Information extraction (IE) needs vary over time, where a flexible information extraction (IE) system can be useful. Despite this, existing IE systems are either fully supervised, requiring expensive human annotations, or fully unsupervised, extracting information that often do not cater to user’s needs. To address these issues, we formally introduce the task of “IE on-the-fly”, and address the problem using our proposed Adaptive IE framework that uses human-in-the-loop refinement to adapt to changing user questions. Through human experiments on three diverse datasets, we demonstrate that Adaptive IE is a domain-agnostic, responsive, efficient framework for helping users access useful information while quickly reorganizing information in response to evolving information needs.

Seyedali Mohammadi, Bhaskara Hanuma Vedula, Hemank Lamba, Edward Raff, Ponnurangam Kumaraguru, Francis Ferraro, Manas Gaur. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

A strategy used by malicious actors is to "live off the land," where benign systems and tools already available on a victim’s systems are used and repurposed for the malicious actor’s intent. In this work, we ask if there is a way for antivirus developers to similarly re-purpose existing work to improve their malware detection capability. We show that this is plausible via YARA rules, which use human-written signatures to detect specific malware families, functionalities, or other markers of interest. By extracting sub-signatures from publicly available YARA rules, we assembled a set of features that can more effectively discriminate malicious samples from benign ones. Our experiments demonstrate that these features add value beyond traditional features on the EMBER 2018 dataset. Manual analysis of the added sub-signatures shows a power-law behavior in a combination of features that are specific and unique, as well as features that occur often. A prior expectation may be that the features would be limited in being overly specific to unique malware families. This behavior is observed, and is apparently useful in practice. In addition, we also find sub-signatures that are dual-purpose (e.g., detecting virtual machine environments) or broadly generic (e.g., DLL imports).

To support positive, ethical human-robot interactions, robots need to be able to respond to unexpected situations in which societal norms are violated, including rejecting unethical commands. Implementing robust communication for robots is inherently difficult due to the variability of context in real-world settings and the risks of unintended influence during robots’ communication. HRI researchers have begun exploring the potential use of LLMs as a solution for language-based communication, which will require an in-depth understanding and evaluation of LLM applications in different contexts. In this work, we explore how an existing LLM responds to and reasons about a set of norm-violating requests in HRI contexts. We ask human participants to assess the performance of a hypothetical GPT-4-based robot on moral reasoning and explanatory language selection as it compares to human intuitions. Our findings suggest that while GPT-4 performs well at identifying norm violation requests and suggesting non-compliant responses, its flaws in not matching the linguistic preferences and context sensitivity of humans prevent it from being a comprehensive solution for moral communication between humans and robots. Based on our results, we provide a four-point recommendation for the community in incorporating LLMs into HRI systems.

As the size of datasets used in statistical learning continues to grow, distributed training of models has attracted increasing attention. These methods partition the data and exploit parallelism to reduce memory and runtime, but suffer increasingly from communication costs as the data size or the number of iterations grows. Recent work on linear models has shown that a surrogate likelihood can be optimized locally to iteratively improve on an initial solution in a communication-efficient manner. However, existing versions of these methods experience multiple shortcomings as the data size becomes massive, including diverging updates and efficiently handling sparsity. In this work we develop solutions to these problems which enable us to learn a communication-efficient distributed logistic regression model even beyond millions of features. In our experiments we demonstrate a large improvement in accuracy over distributed algorithms with only a few distributed update steps needed, and similar or faster runtimes. Our code is available at https://github.com/FutureComputing4AI/ProxCSL.

The existence of external ("side") semantic knowledge has been shown to result in more expressive computational event models.To enable the use of side information that may be noisy or missing, we propose a semi-supervised information bottleneck-based discrete latent variable model.We reparameterize the model's discrete variables with auxiliary continuous latent variables and a light-weight hierarchical structure.Our model is learned to minimize the mutual information between the observed data and optional side knowledge that is not already captured by the new, auxiliary variables.We theoretically show that our approach generalizes past approaches, and perform an empirical case study of our approach on event modeling.We corroborate our theoretical results with strong empirical experiments, showing that the proposed method outperforms previous proposed approaches on multiple datasets.

Abstract The events in a narrative are understood as a coherent whole via the underlying states of their participants. Often, these participant states are not explicitly mentioned, instead left to be inferred by the reader. A model that understands narratives should likewise infer these implicit states, and even reason about the impact of changes to these states on the narrative. To facilitate this goal, we introduce a new crowdsourced English-language, Participant States dataset, PASTA. This dataset contains inferable participant states; a counterfactual perturbation to each state; and the changes to the story that would be necessary if the counterfactual were true. We introduce three state-based reasoning tasks that test for the ability to infer when a state is entailed by a story, to revise a story conditioned on a counterfactual state, and to explain the most likely state change given a revised story. Experiments show that today’s LLMs can reason about states to some degree, but there is large room for improvement, especially in problems requiring access and ability to reason with diverse types of knowledge (e.g., physical, numerical, factual).1

Learning to understand grounded language, which connects natural language to percepts, is a critical research area. Prior work in grounded language acquisition has focused primarily on textual inputs. In this work, we demonstrate the feasibility of performing grounded language acquisition on paired visual percepts and raw speech inputs. This will allow human-robot interactions in which language about novel tasks and environments is learned from end-users, reducing dependence on textual inputs and potentially mitigating the effects of demographic bias found in widely available speech recognition systems. We leverage recent work in self-supervised speech representation models and show that learned representations of speech can make language grounding systems more inclusive towards specific groups while maintaining or even increasing general performance.

