regard (DSR) as a problem of multi-entity aspect-based sentiment analysis (ME-ABSA), which models the degree of intensity of multiple sentiments that are associated with entities described by a text document. Our DSR schema is informed by Bandura’s psychosocial theory of moral disengagement and by recent work in ABSA. We present a dataset of over 2,900 posts and sentences, comprising over 24,000 entities annotated for DSR over nine psychosocial dimensions by three annotators. We present a novel transformer-based ME-ABSA model for DSR, achieving favorable preliminary results on this dataset.

pdf bib
Exploring The Landscape of Distributional Robustness for Question Answering Models
Anas Awadalla | Mitchell Wortsman | Gabriel Ilharco | Sewon Min | Ian Magnusson | Hannaneh Hajishirzi | Ludwig Schmidt
Findings of the Association for Computational Linguistics: EMNLP 2022

We conduct a large empirical evaluation to investigate the landscape of distributional robustness in question answering. Our investigation spans over 350 models and 16 question answering datasets, including a diverse set of architectures, model sizes, and adaptation methods (e.g., fine-tuning, adapter tuning, in-context learning, etc.). We find that, in many cases, model variations do not affect robustness and in-distribution performance alone determines out-of-distribution performance.Moreover, our findings indicate thati) zero-shot and in-context learning methods are more robust to distribution shifts than fully fine-tuned models;ii) few-shot prompt fine-tuned models exhibit better robustness than few-shot fine-tuned span prediction models;iii) parameter-efficient and robustness enhancing training methods provide no significant robustness improvements.In addition, we publicly release all evaluations to encourage researchers to further analyze robustness trends for question answering models.

pdf bib
Just-DREAM-about-it: Figurative Language Understanding with DREAM-FLUTE
Yuling Gu | Yao Fu | Valentina Pyatkin | Ian Magnusson | Bhavana Dalvi Mishra | Peter Clark
Proceedings of the 3rd Workshop on Figurative Language Processing (FLP)

Figurative language (e.g., “he flew like the wind”) is challenging to understand, as it is hard to tell what implicit information is being conveyed from the surface form alone. We hypothesize that to perform this task well, the reader needs to mentally elaborate the scene being described to identify a sensible meaning of the language. We present DREAM-FLUTE, a figurative language understanding system that does this, first forming a “mental model” of situations described in a premise and hypothesis before making an entailment/contradiction decision and generating an explanation. DREAM-FLUTE uses an existing scene elaboration model, DREAM, for constructing its “mental model.” In the FigLang2022 Shared Task evaluation, DREAM-FLUTE achieved (joint) first place (Acc@60=63.3%), and can perform even better with ensemble techniques, demonstrating the effectiveness of this approach. More generally, this work suggests that adding a reflective component to pretrained language models can improve their performance beyond standard fine-tuning (3.3% improvement in Acc@60).

2021

pdf bib
Extracting Fine-Grained Knowledge Graphs of Scientific Claims: Dataset and Transformer-Based Results
Ian Magnusson | Scott Friedman
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

Recent transformer-based approaches demonstrate promising results on relational scientific information extraction. Existing datasets focus on high-level description of how research is carried out. Instead we focus on the subtleties of how experimental associations are presented by building SciClaim, a dataset of scientific claims drawn from Social and Behavior Science (SBS), PubMed, and CORD-19 papers. Our novel graph annotation schema incorporates not only coarse-grained entity spans as nodes and relations as edges between them, but also fine-grained attributes that modify entities and their relations, for a total of 12,738 labels in the corpus. By including more label types and more than twice the label density of previous datasets, SciClaim captures causal, comparative, predictive, statistical, and proportional associations over experimental variables along with their qualifications, subtypes, and evidence. We extend work in transformer-based joint entity and relation extraction to effectively infer our schema, showing the promise of fine-grained knowledge graphs in scientific claims and beyond.