Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Claire CardieSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works368 from public data
- HellaSwag: Can a Machine Really Finish Your Sentence?5,095
The construction of HellaSwag, a new challenge dataset, and its resulting difficulty, sheds light on the inner workings of deep pretrained models, and suggests a new path forward for NLP research, in which benchmarks co-evolve with the evolving state-of-the-art in an adversarial way, so as to present ever-harder challenges.
- The Curious Case of Neural Text Degeneration4,651
By sampling text from the dynamic nucleus of the probability distribution, which allows for diversity while effectively truncating the less reliable tail of the distribution, the resulting text better demonstrates the quality of human text, yielding enhanced diversity without sacrificing fluency and coherence.
- PIQA: Reasoning about Physical Commonsense in Natural Language3,673
The task of physical commonsense reasoning and a corresponding benchmark dataset Physical Interaction: Question Answering or PIQA are introduced and analysis about the dimensions of knowledge that existing models lack are provided, which offers significant opportunities for future research.
- CLIPScore: A Reference-free Evaluation Metric for Image Captioning3,343
The surprising empirical finding that CLIP (Radford et al., 2021), a cross-modal model pretrained on 400M image+caption pairs from the web, can be used for robust automatic evaluation of image captioning without the need for references is reported.
- Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks2,301
This paper proposes a new learning method Oscar (Object-Semantics Aligned Pre-training), which uses object tags detected in images as anchor points to significantly ease the learning of alignments.
- RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models1,890
It is found that pretrained LMs can degenerate into toxic text even from seemingly innocuous prompts, and empirically assess several controllable generation methods find that while data- or compute-intensive methods are more effective at steering away from toxicity than simpler solutions, no current method is failsafe against neural toxic degeneration.
- Neural Motifs: Scene Graph Parsing with Global Context1,207
This work analyzes the role of motifs: regularly appearing substructures in scene graphs and introduces Stacked Motif Networks, a new architecture designed to capture higher order motifs in scene graph graphs that improves on the previous state-of-the-art by an average of 3.6% relative improvement across evaluation settings.
- 1,137
- From Recognition to Cognition: Visual Commonsense Reasoning1,083
To move towards cognition-level understanding, a new reasoning engine is presented, Recognition to Cognition Networks (R2C), that models the necessary layered inferences for grounding, contextualization, and reasoning.
- The Risk of Racial Bias in Hate Speech Detection1,082
This work proposes *dialect* and *race priming* as ways to reduce the racial bias in annotation, showing that when annotators are made explicitly aware of an AAE tweet’s dialect they are significantly less likely to label the tweet as offensive.
- 1,077
- COMET: Commonsense Transformers for Automatic Knowledge Graph Construction1,061
This investigation reveals promising results when implicit knowledge from deep pre-trained language models is transferred to generate explicit knowledge in commonsense knowledge graphs, and suggests that using generative commonsense models for automatic commonsense KB completion could soon be a plausible alternative to extractive methods.
- Truth of Varying Shades: Analyzing Language in Fake News and Political Fact-Checking957
Experiments show that while media fact-checking remains to be an open research question, stylistic cues can help determine the truthfulness of text.
- MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms937
A large-scale dataset of math word problems and an interpretable neural math problem solver by learning to map problems to their operation programs and a new representation language to model operation programs corresponding to each math problem that aim to improve both the performance and the interpretability of the learned models.
- QuAC: Question Answering in Context931
QuAC introduces challenges not found in existing machine comprehension datasets: its questions are often more open-ended, unanswerable, or only meaningful within the dialog context, as it shows in a detailed qualitative evaluation.
- SWAG: A Large-Scale Adversarial Dataset for Grounded Commonsense Inference804
This paper introduces the task of grounded commonsense inference, unifying natural language inference and commonsense reasoning, and proposes Adversarial Filtering (AF), a novel procedure that constructs a de-biased dataset by iteratively training an ensemble of stylistic classifiers, and using them to filter the data.
