Shijia Liu, David A. Smith. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

We explore the task of quotability identification, in which, given a document, we aim to identify which of its passages are the most quotable, i.e. the most likely to be directly quoted by later derived documents. We approach quotability identification as a passage ranking problem and evaluate how well both feature-based and BERT-based (Devlin et al., 2019) models rank the passages in a given document by their predicted quotability. We explore this problem through evaluations on five datasets that span multiple languages (English, Latin) and genres of literature (e.g. poetry, plays, novels) and whose corresponding derived documents are of multiple types (news, journal articles). Our experiments confirm the relatively strong performance of BERT-based models on this task, with the best model, a RoBERTA sequential sentence tagger, achieving an average rho of 0.35 and NDCG@1, 5, 50 of 0.26, 0.31 and 0.40, respectively, across all five datasets.

Code-switching has long interested linguists, with computational work in particular focusing on speech and social media data (Sitaram et al., 2019).This paper contrasts these informal instances of code-switching to its appearance in more formal registers, by examining the mixture of languages in the Deutsches Textarchiv (DTA), a corpus of 1406 primarily German books from the 17th to 19th centuries.We automatically annotate and manually inspect spans of six embedded languages (Latin, French, English, Italian, Spanish, and Greek) in the corpus.We quantitatively analyze the differences between code-switching patterns in these books and those in more typically studied speech and social media corpora.Furthermore, we address the practical task of predicting code-switching from features of the matrix language alone in the DTA corpus.Such classifiers can help reduce errors when optical character recognition or speech transcription is applied to a large corpus with rare embedded languages.

Human communication mediated by computers and Augmented Reality devices will enable us to dynamically express, share and explore new ideas with each other via live simulations as easily as we talk about the weather. This collaboration provides a "shared truth" - what you see is exactly what I see, I see you perform an action as you do it, and we both see exactly the same dynamic transformation of this shared information space. When you express an idea, the computer, a full participant in this conversation, instantly makes it real for both of us enabling us to critique and negotiate the meaning of it. This shared virtual world will be as live, dynamic, pervasive, and visceral as the physical.

This paper proposes a model of information cascades as directed spanning trees (DSTs) over observed documents. In addition, we propose a contrastive training procedure that exploits partial temporal ordering of node infections in lieu of labeled training links. This combination of model and unsupervised training makes it possible to improve on models that use infection times alone and to exploit arbitrary features of the nodes and of the text content of messages in information cascades. With only basic node and time lag features similar to previous models, the DST model achieves performance with unsupervised training comparable to strong baselines on a blog network inference task. Unsupervised training with additional content features achieves significantly better results, reaching half the accuracy of a fully supervised model.

I have been working in VR and interactive 3D for a long time. I have had the pleasure of knowing and working with many of the people that are directly responsible for creating the magic in the world that we live in today. These are the people that started with a virtual blank page and created their own reality. Their vision defined a vector into their future that we have had the privilege of extending into ours. Knowing where this vector started gives us an incredible perspective on where it is today and where it is going. I will describe my personal journey along this vector over the last 35 years, demonstrate a few things I am working on today, and speculate about where this vector into the future may take us.

For topic models, such as LDA, that use a bag-of-words assumption, it becomes especially important to break the corpus into appropriately-sized "documents".Since the models are estimated solely from the term cooccurrences, extensive documents such as books or long journal articles lead to diffuse statistics, and short documents such as forum posts or product reviews can lead to sparsity.This paper describes practical inference procedures for hierarchical models that smooth topic estimates for smaller sections with hyperpriors over larger documents.Importantly for large collections, these online variational Bayes inference methods perform a single pass over a corpus and achieve better perplexity than "flat" topic models on monolingual and multilingual data.Furthermore, on the task of detecting document translation pairs in large multilingual collections, polylingual topic models (PLTM) with multi-level hyperpriors (mlhPLTM) achieve significantly better performance than existing online PLTM models while retaining computational efficiency.

We present a novel approach for efficiently evaluating the performance of retrieval models and introduce two evaluation metrics: Distributional Overlap (DO), which compares the clustering of scores of relevant and non-relevant documents, and Histogram Slope Analysis (HSA), which examines the log of the empirical distributions of relevant and non-relevant documents. Unlike rank evaluation metrics such as mean average precision (MAP) and normalized discounted cumulative gain (NDCG), DO and HSA only require calculating model scores of queries and a fixed sample of relevant and non-relevant documents rather than scoring the entire collection, even implicitly by means of an inverted index. In experimental meta-evaluations, we find that HSA achieves high correlation with MAP and NDCG on a monolingual and a cross-language document similarity task; on four ad-hoc web retrieval tasks; and on an analysis of ten TREC tasks from the past ten years. In addition, when evaluating latent Dirichlet allocation (LDA) models on document similarity tasks, HSA achieves better correlation with MAP and NCDG than perplexity, an intrinsic metric widely used with topic models.

