Papers by Annie Louis
LAIT: Efficient Multi-Segment Encoding in Transformers with Layer-Adjustable Interaction (2023.acl-long)
Copied to clipboard
Jeremiah Milbauer, Annie Louis, Mohammad Javad Hosseini, Alex Fabrikant, Donald Metzler, Tal Schuster
| Challenge: | In many NLP tasks, the input text can be seen as a sequence of related segments. |
| Approach: | They propose a layer-adjustable interactions framework that contextualizes token representations by attending to all other tokens at each layer, leading to quadratic increase in compute effort with the input length. |
| Outcome: | The proposed model reduces 30-50% of attention FLOPs while maintaining high accuracy. |
Countering the Effects of Lead Bias in News Summarization via Multi-Stage Training and Auxiliary Losses (D19-1)
Copied to clipboard
| Challenge: | Sentence position is a strong feature for news summarization, since the lead often summarizes the key points of the article. |
| Approach: | They propose two techniques to make neural systems sensitive to the importance of content in different parts of the article by using random shuffled sentences to pretrain the model. |
| Outcome: | The proposed techniques improve the performance of a competitive reinforcement learning based extractive system, with the auxiliary loss being more powerful than pretraining. |
Deep Dungeons and Dragons: Learning Character-Action Interactions from Role-Playing Game Transcripts (N18-2)
Copied to clipboard
| Challenge: | a novel approach to understanding narratives involves modelling the interaction between characters and actions . we propose role-playing games as a testbed for inferring interactions between characters in narratives . |
| Approach: | They propose role-playing games as a testbed for learning latent ties between characters and actions . they propose to combine character and action descriptions from online discussion forums . |
| Outcome: | The proposed model can capture interactions between characters and actions in narratives . it can predict actions better when character attributes are taken into account . |
A synthetic data approach for domain generalization of NLI models (2024.acl-long)
Copied to clipboard
| Challenge: | Natural Language Inference (NLI) datasets are important benchmark tasks for LLMs . however, their realistic performance on out-of-distribution/domain data is less well-understood . a T5-small model trained with our data improves around 7% on average compared to the best alternative dataset . |
| Approach: | They propose a new approach for generating NLI data in diverse domains and lengths . they show that models trained on this data have the best generalization to completely new downstream test settings . |
| Outcome: | The proposed model can be trained on datasets with high-quality examples with meaningful premises and high accuracy. |
OpineSum: Entailment-based self-training for abstractive opinion summarization (2023.findings-acl)
Copied to clipboard
| Challenge: | Abstractive summarization is promising for fluently comparing opinions from a set of reviews about a place or product. |
| Approach: | They propose a novel method that automatically leverages common opinions across reviews to create powerful abstractive models. |
| Outcome: | The proposed method outperforms strong peer systems in both settings. |
TESA: A Task in Entity Semantic Aggregation for Abstractive Summarization (2020.emnlp-main)
Copied to clipboard
| Challenge: | Abstractive summarization systems focus on paraphrasing and simplifying the source content, to the exclusion of such semantic abstraction capabilities. |
| Approach: | They propose a dataset and task to fine tune an abstractive summarization model to generate aggregations of 5.3K entities from a crowd-sourced dataset. |
| Outcome: | The proposed task and dataset show that the proposed model can generate aggregations at a semantic level, but that it is too complex to use. |
Little Red Riding Hood Goes around the Globe: Crosslingual Story Planning and Generation with Large Language Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing work has demonstrated the effectiveness of planning for story generation exclusively in a monolingual setting focusing primarily on English. |
| Approach: | They propose a task of crosslingual story generation with planning to leverage the creative and reasoning capabilities of large pretrained language models to generate stories in multiple languages. |
| Outcome: | The proposed task combines planning and planning in a monolingual setting and demonstrates that plans which structure stories into three acts lead to more coherent and interesting narratives while allowing to explicitly control their content and structure. |
“I’d rather just go to bed”: Understanding Indirect Answers (2020.emnlp-main)
Copied to clipboard
| Challenge: | Humans produce and interpret complex utterances even in simple scenarios. |
| Approach: | They present a large-scale English language corpus with 34,268 (polar question, indirect answer) pairs to enable progress on this task. |
| Outcome: | The proposed corpus contains 34,268 (polar question, indirect answer) pairs, and reaches 82-88% accuracy for a 4-class distinction, and 64-85% for 6 classes. |
Getting to “Hearer-old”: Charting Referring Expressions Across Time (D18-1)
Copied to clipboard
| Challenge: | a study of how expressions that refer to an entity develop over time is done . we track thousands of person and organization entities over 20 years of NYT . |
| Approach: | They track thousands of person and organization entities over 20 years of the NYT . they find that referring expressions evolve over time as entities move from hearer-new to hearer old . |
| Outcome: | The proposed model improves on a majority-class baseline by 10-30% . it shows that the expressions evolve as the entity becomes accepted into common knowledge . |
Conditional Generation with a Question-Answering Blueprint (2023.tacl-1)
Copied to clipboard
Shashi Narayan, Joshua Maynez, Reinald Kim Amplayo, Kuzman Ganchev, Annie Louis, Fantine Huot, Anders Sandholm, Dipanjan Das, Mirella Lapata
| Challenge: | Neural generation models often struggle to identify which content units are salient. |
| Approach: | They propose a new conceptualization of text plans as a sequence of question-answer pairs . they propose QA blueprints as QA proxy for content selection and planning . |
| Outcome: | The proposed model improves existing datasets with QA blueprints as proxy for content selection and planning. |
Resolving Indirect Referring Expressions for Entity Selection (2023.acl-long)
Copied to clipboard
| Challenge: | Recent advances in language modeling have enabled new conversational systems. |
| Approach: | They propose to use a dataset of indirect referring expressions to solve the problem of reference resolution when people use natural expressions . they propose to model the problem using 42K indirect referred expressions across three domains and a public dataset of entity pairs and utterances. |
| Outcome: | The proposed models achieve 82%-87% accuracy in realistic settings, while reasonable invites further advances. |
Source-summary Entity Aggregation in Abstractive Summarization (2022.coling-1)
Copied to clipboard
| Challenge: | Existing studies on the semantics of text generated by abstractive summarization systems have focused on summary n-grams that are not found in the source text. |
| Approach: | They study how entities from a source text can be referred to in later discourse by a more general description. |
| Outcome: | The proposed method shows that state-of-the-art summarization systems produce semantically correct aggregations. |