- 740
- 679
- Social Bias Frames: Reasoning about Social and Power Implications of Language672
It is found that while state-of-the-art neural models are effective at high-level categorization of whether a given statement projects unwanted social bias, they are not effective at spelling out more detailed explanations in terms of Social Bias Frames.
- Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics648
The results indicate that a shift in focus from quantity to quality of data could lead to robust models and improved out-of-distribution generalization, and a model-based tool to characterize and diagnose datasets.
- OpinionFinder: a system for subjectivity analysis605
OpinionFinder is a system that performs subjectivity analysis, automatically identifying when opinions, sentiments, speculations, and other private states are present in text.
- DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts576
This work highlights the promise of tuning small LMs on text with (un)desirable attributes for efficient decoding-time steering and applies DExperts to language detoxification and sentiment-controlled generation, where it outperform existing controllable generation methods on both automatic and human evaluations.
- Syntactic Stylometry for Deception Detection568
This paper investigates syntactic stylometry for deception detection, adding a somewhat unconventional angle to prior literature and demonstrating that features driven from Context Free Grammar (CFG) parse trees consistently improve the detection performance over several baselines that are based only on shallow lexico-syntactic features.
- 544
- Cosmos QA: Machine Reading Comprehension with Contextual Commonsense Reasoning529
This paper introduces Cosmos QA, a large-scale dataset of 35,600 problems that require commonsense-based reading comprehension, formulated as multiple-choice questions, and proposes a new architecture that improves over the competitive baselines.
- (Comet-) Atomic 2020: On Symbolic and Neural Commonsense Knowledge Graphs477
It is proposed that manually constructed CSKGs will never achieve the coverage necessary to be applicable in all situations encountered by NLP agents, and a new evaluation framework for testing the utility of KGs based on how effectively implicit knowledge representations can be learned from them is proposed.
- 476
- 472
- Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detection443
This work disentangle what is annotated as toxic by considering posts with three characteristics: anti-Black language, African American English dialect, and vulgarity, and shows strong associations between annotator identity and beliefs and their ratings of toxicity.
- Generated Knowledge Prompting for Commonsense Reasoning427
Generated knowledge prompting develops generated knowledge prompting, which consists of generating knowledge from a language model, then providing the knowledge as additional input when answering a question, and improves performance of large-scale, state-of-the-art models on four commonsense reasoning tasks.
- CommonGen: A Constrained Text Generation Challenge for Generative Commonsense Reasoning426
A constrained text generation task, CommonGen associated with a benchmark dataset, to explicitly test machines for the ability of generative commonsense reasoning, and demonstrates that the learned generative Commonsense reasoning capability can be transferred to improve downstream tasks such as CommonsenseQA by generating additional context.
- Identifying sources of opinions with conditional random fields and extraction patterns420
This work adopts a hybrid approach that combines Conditional Random Fields (Lafferty et al., 2001) and a variation of AutoSlog (Riloff, 1996a), and shows that the combination of these two methods performs better than either one alone.
- Composing Simple Image Descriptions using Web-scale N-grams418
A simple yet effective approach to automatically compose image descriptions given computer vision based inputs and using web-scale n-grams, which indicates that it is viable to generate simple textual descriptions that are pertinent to the specific content of an image, while permitting creativity in the description -- making for more human-like annotations than previous approaches.
- Symbolic Knowledge Distillation: from General Language Models to Commonsense Models403
It is demonstrated that careful prompt engineering and a separately trained critic model allow us to selectively distill high-quality causal commonsense from GPT-3, a general language model, and results in a neural commonsense model that surpasses the teacher model's commonsense capabilities despite its 100x smaller size.
- Collective Generation of Natural Image Descriptions394
A holistic data-driven approach to image description generation, exploiting the vast amount of (noisy) parallel image data and associated natural language descriptions available on the web to generate novel descriptions for query images.
- Social Chemistry 101: Learning to Reason about Social and Moral Norms383
A new conceptual formalism to study people's everyday social norms and moral judgments over a rich spectrum of real life situations described in natural language and a model framework, Neural Norm Transformer, learns and generalizes Social-Chem-101 to successfully reason about previously unseen situations, generating relevant (and potentially novel) attribute-aware social rules-of-thumb.
