United States

Distillation efforts have led to language models that are more compact and efficient without serious drops in performance. The standard approach to distillation trains a student model against two objectives: a task-specific objective (e.g., language modeling) and an imitation objective that encourages the hidden states of the student model to be similar to those of the larger teacher model. In this paper, we show that it is beneficial to augment distillation with a third objective that encourages the student to imitate the \emphcausal dynamics of the teacher through a distillation interchange intervention training objective (DIITO). DIITO pushes the student model to become a \emphcausal abstraction of the teacher model -- a faithful model with simpler causal structure. DIITO is fully differentiable, easily implemented, and combines flexibly with other objectives. Compared against standard distillation with the same setting, DIITO results in lower perplexity on the WikiText-103M corpus (masked language modeling) and marked improvements on the GLUE benchmark (natural language understanding), SQuAD (question answering), and CoNLL-2003 (named entity recognition).

NAACL 2022

Causal Distillation for Language Models

interchange intervention training

intervention analysis

model distillation

causality

language model

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2021

poster

## Welcome to NAACL 2022!
The Annual Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL) is taking place July 10-15, 2022 as a hybrid event, in Seattle, WA and online. We are happy to welcome all of you to conference! 
 
The main conference program features oral presentations, in-person and virtual posters and demo sessions, a plenary session for our best paper presentations and awards, keynote presentations, a plenary panel on the place of linguistics and symbolic structures in NLP, and an Industrial Track panel on Careers in NLP.
 
Posters (including Findings of NAACL 2022) and demos are grouped by areas for both the in-person and the virtual sessions. For the virtual component, the talks will be on Zoom and the posters and the demos will be in GatherTown. The Student Research Workshop and will have an oral session and a poster session.
 
The program also features 6 Tutorials and 26 Workshops.
 
 We wish you a wonderful conference!
 
[The NAACL 2022 Organizing Committee](https://2022.naacl.org/committees/organization/)

2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics

Word embeddings are one of the most fundamental technologies used in natural language processing. Existing word embeddings are high-dimensional and consume considerable computational resources. In this study, we propose WordTour, unsupervised one-dimensional word embeddings. To achieve the challenging goal, we propose a decomposition of the desiderata of word embeddings into two parts, completeness and soundness, and focus on soundness in this paper. Owing to the single dimensionality, WordTour is extremely efficient and provides a minimal means to handle word embeddings. We experimentally confirmed the effectiveness of the proposed method via user study and document classification.

Word Tour: One-dimensional Word Embeddings via the Traveling Salesman Problem

We target on the document-level relation extraction in an end-to-end setting, where the model needs to jointly perform mention extraction, coreference resolution (COREF) and relation extraction (RE) at once, and gets evaluated in an entity-centric way. Especially, we address the two-way interaction between COREF and RE that has not been the focus by previous work, and propose to introduce explicit interaction namely Graph Compatibility (GC) that is specifically designed to leverage task characteristics, bridging decisions of two tasks for direct task interference. Our experiments are conducted on DocRED and DWIE; in addition to GC, we implement and compare different multi-task settings commonly adopted in previous work, including pipeline, shared encoders, graph propagation, to examine the effectiveness of different interactions. The result shows that GC achieves the best performance by up to 2.3/5.1 F1 improvement over the baseline.

Modeling Task Interactions in Document-Level Joint Entity and Relation Extraction

General domain Named Entity Recognition (NER) datasets like CoNLL-2003 mostly annotate coarse-grained location entities such as a country or a city. But many applications require identifying fine-grained locations from texts and mapping them precisely to geographic sites, e.g., a crossroad, an apartment building, or a grocery store. In this paper, we introduce a new dataset HarveyNER with fine-grained locations annotated in tweets. This dataset presents unique challenges and characterizes many complex and long location mentions in informal descriptions. We built strong baseline models using Curriculum Learning and experimented with different heuristic curricula to better recognize difficult location mentions. Experimental results show that the simple curricula can improve the system's performance on hard cases and its overall performance, and outperform several other baseline systems. The dataset and the baseline models can be found at https://github.com/brickee/HarveyNER.

Crossroads, Buildings and Neighborhoods: A Dataset for Fine-grained Location Recognition

Semantic typing aims at classifying tokens or spans of interest in a textual context into semantic categories such as relations, entity types, and event types. The inferred labels of semantic categories meaningfully interpret how machines understand components of text. In this paper, we present UniST, a unified framework for semantic typing that captures label semantics by projecting both inputs and labels into a joint semantic embedding space. To formulate different lexical and relational semantic typing tasks as a unified task, we incorporate task descriptions to be jointly encoded with the input, allowing UniST to be adapted to different tasks without introducing task-specific model components. UniST optimizes a margin ranking loss such that the semantic relatedness of the input and labels is reflected from their embedding similarity. Our experiments demonstrate that UniST achieves strong performance across three semantic typing tasks: entity typing, relation classification and event typing. Meanwhile, UniST effectively transfers semantic knowledge of labels and substantially improves generalizability on inferring rarely seen and unseen types. In addition, multiple semantic typing tasks can be jointly trained within the unified framework, leading to a single compact multi-tasking model that performs comparably to dedicated single-task models, while offering even better transferability.