Software distribution and installation is a logistical issue for large enterprises. Web applications are often a good solution because users can instantly receive application updates on any device without needing special permissions to install them on their hardware. Until recently, it was not possible to create 3D multiuser virtual environment-based web applications that didn't require installing a browser plugin. However, recent web standards have made it possible. We present the Virtual World Framework (VWF), a software framework for creating 3D multiuser web applications. We are using VWF to create applications for team training and collaboration. VWF can be downloaded at http://virtual.wf.

Texts propagate through many social networks and provide evidence for their structure. We present efficient algorithms for detecting clusters of reused passages embedded within longer documents in large collections. We apply these techniques to analyzing the culture of reprinting in the United States before the Civil War. Without substantial copyright enforcement, stories, poems, news, and anecdotes circulated freely among newspapers, magazines, and books. From a collection of OCR'd newspapers, we extract a new corpus of reprinted texts, explore the geographic spread and network connections of different publications, and analyze the time dynamics of different genres.

This paper presents a new language model for text recognition in natural images. Many existing techniques incorporate n-gram information as an additional source of information. One problem is that some n-grams are very uncommon, but will still appear in a word across a syllable boundary. These words are given a low probability under an n-gram model. To overcome this problem, we introduce a probabilistic syllable model that uses a probabilistic context-free grammar to generate recognized word labels that are consistent with syllables. In other words, labels generated by this model are pronounceable. This is important for scene text recognition where text often includes proper nouns and standard dictionary information cannot be a useful resource. We show that this language model leads to increased recognition accuracy over a big ram model and discuss the benefits over a dictionary model.

Conventional retrieval systems view documents as a unit and look at different retrieval types within a document. We introduce Proteus, a frame-work for seamlessly navigating books as dynamic collections which are defined on the fly. Proteus allows us to search various retrieval types. Navigable types include pages, books, named persons, locations, and pictures in a collection of books taken from the Internet Archive. The demonstration shows the value of multi-type browsing in dynamic collections to peruse new data.

Many forms of linguistic analysis, such as part of speech tagging, named entity recognition, and other sequence labeling tasks are performed on short spans of text and assume statistical dependence within a window of only a few tokens. We propose using passage retrieval to induce non-local dependencies in structured classification that generalizes earlier work in context aggregation for named-entity recognition. We introduce a new method for feature expansion inspired by psuedo-relevance feedback (PRF). Our results on the CoNLL 2003 task show that features from cross-document feature expansion improves NER effectiveness over previous aggregation models. Utilizing all the tokens in a sentence for query context consistently perform best on both intrinsic and extrinsic evaluations. Tagging models incorporating feature expansion outperform the leading NER system when evaluated on out of domain data, a collection of publicly available scanned books on the topic of historic Deerfield, MA. Finally, the results show that retrieval based feature expansion using an external collection of unlabeled text can result in further effectiveness improvements.

Recent studies suggest that associative browsing can be beneficial for personal information access. Associative browsing is intuitive for the user and complements other methods of accessing personal information, such as keyword search. In our previous work, we proposed an associative browsing model of personal information in which users can navigate through the space of documents and concepts (e.g., person names, events, etc.). Our approach differs from other systems in that it presented a ranked list of associations by combining multiple measures of similarity, whose weights are improved based on click feedback from the user.

Incorporating syntactic features in a retrieval model has had very limited success in the past, with the exception of binary term dependencies. This paper presents a new term dependency modeling approach based on syntactic dependency parsing for both queries and documents. Our model is inspired by a quasi-synchronous stochastic process for machine translation[21]. We model four different types of relationships between syntactically dependent term pairs to perform inexact matching between documents and queries. We also propose a machine learning technique for predicting optimal parameter settings for a retrieval model incorporating syntactic relationships. The results on TREC collections show that the quasi-synchronous dependence model can improve retrieval performance and outperform a strong state-of-art sequential dependence baseline when we use predicted optimal parameters.

Marking up queries with annotations such as part-of-speech tags, capitalization, and segmentation, is an important part of many approaches to query processing and understanding. Due to their brevity and idiosyncratic structure, search queries pose a challenge to existing annotation tools that are commonly trained on full-length documents. To address this challenge, we view the query as an explicit representation of a latent information need, which allows us to use pseudo-relevance feedback, and to leverage additional information from the document corpus, in order to improve the quality of query annotation.