- Learning with compositional semantics as structural inference for subsentential sentiment analysis365
A novel learning-based approach that incorporates structural inference motivated by compositional semantics into the learning procedure is presented, and it is found that expression-level classification accuracy uniformly decreases as additional, potentially disambiguating, context is considered.
- WinoGrande360
WinoGrande is introduced, a large-scale dataset of 44k problems, inspired by the original WSC, but adjusted to improve both the scale and the hardness of the dataset, and demonstrates the effectiveness of WINOGRANDE when used as a resource for transfer learning.
- Surface Form Competition: Why the Highest Probability Answer Isn’t Always Right346
Domain Conditional Pointwise Mutual Information is introduced, an alternative scoring function that directly compensates for surface form competition by simply reweighing each option according to its a priori likelihood within the context of a specific task.
- DREAM: A Challenge Data Set and Models for Dialogue-Based Reading Comprehension323
Experimental results on the DREAM data set show the effectiveness of dialogue structure and general world knowledge, the first dialogue-based multiple-choice reading comprehension data set to focus on in-depth multi-turn multi-party dialogue understanding.
- Deep Communicating Agents for Abstractive Summarization323
Empirical results demonstrate that multiple communicating encoders lead to a higher quality summary compared to several strong baselines, including those based on a single encoder or multiple non-communicating encoder.
- Neural AMR: Sequence-to-Sequence Models for Parsing and Generation319
This work presents a novel training procedure that can lift the limitation of the relatively limited amount of labeled data and the non-sequential nature of the AMR graphs, and presents strong evidence that sequence-based AMR models are robust against ordering variations of graph-to-sequence conversions.
- Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization309
It is shown that RL techniques are generally better than supervised methods at aligning LMs to human preferences; and that NLPO exhibits greater stability and performance than previous policy gradient methods (e.g., PPO) based on both automatic and human evaluations.
- Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory303
This work proposes ConfAIde, a benchmark designed to identify critical weaknesses in the privacy reasoning capabilities of instruction-tuned LLMs, and underscores the immediate need to explore novel inference-time privacy-preserving approaches, based on reasoning and theory of mind.
- Unsupervised Commonsense Question Answering with Self-Talk289
An unsupervised framework based on self-talk as a novel alternative to multiple-choice commonsense tasks, inspired by inquiry-based discovery learning, which improves performance on several benchmarks and competes with models that obtain knowledge from external KBs.
- T ree T alk : Composition and Compression of Trees for Image Descriptions289
This work is the first to show the empirical benefit of automatically generalized captions for composing natural image descriptions, and significantly better performance than previous approaches for both image caption generalization and generation.
- Agent AI: Surveying the Horizons of Multimodal Interaction288
It is argued that by developing agentic AI systems in grounded environments, one can also mitigate the hallucinations of large foundation models and their tendency to generate environmentally incorrect outputs.
- Reframing Instructional Prompts to GPTk’s Language286
This work studies several classes of reframing techniques for manual reformulation of prompts into more effective ones, and hopes these empirically-driven techniques will pave the way towards more effective future prompting algorithms.
- Identifying expressions of opinion in context284
An approach for identifying opinion expressions that uses conditional random fields and achieves expression-level performance within 5% of the human interannotator agreement is presented.
- WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creation274
This work introduces a novel approach for dataset creation based on worker and AI collaboration, which brings together the generative strength of language models and the evaluation strength of humans, and demonstrates the promise of leveraging natural language generation techniques and re-imagining the role of humans in the dataset creation process.
- Learning to Write with Cooperative Discriminators257
Human evaluation demonstrates that text generated by the unified learning framework is preferred over that of baselines by a large margin, significantly enhancing the overall coherence, style, and information of the generations.
- Distributional Footprints of Deceptive Product Reviews233
A range of experiments confirm the hypothesized connection between the distributional anomaly and deceptive reviews and provide novel quantitative insights into the characteristics of natural distributions of opinions in the TripAdvisor hotel review and the Amazon product review domains.