Knowledge about outcomes is critical for complex event understanding but is hard to acquire.We show that by pre-identifying a participant in a complex event, crowdworkers are able to (1) infer the collective impact of salient events that make up the situation, (2) annotate the volitional engagement of participants in causing the situation, and (3) ground the outcome of the situation in state changes of the participants.By creating a multi-step interface and a careful quality control strategy, we collect a high quality annotated dataset of 8K short newswire narratives and ROCStories with high inter-annotator agreement (0.74-0.96 weighted Fleiss Kappa).Our dataset, POQue (Participant Outcome Questions), enables the exploration and development of models that address multiple aspects of semantic understanding.Experimentally, we show that current language models lag behind human performance in subtle ways through our task formulations that target abstract and specific comprehension of a complex event, its outcome, and a participant's influence over the event culmination.

We present a framework to statistically audit the privacy guarantee conferred by a differentially private machine learner in practice. While previous works have taken steps toward evaluating privacy loss through poisoning attacks or membership inference, they have been tailored to specific models or have demonstrated low statistical power. Our work develops a general methodology to empirically evaluate the privacy of differentially private machine learning implementations, combining improved privacy search and verification methods with a toolkit of influence-based poisoning attacks. We demonstrate significantly improved auditing power over previous approaches on a variety of models including logistic regression, Naive Bayes, and random forest. Our method can be used to detect privacy violations due to implementation errors or misuse. When violations are not present, it can aid in understanding the amount of information that can be leaked from a given dataset, algorithm, and privacy specification.

Ordinal regression is a classification task where classes have an order and prediction error increases the further the predicted class is from the true class. The standard approach for modeling ordinal data involves fitting parallel separating hyperplanes that optimize a certain loss function. This assumption offers sample efficient learning via inductive bias, but is often too restrictive in real-world datasets where features may have varying effects across different categories. Allowing class-specific hyperplane slopes creates generalized logistic ordinal regression, increasing the flexibility of the model at a cost to sample efficiency. We explore an extension of the generalized model to the all-thresholds logistic loss and propose a regularization approach that interpolates between these two extremes. Our method, which we term continuously generalized ordinal logistic, significantly outperforms the standard ordinal logistic model over a thorough set of ordinal regression benchmark datasets. We further extend this method to deep learning and show that it achieves competitive or lower prediction error compared to previous models over a range of datasets and modalities. Furthermore, two primary alternative models for deep learning ordinal regression are shown to be special cases of our framework.

Within the context of event modeling and understanding, we propose a new method for neural sequence modeling that takes partially-observed sequences of discrete, external knowledge into account.We construct a sequential neural variational autoencoder, which uses Gumbel-Softmax reparametrization within a carefully defined encoder, to allow for successful backpropagation during training.The core idea is to allow semisupervised external discrete knowledge to guide, but not restrict, the variational latent parameters during training.Our experiments indicate that our approach not only outperforms multiple baselines and the state-of-the-art in narrative script induction, but also converges more quickly.

We propose a learning system in which language is grounded in visual percepts without specific pre-defined categories of terms. We present a unified generative method to acquire a shared semantic/visual embedding that enables the learning of language about a wide range of real-world objects. We evaluate the efficacy of this learning by predicting the semantics of objects and comparing the performance with neural and non-neural inputs. We show that this generative approach exhibits promising results in language grounding without pre-specifying visual categories under low resource settings. Our experiments demonstrate that this approach is generalizable to multilingual, highly varied datasets.

We present CASIE, a system that extracts information about cybersecurity events from text and populates a semantic model, with the ultimate goal of integration into a knowledge graph of cybersecurity data. It was trained on a new corpus of 1,000 English news articles from 2017–2019 that are labeled with rich, event-based annotations and that covers both cyberattack and vulnerability-related events. Our model defines five event subtypes along with their semantic roles and 20 event-relevant argument types (e.g., file, device, software, money). CASIE uses different deep neural networks approaches with attention and can incorporate rich linguistic features and word embeddings. We have conducted experiments on each component in the event detection pipeline and the results show that each subsystem performs well.