A typical collection of personal information contains many documents and mentions many concepts (e.g., person names, events, etc.). In this environment, associative browsing between these concepts and documents can be useful as a complement for search. Previous approaches in the area of semantic desktops aimed at addressing this task. However, they were not practical because they require tedious manual annotation by the user.

Query reformulation modifies the original query with the aim of better matching the vocabulary of the relevant documents, and consequently improving ranking effectiveness. Previous techniques typically generate words and phrases related to the original query, but do not consider how these words and phrases would fit together in new queries. In this paper, we focus on an implementation of an approach that models reformulation as a distribution of queries, where each query is a variation of the original query. This approach considers a query as a basic unit and can capture important dependencies between words and phrases in the query. The implementation discussed here is based on passage analysis of the target corpus. Experiments on the TREC collection show that the proposed model for query reformulation significantly outperforms state-of-the-art methods.

Online communities are valuable information sources where knowledge is accumulated by interactions between people. Search services provided by online community sites such as forums are often, however, quite poor. To address this, we investigate retrieval techniques that exploit the hierarchical thread structures in community sites. Since these structures are sometimes not explicit or accurately annotated, we use structure discovery techniques. We then make use of thread structures in retrieval experiments. Our results show that using thread structures that have been accurately annotated can lead to significant improvements in retrieval performance compared to strong baselines.

Modeling term dependence has been shown to have a significant positive impact on retrieval. Current models, however, use sequential term dependencies, leading to an increased query latency, especially for long queries. In this paper, we examine two query segmentation models that reduce the number of dependencies. We find that two-stage segmentation based on both query syntactic structure and external information sources such as query logs, attains retrieval performance comparable to the sequential dependence model, while achieving a 50% reduction in query latency.

Previews and overviews of large, heterogeneous information resources help users comprehend the scope of collections and focus on particular subsets of interest. For narrative docu-ments, questions of “what happened? where? and when?” are natural points of entry. Building on our earlier work at the Perseus Project with detecting terms, place names, and dates, we have exploited co-occurrences of dates and place names to detect and describe likely events in document col-lections. We compare statistical measures for determining the relative significance of various events. We have built in-terfaces that help users preview likely regions of interest for a given range of space and time by plotting the distribution and relevance of various collocations. Users can also control the amount of collocation information in each view. Once particular collocations are selected, the system can identify key phrases associated with each possible event to organize browsing of the documents themselves.

The background needed to understand the integrated-optic collinear acoustically tunable optical filters (ATOFs) and the fabrication and performance of both simple and multielement acoustooptic tunable filters is presented. The most important sources of crosstalk between channels simultaneously selected by a single device are discussed. ATOFs have the combined virtues of narrow passband (subnanometer bandwidth), broad tuning range (hundreds of nanometers have been demonstrated), and simultaneous independent multiple-channel filtering. The theory and practice of collinear integrated optic acoustooptic filters is presented. Devices which involve higher-order integration are discussed, including multiple-state filters for enhanced sidelobe suppression and polarization-independent configurations. Various sources of interchannel crosstalk are derived and demonstrated in multiwavelength filtering experiments.>

We present a general acceleration algorithm for alternating and monotone scalar sequences and series.The main components of the algomthm are the subroutines WHIZ and HURRY.WHIZ is a recursive implementation of Levin's u transform, and HURRY {which calls WHIZ) estimates truncation and round-off errors to make a near-optimal stopping decision and provide a very good estimate of the accuracy of the computed answer.We also present a test driver program that demonstrates the capabilities of HURRY when applied to a wide varmty of convergent and divergent sequences and series Categories and Subject Descriptors: G 1 0 [Numerical Analysis]: General--numerical algortthms; G

The algorithm presented here accompanies [1], which gives the description, test results, and references.The algorithm consists of three FORTRAN subroutines, HURRY(), WHIZ(), and W H I Z l ( ) .HURRY( ) controls the logical flow, makes the stopping decision, and calls both WHIZ( ) and VCHIZI(), which implement Levin's u transform for acceleration of convergence.The two implementations are identical except that W H I Z l ( ) avoids recomputation of initial information, and thus is called only after WHIZ( ) has been called once.Also included with the algorithm is a demonstration driver program called XACCEL, which is described in [1, Sect.4], and sample input data that were used to produce the output summarized in [1, Table I].