- Ultra-Fine Entity Typing232
A model that can predict ultra-fine types is presented, and is trained using a multitask objective that pools the authors' new head-word supervision with prior supervision from entity linking, and achieves state of the art performance on an existing fine-grained entity typing benchmark, and sets baselines for newly-introduced datasets.
- Joint extraction of entities and relations for opinion recognition232
An integer linear programming approach is employed to solve the joint opinion recognition task, and it is shown that global, constraint-based inference can significantly boost the performance of both relation extraction and the extraction of opinion-related entities.
- Globally Coherent Text Generation with Neural Checklist Models225
The neural checklist model is presented, a recurrent neural network that models global coherence by storing and updating an agenda of text strings which should be mentioned somewhere in the output by dynamically adjusting the interpolation among a language model and a pair of attention models.
- NeuroLogic A*esque Decoding: Constrained Text Generation with Lookahead Heuristics213
NeuroLogic A*esque is proposed, a decoding algorithm that incorporates heuristic estimates of future cost that develops lookahead heuristics that are efficient for large-scale language models, making this method a drop-in replacement for common techniques such as beam search and top-k sampling.
- Event2Mind: Commonsense Inference on Events, Intents, and Reactions212
It is demonstrated how commonsense inference on people’s intents and reactions can help unveil the implicit gender inequality prevalent in modern movie scripts.
- Adapting a polarity lexicon using integer linear programming for domain-specific sentiment classification207
A novel method based on integer linear programming that can adapt an existing lexicon into a new one to reflect the characteristics of the data more directly, and Experimental results show that the lexicon adaptation technique improves the performance of fine-grained polarity classification.
- Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequences204
Moral Stories, a crowd-sourced dataset of structured, branching narratives for the study of grounded, goal-oriented social reasoning, is introduced and decoding strategies that combine multiple expert models to significantly improve the quality of generated actions, consequences, and norms compared to strong baselines are proposed.
- NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraints197
This work proposes NeuroLogic Decoding, a simple yet effective algorithm that enables neural language models – supervised or not – to generate fluent text while satisfying complex lexical constraints, and suggests the limit of large-scale neural networks for fine-grained controllable generation and the promise of inference-time algorithms.
- PlotMachines: Outline-Conditioned Generation with Dynamic Plot State Tracking190
PlotMachines is presented, a neural narrative model that learns to transform an outline into a coherent story by tracking the dynamic plot states, and is enriched with high-level discourse structure so that the model can learn different styles of writing corresponding to different parts of the narrative.
- Tactical Rewind: Self-Correction via Backtracking in Vision-And-Language Navigation188
The Frontier Aware Search with backTracking (FAST) Navigator is presented, a general framework for action decoding, that achieves state-of-the-art results on the 2018 Room-to-Room (R2R) Vision-and-Language navigation challenge.
- Reframing Human-AI Collaboration for Generating Free-Text Explanations180
This work creates a pipeline that combines GPT-3 with a supervised filter that incorporates binary acceptability judgments from humans in the loop and demonstrates that acceptability is partially correlated with various fine-grained attributes of explanations.
- Counterfactual Story Reasoning and Generation176
This paper proposes Counterfactual Story Rewriting: given an original story and an intervening counterfactual event, the task is to minimally revise the story to make it compatible with the given counterfactually event.
- SCRUPLES: A Corpus of Community Ethical Judgments on 32,000 Real-Life Anecdotes166
This work introduces SCRUPLES, the first large-scale dataset with 625,000 ethical judgments over 32,000 real-life anecdotes, and presents a new method to estimate the best possible performance on such tasks with inherently diverse label distributions, and explores likelihood functions that separate intrinsic from model uncertainty.
- Generating Topical Poetry164
Hafez is described, a program that generates any number of distinct poems on a user-supplied topic that obey rhythmic and rhyme constraints and shows its generality with respect to language and poetic form.
- Is GPT-3 Text Indistinguishable from Human Text? Scarecrow: A Framework for Scrutinizing Machine Text163
This work proposes a new framework called Scarecrow for scrutinizing machine text via crowd annotation, and quantifies measurable gaps between human authored text and generations from models of several sizes, including fourteen configurations of GPT-3.