Ordering the selection of training data using active learning can lead to improvements in learning efficiently from smaller corpora. We present an exploration of active learning approaches applied to three grounded language problems of varying complexity in order to analyze what methods are suitable for improving data efficiency in learning. We present a method for analyzing the complexity of data in this joint problem space, and report on how characteristics of the underlying task, along with design decisions such as feature selection and classification model, drive the results. We observe that representativeness, along with diversity, is crucial in selecting data samples.

We present the Universal Decompositional Semantics (UDS) dataset (v1.0), which is bundled with the Decomp toolkit (v0.1). UDS1.0 unifies five high-quality, decompositional semantics-aligned annotation sets within a single semantic graph specification—with graph structures defined by the predicative patterns produced by the PredPatt tool and real-valued node and edge attributes constructed using sophisticated normalization procedures. The Decomp toolkit provides a suite of Python 3 tools for querying UDS graphs using SPARQL. Both UDS1.0 and Decomp0.1 are publicly available at http://decomp.io.

We show how to learn a neural topic model with discrete random variables---one that explicitly models each word's assigned topic---using neural variational inference that does not rely on stochastic backpropagation to handle the discrete variables. The model we utilize combines the expressive power of neural methods for representing sequences of text with the topic model's ability to capture global, thematic coherence. Using neural variational inference, we show improved perplexity and document understanding across multiple corpora. We examine the effect of prior parameters both on the model and variational parameters, and demonstrate how our approach can compete and surpass a popular topic model implementation on an automatic measure of topic quality.

Articles about cybersecurity events like data breaches and ransomware attacks are common, both in general news and technical sources. Automatically extracting structured information from these can provide valuable information to inform both human analysts and computer systems. In this paper we describe how cybersecurity events can be described via semantic schemas, examined through an initial set of five event types. Using a collection of 1,000 news articles annotated with these event types, including their semantic roles, arguments, realis, and coreference, we detail a modular, deep-learning based information extraction (IE) pipeline, which extracts useful event information with high accuracy. We argue that the event argument set considered here can support many other cybersecurity events, facilitating the extension to new cybersecurity event types, such as distributed denial of service and SQL injection attacks.

Learning the meaning of grounded language - language that references a robot's physical environment and perceptual data - is an important and increasingly widely studied problem in robotics and human-robot interaction. However, with a few exceptions, research in robotics has focused on learning groundings for a single natural language pertaining to rich perceptual data. We present experiments on taking an existing natural language grounding system designed for English and applying it to a novel multilingual corpus of descriptions of objects paired with RGB-D perceptual data. We demonstrate that this specific approach transfers well to different languages, but also present possible design constraints to consider for grounded language learning systems intended for robots that will function in a variety of linguistic settings.

We present the first probabilistic model to capture all levels of the Minsky Frame structure, with the goal of corpus-based induction of scenario definitions. Our model unifies prior efforts in discourse-level modeling with that of Fillmore's related notion of frame, as captured in sentence-level, FrameNet semantic parses; as part of this, we resurrect the coupling among Minsky's frames, Schank's scripts and Fillmore's frames, as originally laid out by those authors. Empirically, our approach yields improved scenario representations, reflected quantitatively in lower surprisal and more coherent latent scenarios.

Ting-Hao Kenneth Huang, Francis Ferraro, Nasrin Mostafazadeh, Ishan Misra, Aishwarya Agrawal, Jacob Devlin, Ross Girshick, Xiaodong He, Pushmeet Kohli, Dhruv Batra, C. Lawrence Zitnick, Devi Parikh, Lucy Vanderwende, Michel Galley, Margaret Mitchell. Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2016.

Francis Ferraro, Nasrin Mostafazadeh, Ting-Hao Huang, Lucy Vanderwende, Jacob Devlin, Michel Galley, Margaret Mitchell. Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. 2015.

We compare the multinomial i-vector framework from the speech community with LDA, SAGE, and LSA as feature learners for topic ID on multinomial speech and text data.We also compare the learned representations in their ability to discover topics, quantified by distributional similarity to gold-standard topics and by human interpretability.We find that topic ID and topic discovery are competing objectives.We argue that LSA and i-vectors should be more widely considered by the text processing community as pre-processing steps for downstream tasks, and also speculate about speech processing tasks that could benefit from more interpretable representations like SAGE.

The narrative cloze is an evaluation metric commonly used for work on automatic script induction.While prior work in this area has focused on count-based methods from distributional semantics, such as pointwise mutual information, we argue that the narrative cloze can be productively reframed as a language modeling task.By training a discriminative language model for this task, we attain improvements of up to 27 percent over prior methods on standard narrative cloze metrics.

Nanyun Peng, Francis Ferraro, Mo Yu, Nicholas Andrews, Jay DeYoung, Max Thomas, Matthew R. Gormley, Travis Wolfe, Craig Harman, Benjamin Van Durme, Mark Dredze. Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. 2015.

We present the first large-scale, corpus based verification of Dowty’s seminal theory of proto-roles. Our results demonstrate both the need for and the feasibility of a property-based annotation scheme of semantic relationships, as opposed to the currently dominant notion of categorical roles.