Search

Network point processes often exhibit latent structure that govern the behaviour of the sub-processes. It is not always reasonable to assume that this latent structure is static, and detecting when and how this driving structure changes is often of interest. In this paper, we introduce a novel online methodology for detecting changes within the latent structure of a network point process. We focus on block-homogeneous Poisson processes, where latent node memberships determine the rates of the edge processes. We propose a scalable variational procedure which can be applied on large networks in an online fashion via a Bayesian forgetting factor applied to sequential variational approximations to the posterior distribution. The proposed framework is tested on simulated and real-world data, and it rapidly and accurately detects changes to the latent edge process rates, and to the latent node group memberships, both in an online manner. In particular, in an application on the Santander Cycles bike-sharing network in central London, we detect changes within the network related to holiday periods and lockdown restrictions between 2019 and 2020., Comment: 34 pages, 17 figures

The most popular Cross-Document Event Coreference Resolution (CDEC) datasets fail to convey the true difficulty of the task, due to the lack of lexical diversity between coreferring event triggers (words or phrases that refer to an event). Furthermore, there is a dearth of event datasets for figurative language, limiting a crucial avenue of research in event comprehension. We address these two issues by introducing ECB+META, a lexically rich variant of Event Coref Bank Plus (ECB+) for CDEC on symbolic and metaphoric language. We use ChatGPT as a tool for the metaphoric transformation of sentences in the documents of ECB+, then tag the original event triggers in the transformed sentences in a semi-automated manner. In this way, we avoid the re-annotation of expensive coreference links. We present results that show existing methods that work well on ECB+ struggle with ECB+META, thereby paving the way for CDEC research on a much more challenging dataset. Code/data: https://github.com/ahmeshaf/llms_coref, Comment: Short Paper, ACL 2024

This paper is dedicated to the design and evaluation of the first AMR parser tailored for clinical notes. Our objective was to facilitate the precise transformation of the clinical notes into structured AMR expressions, thereby enhancing the interpretability and usability of clinical text data at scale. Leveraging the colon cancer dataset from the Temporal Histories of Your Medical Events (THYME) corpus, we adapted a state-of-the-art AMR parser utilizing continuous training. Our approach incorporates data augmentation techniques to enhance the accuracy of AMR structure predictions. Notably, through this learning strategy, our parser achieved an impressive F1 score of 88% on the THYME corpus's colon cancer dataset. Moreover, our research delved into the efficacy of data required for domain adaptation within the realm of clinical notes, presenting domain adaptation data requirements for AMR parsing. This exploration not only underscores the parser's robust performance but also highlights its potential in facilitating a deeper understanding of clinical narratives through structured semantic representations., Comment: Accepted to the 6th Clinical NLP Workshop at NAACL, 2024

Event coreference resolution (ECR) is the task of determining whether distinct mentions of events within a multi-document corpus are actually linked to the same underlying occurrence. Images of the events can help facilitate resolution when language is ambiguous. Here, we propose a multimodal cross-document event coreference resolution method that integrates visual and textual cues with a simple linear map between vision and language models. As existing ECR benchmark datasets rarely provide images for all event mentions, we augment the popular ECB+ dataset with event-centric images scraped from the internet and generated using image diffusion models. We establish three methods that incorporate images and text for coreference: 1) a standard fused model with finetuning, 2) a novel linear mapping method without finetuning and 3) an ensembling approach based on splitting mention pairs by semantic and discourse-level difficulty. We evaluate on 2 datasets: the augmented ECB+, and AIDA Phase 1. Our ensemble systems using cross-modal linear mapping establish an upper limit (91.9 CoNLL F1) on ECB+ ECR performance given the preprocessing assumptions used, and establish a novel baseline on AIDA Phase 1. Our results demonstrate the utility of multimodal information in ECR for certain challenging coreference problems, and highlight a need for more multimodal resources in the coreference resolution space., Comment: To appear at LREC-COLING 2024

Event Coreference Resolution (ECR) as a pairwise mention classification task is expensive both for automated systems and manual annotations. The task's quadratic difficulty is exacerbated when using Large Language Models (LLMs), making prompt engineering for ECR prohibitively costly. In this work, we propose a graphical representation of events, X-AMR, anchored around individual mentions using a \textbf{cross}-document version of \textbf{A}bstract \textbf{M}eaning \textbf{R}epresentation. We then linearize the ECR with a novel multi-hop coreference algorithm over the event graphs. The event graphs simplify ECR, making it a) LLM cost-effective, b) compositional and interpretable, and c) easily annotated. For a fair assessment, we first enrich an existing ECR benchmark dataset with these event graphs using an annotator-friendly tool we introduce. Then, we employ GPT-4, the newest LLM by OpenAI, for these annotations. Finally, using the ECR algorithm, we assess GPT-4 against humans and analyze its limitations. Through this research, we aim to advance the state-of-the-art for efficient ECR and shed light on the potential shortcomings of current LLMs at this task. Code and annotations: \url{https://github.com/ahmeshaf/gpt_coref}, Comment: LREC-COLING 2024 main conference

We study a multispecies $t$-PushTASEP system on a finite ring of $n$ sites with site-dependent rates $x_1,\dots,x_n$. Let $\lambda=(\lambda_1,\dots,\lambda_n)$ be a partition whose parts represent the species of the $n$ particles on the ring. We show that for each composition $\eta$ obtained by permuting the parts of $\lambda$, the stationary probability of being in state $\eta$ is proportional to the ASEP polynomial $F_{\eta}(x_1,\dots,x_n; q,t)$ at $q=1$; the normalizing constant (or partition function) is the Macdonald polynomial $P_{\lambda}(x_1,\dots,x_n;q,t)$ at $q=1$. Our approach involves new relations between the families of ASEP polynomials and of non-symmetric Macdonald polynomials at $q=1$. We also use multiline diagrams, showing that a single jump of the PushTASEP system is closely related to the operation of moving from one line to the next in a multiline diagram. We derive symmetry properties for the system under permutation of its jump rates, as well as a formula for the current of a single-species system., Comment: 29 pages, 4 figures, comments welcome

This paper presents a novel Cross-document Abstract Meaning Representation (X-AMR) annotation tool designed for annotating key corpus-level event semantics. Leveraging machine assistance through the Prodigy Annotation Tool, we enhance the user experience, ensuring ease and efficiency in the annotation process. Through empirical analyses, we demonstrate the effectiveness of our tool in augmenting an existing event corpus, highlighting its advantages when integrated with GPT-4. Code and annotations: https://github.com/ahmeshaf/gpt_coref, Comment: EACL 2024 System Demonstration