- Challenges in Automated Debiasing for Toxic Language Detection163
The findings show that debiasing a model trained on biased toxic language data is not as effective as simply relabeling the data to remove existing biases, and proposes an automatic, dialect-aware data correction method, as a proof-of-concept.
- Commonsense Knowledge Base Completion with Structural and Semantic Context161
This paper investigates two key ideas: (1) learning from local graph structure, using graph convolutional networks and automatic graph densification and (2) transfer learning from pre-trained language models to knowledge graphs for enhanced contextual representation of knowledge.
- VisualCOMET: Reasoning About the Dynamic Context of a Still Image156
This work proposes VisualComet, the novel framework of visual commonsense reasoning tasks to predict events thatmight have happened before, events that might happen next, and the intents of the people at present, and introduces the first large-scale repository of Visual Commonsense Graphs, allowing for tighter integration between images and text.
- UNICORN on RAINBOW: A Universal Commonsense Reasoning Model on a New Multitask Benchmark153
Two new ways to evaluate commonsense models are proposed, emphasizing their generality on new tasks and building on diverse, recently introduced benchmarks, and a novel evaluation, the cost equivalent curve, is proposed, that sheds new insight on how the choice of source datasets, pretrained language models, and transfer learning methods impacts performance and data efficiency.
- 148
- Neural Metaphor Detection in Context147
These end-to-end neural models establish a new state-of-the-art on existing verb metaphor detection benchmarks, and show strong performance on jointly predicting the metaphoricity of all words in a running text.
- Dynamic Neuro-Symbolic Knowledge Graph Construction for Zero-shot Commonsense Question Answering140
This paper presents a novel approach that generates contextually-relevant symbolic knowledge structures on demand using generative neural commonsense knowledge models and achieves significant performance boosts over pretrained language models and vanilla knowledge models, all while providing interpretable reasoning paths for its predictions.
- Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Duties139
Value Kaleidoscope, an open, light-weight, and structured language-based multi-task model that generates, explains, and assesses the relevance and valence of human values, rights, and duties within a specific context, and shows that Kaleido’s representations transfer to other philosophical frameworks and datasets, confirming the benefit of an explicit, modular, and interpretable approach to value pluralism.
- The Generative AI Paradox: "What It Can Create, It May Not Understand"136
The hypothesis that models' generative capability may not be contingent upon understanding capability, and call for caution in interpreting artificial intelligence by analogy to human intelligence, is supported.
- Robust Navigation with Language Pretraining and Stochastic Sampling133
This paper adapts large-scale pretrained language models to learn text representations that generalize better to previously unseen instructions and proposes a stochastic sampling scheme to reduce the considerable gap between the expert actions in training and sampled actions in test.
- Thinking Like a Skeptic: Defeasible Inference in Natural Language131
From Defeasible NLI, both a classification and generation task for defeasible inference are developed, and it is demonstrated that the generation task is much more challenging.
- Contrastive Explanations for Model Interpretability128
The ability of label-contrastive explanations to provide fine-grained interpretability of model decisions is demonstrated, via both high-level abstract concept attribution and low-level input token/span attribution for two NLP classification benchmarks.
- A Data Augmentation Approach for Sign-Language-To-Text Translation In-The-Wild128
The goal is to build a state-of-the-art system for the automatic interpretation of sign language in unpredictable video framing conditions and the set of hypotheses to build such a system are described and the list of experiments that will lead to their verification are described.
- Connotation Frames of Power and Agency in Modern Films127
This work uses connotation frames of power and agency frames to measure the subtle, but prevalent, gender bias in the portrayal of modern film characters and provide insights that deviate from the well-known Bechdel test.
- Gender Attribution: Tracing Stylometric Evidence Beyond Topic and Genre125
This work is the first that consciously avoids gender bias in topics, thereby providing stronger evidence to gender-specific styles in language beyond topic, and the comparative study provides new insights into robustness of various stylometric techniques across topic and genre.