Unified Semantic Typing with Meaningful Label Inference

As a fundamental task in natural language processing, named entity recognition (NER) aims to locate and classify named entities in unstructured text. However, named entities are always the minority among all tokens in the text. This data imbalance problem presents a challenge to machine learning models as their learning objective is usually dominated by the majority of non-entity tokens. To alleviate data imbalance, we propose a set of sentence-level resampling methods  where the importance of each training sentence is computed based on its tokens and entities. We study the generalizability of these resampling methods on a wide variety of NER models (CRF, Bi-LSTM, and BERT) across corpora from diverse domains (general, social, and medical texts). Extensive experiments show that the proposed methods improve span-level macro F1-scores of the evaluated NER models on multiple corpora, frequently outperforming sub-sentence-level resampling, data augmentation, and special loss functions such as focal and Dice loss.

Sentence-Level Resampling for Named Entity Recognition

Recent work incorporates pre-trained word embeddings such as BERT embeddings into Neural Topic Models (NTMs), generating highly coherent topics. However, with high-quality contextualized document representations, do we really need sophisticated neural models to obtain coherent and interpretable topics? In this paper, we conduct thorough experiments showing that directly clustering high-quality sentence embeddings with an appropriate word selecting method can generate more coherent and diverse topics than NTMs, achieving also higher efficiency and simplicity.

Is Neural Topic Modelling Better than Clustering? An Empirical Study on Clustering with Contextual Embeddings for Topics

Many clinical informatics tasks that are based on electronic health records (EHR) need relevant patient cohorts to be selected based on findings, symptoms and diseases. Frequently, these conditions are described in radiology reports which can be retrieved using information retrieval (IR) methods. The latest of these techniques utilize neural IR models such as BERT trained on clinical text. However, these methods still lack semantic understanding of the underlying clinical conditions as well as ruled out findings, resulting in poor precision during retrieval. In this paper we combine clinical finding detection with supervised query match learning. Specifically, we use lexicon-driven concept detection to detect relevant findings in sentences. These findings are used as queries to train a Sentence-BERT (SBERT) model using triplet loss on matched and unmatched query-sentence pairs. We show that the proposed supervised training task remarkably improves the retrieval performance of SBERT. The trained model generalizes well to unseen queries and reports from different collections.

Improving Neural Models for Radiology Report Retrieval with Lexicon-based Automated Annotation

With the growing popularity of deep-learning models, model understanding becomes more important. Much effort has been devoted to demystify deep neural networks for better explainability. Some feature attribution methods have shown promising results in computer vision, especially the gradient-based methods where effectively smoothing the gradients with reference data is the key to a robust and faithful result. However, direct application of these gradient-based methods to NLP tasks is not trivial due to the fact that the input consists of discrete tokens and the “reference” tokens are not explicitly defined. In this work, we propose Locally Aggregated Feature Attribution (LAFA), a novel gradient-based feature attribution method for NLP models. Instead of relying on obscure reference tokens, it smooths gradients by aggregating similar reference texts derived from language model embeddings. For evaluation purpose, we also design experiments on different NLP tasks including Entity Recognition and Sentiment Analysis on public datasets and key words detection on constructed Amazon catalogue dataset. The superior performance of the proposed method is demonstrated through experiments. 

Locally Aggregated Feature Attribution on Natural Language Model Understanding

Although self-attention based models such as Transformers have achieved remarkable successes on natural language processing (NLP)
tasks, recent studies reveal that they have limitations on modeling sequential transformations (Hahn, 2020), which may prompt
re-examinations of recurrent neural networks (RNNs) that demonstrated impressive results on handling sequential data. Despite many
prior attempts to interpret RNNs, their internal mechanisms have not been fully understood, and the question on how exactly they capture
sequential features remains largely unclear. In this work, we present a study that shows there actually exist some explainable components
that reside within the hidden states, which are reminiscent of the classical n-grams features. We evaluated such extracted explainable features from trained RNNs on downstream sentiment analysis tasks and found they could be used to model interesting linguistic phenomena such as negation and intensification. Furthermore, we examined the efficacy of using such n-gram components alone as encoders on tasks such as sentiment analysis and language modeling, revealing they could be playing important roles in contributing to the overall performance of RNNs. We hope our findings could add interpretability to RNN architectures, and also provide inspirations for proposing new architectures for sequential data.

Implicit n-grams Induced by Recurrence

Stories or narratives are comprised of a sequence of events. To compose interesting stories, professional writers often leverage a creative writing technique called *flashback* that inserts past events into current storylines as we commonly observe in novels and plays. However,  it is challenging for machines to generate *flashback* as it requires a solid understanding of event **temporal order** (e.g. *feeling hungry* before *eat*, not vice versa), and the creativity to arrange storylines so that earlier events do not always appear first in **narrative order**. Two major issues in existing systems that exacerbate the challenges: 1) temporal bias in pertaining and story datasets that leads to monotonic event temporal orders; 2) lack of explicit guidance that helps machines decide where to insert *flashbacks*. We propose to address these issues using structured storylines to encode events and their pair-wise temporal relations (before, after and vague) as **temporal prompts** that guide how stories should unfold temporally. We leverage a Plan-and-Write framework enhanced by reinforcement learning to generate storylines and stories end-to-end. Evaluation results show that the proposed method can generate more interesting stories with *flashbacks* while maintaining textual diversity, fluency, and temporal coherence.

Downloads

Next from NAACL 2022

Word Tour: One-dimensional Word Embeddings via the Traveling Salesman Problem

Similar lecture

Reinforced Multi-Teacher Selection for Knowledge Distillation