- Connotation Lexicon: A Dash of Sentiment Beneath the Surface Meaning124
This paper proposes induction algorithms encoding a diverse set of linguistic insights (semantic prosody, distributional similarity, semantic parallelism of coordination) and prior knowledge drawn from lexical resources, resulting in the first broad-coverage connotation lexicon.
- Hierarchical Sequential Learning for Extracting Opinions and Their Attributes124
This paper applies a hierarchical parameter sharing technique using Conditional Random Fields for fine-grained opinion analysis, jointly detecting the boundaries of opinion expressions as well as determining two of their key attributes --- polarity and intensity.
- Mise en Place: Unsupervised Interpretation of Instructional Recipes123
An unsupervised hard EM approach to automatically mapping instructional recipes to action graphs, which define what actions should be performed on which objects and in what order, which incorporates aspects of procedural semantics and world knowledge.
- The Effect of Different Writing Tasks on Linguistic Style: A Case Study of the ROC Story Cloze Task121
It is shown how variants of the same writing task can lead to measurable differences in writing style, and a simple linear classifier informed by stylistic features is able to successfully distinguish among the three cases.
- Modeling Naive Psychology of Characters in Simple Commonsense Stories118
A new annotation framework is introduced to explain naive psychology of story characters as fully-specified chains of mental states with respect to motivations and emotional reactions and establishes baseline performance on several new tasks, suggesting avenues for future research.
- 117
- GO FIGURE: A Meta Evaluation of Factuality in Summarization115
GO FIGURE, a meta-evaluation framework for evaluating factuality evaluation metrics, is introduced and it is revealed that while QA metrics generally improve over standard metrics that measure factuality across domains, performance is highly dependent on the way in which questions are generated.
- Success with Style: Using Writing Style to Predict the Success of Novels114
This study reports for the first time that statistical stylometry can be surprisingly effective in discriminating highly successful literature from less successful counterpart, achieving accuracy up to 84%, and identifies characteristic stylistic elements that are more prominent in successful writings.
- 112
- Conversing by Reading: Contentful Neural Conversation with On-demand Machine Reading111
A new end-to-end approach to contentful neural conversation that jointly models response generation and on-demand machine reading is presented, allowing for more focused integration of external knowledge than has been possible in prior approaches.
- Generative Data Augmentation for Commonsense Reasoning107
This work proposes a novel generative data augmentation technique, G-DAUGˆC, that aims to achieve more accurate and robust learning in a low-resource setting and produces a diverse set of fluent training examples, demonstrating that its selection and training approaches are important for performance.
- TIMEDIAL: Temporal Commonsense Reasoning in Dialog99
This paper presents the first study to investigate pre-trained LMs for their temporal reasoning capabilities in dialogs by introducing a new task and a crowd-sourced English challenge set, TimeDial, and reveals that the models fail to reason about dialog context correctly; instead, they rely on shallow cues based on existing temporal patterns in context.
- Connotation Frames: A Data-Driven Investigation99
This work presents models for predicting the connotation frames of verb predicates based on their distributional word representations and the interplay between different types of connotative relations, and investigates the feasibility of obtaining con notative labels through crowdsourcing experiments.
- 96
- Dynamic Entity Representations in Neural Language Models94
A new type of language model is presented that can explicitly model entities, dynamically update their representations, and contextually generate their mentions and can model an arbitrary number of entities in context while generating each entity mention at an arbitrary length.
- Generalizing Image Captions for Image-Text Parallel Corpus89
This work introduces the new task of image caption generalization, formulated as visually-guided sentence compression, and presents an efficient algorithm based on dynamic beam search with dependency-based constraints and releases a new large-scale corpus with 1 million image-caption pairs achieving tighter content alignment between images and text.
- Verb Physics: Relative Physical Knowledge of Actions and Objects88
An approach to infer relative physical knowledge of actions and objects along five dimensions (e.g., size, weight, and strength) from unstructured natural language text is presented.
- Discourse-Aware Neural Rewards for Coherent Text Generation84
Empirical results demonstrate that a generator trained with the learned reward produces more coherent and less repetitive text than models trained with cross-entropy or with reinforcement learning with commonly used scores as rewards.
- PIGLeT: Language Grounding Through Neuro-Symbolic Interaction in a 3D World83
A model that learns physical commonsense knowledge through interaction, and then uses this knowledge to ground language, and is able to correctly forecast “what happens next” given an English sentence over 80% of the time, outperforming a 100x larger, text-to-text approach by over 10%.
- 83
- Large Scale Retrieval and Generation of Image Descriptions83
The end result is two simple, but effective, methods for harnessing the power of big data to produce image captions that are altogether more general, relevant, and human-like than previous attempts.
- 79
- Event Detection and Factuality Assessment with Non-Expert Supervision79
It is found that non-experts, with very little training, can reliably provide judgments about what events are mentioned and the extent to which the author thinks they actually happened.
- 76
- It’s not Rocket Science: Interpreting Figurative Language in Narratives73
This paper studies the interpretation of two non-compositional figurative languages (idioms and similes), and proposes knowledge-enhanced models, adopting human strategies for interpreting figurative language types: inferring meaning from the context and relying on the constituent words’ literal meanings.
- BottleSum: Unsupervised and Self-supervised Sentence Summarization using the Information Bottleneck Principle70
This paper proposes a novel approach to unsupervised sentence summarization by mapping the Information Bottleneck principle to a conditional language modelling objective: given a sentence, the approach seeks a compressed sentence that can best predict the next sentence.
- 69
- 69
- Natural Language Rationales with Full-Stack Visual Reasoning: From Pixels to Semantic Frames to Commonsense Graphs68
This study presents RationaleˆVT Transformer, an integrated model that learns to generate free-text rationales by combining pretrained language models with object recognition, grounded visual semantic frames, and visual commonsense graphs, and finds that integration of richer semantic and pragmatic visual features improves visual fidelity of rationales.
- Do Neural Language Models Overcome Reporting Bias?68
It is found that while pre-trained language models' generalization capacity allows them to better estimate the plausibility of frequent but unspoken of actions, outcomes, and properties, they also tend to overestimate that of the very rare, amplifying the bias that already exists in their training corpus.
- Learning Interpretable Spatial Operations in a Rich 3D Blocks World65
A new dataset that pairs complex 3D spatial operations to rich natural language descriptions that require complex spatial and pragmatic interpretations and a new neural architecture that achieves competitive results while automatically discovering an inventory of interpretable spatial operations are introduced.
- Automatically Generating Annotator Rationales to Improve Sentiment Classification65
This work explores methods to automatically generate annotator rationales for document-level sentiment classification and finds the automatically generated rationales just as helpful as human rationales.
- Keystroke Patterns as Prosody in Digital Writings: A Case Study with Deceptive Reviews and Essays64
Empirical results show that incorporating keystrokebased features lead to improved performance in deception detection in two different domains: online reviews and essays.
- 61
- Sounding Board: A User-Centric and Content-Driven Social Chatbot61
The system architecture consists of several components including spoken language processing, dialogue management, language generation, and content management, with emphasis on user-centric and content-driven design.
- 61
- 60
- Learning General Connotation of Words using Graph-based Algorithms60
A connotation lexicon is introduced, a new type of lexicon that lists words with connotative polarity, i.e., words with positive connotation and words with negative connotation that is of great value for sentiment analysis complementing existing sentiment lexicons.
- 59
- Zero-Shot Activity Recognition with Verb Attribute Induction56
Experimental results confirm that action attributes inferred from language can provide a predictive signal for zero-shot prediction of previously unseen activities.
- Are Elephants Bigger than Butterflies? Reasoning about Sizes of Objects54
This paper introduces a method to automatically infer object sizes, leveraging visual and textual information from web, and shows that the method outperforms competitive textual and visual baselines in reasoning about size comparisons.
- ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning53
ThinkMorph is built, a unified model fine-tuned on approximately 24K high-quality interleaved reasoning traces spanning tasks with varying visual engagement that learns to generate progressive text-image reasoning steps that concretely manipulate visual content while maintaining coherent verbal logic.
- 53
- Using landing pages for sponsored search ad selection53
This work compares the use of the ad's intrinsic content to augmenting the ad with the whole, or parts, of the landing page to improve the quality of ad selection, and explores two types of extractive summarization techniques to select useful regions from the landing pages.
- 51
- Detecting visual text49
This work concretely defines what it means to be visual, annotate visual text and develops algorithms to automatically classify noun phrases as visual or non-visual, and finds that using text alone, it is able to achieve high accuracies at this task, and that incorporating features derived from computer vision algorithms improves performance.
- Story Cloze Task: UW NLP System48
University of Washington NLP’s submission for the Linking Models of Lexical, Sentential and Discourse-level Semantics (LSDSem 2017) shared task—the Story Cloze Task is described, which is a linear classifier with a variety of features, including both the scores of a neural language model and style features.
- The Curious Case of Commonsense Intelligence47
The significance of language for modeling intuitive reasoning, the fundamental limitations of logic formalisms despite their intellectual appeal, the case for on-the-fly generative reasoning through language, the continuum between knowledge and reasoning, and the blend between symbolic and neural knowledge representations are discussed.
- 46
- 46
- Recollection versus Imagination: Exploring Human Memory and Cognition via Neural Language Models45
A measure of narrative flow is introduced and used to examine the narratives for imagined and recalled events and reveals the effect of narrativization of memories in stories (e.g., stories about frequently recalled memories flow more linearly; Bartlett, 1932).
- Language of Vandalism: Improving Wikipedia Vandalism Detection via Stylometric Analysis43
It is hypothesized that textual vandalism constitutes a unique genre where a group of people share a similar linguistic behavior and that deep syntactic patterns based on probabilistic context free grammars (PCFG) discriminate vandalism more effectively than shallow lexico-syntactic patternsbased on n-grams.
- 42
- Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering41
Empirical results on the SocialIQa and StoryCommonsense datasets in a zero-shot setting demonstrate that using commonsense knowledge models to dynamically construct and reason over knowledge graphs achieves performance boosts over pre-trained language models and usingknowledge models to directly evaluate answers.
- Neural Poetry Translation40
The first neural poetry translation system that always translates a source text to an English poem, and human evaluation of the translations ranks the quality as acceptable 78.2% of the time.
- 39
- Characterizing Stylistic Elements in Syntactic Structure39
A comprehensive exploration of syntactic elements in writing styles, with particular emphasis on interpretable characterization of stylistic elements, is presented.
- 38
- 37
- 36
- 35
- 34
- 30
- 30
- 29
- 29
- 29
- 29
- 28
- 28
- 28
- 27
- 27
- 27
- 26
- 26
- 25
- 25
- 25
- 25
- 24
- 24
- 24
- 23
- Unfolding Spatial Cognition: Evaluating Multimodal Models on Visual Simulations22
STARE is introduced, a benchmark designed to rigorously evaluate multimodal large language models on tasks better solved through multi-step visual simulation, and shows that models excel at reasoning over simpler 2D transformations, but perform close to random chance on more complex tasks.
- 22
- 22
- BotPercent: Estimating Bot Populations in Twitter Communities21
The results reveal that the presence of Twitter bots is not homogeneous, but exhibiting a spatial-temporal distribution with considerable heterogeneity that should be taken into account for content moderation and social media policy making.
- 20
- 20
- 20
- 19
- 19
- 19
- 19
- 18
- 18
- 18
- 17
- Certainly Uncertain: A Benchmark and Metric for Multimodal Epistemic and Aleatoric Awareness17
A taxonomy of uncertainty specific to vision-language AI systems is presented, distinguishing between epistemic uncertainty (arising from a lack of information) and aleatoric uncertainty (due to inherent unpredictability), and further explore finer categories within.
- 17
- 17
- 17
- 17
- 16
- 16
- 16
- 15
- 15
- 15
- 14
- 14
- 14
- 14
- 13
- 13
- 13
- 13
- 13
- 13
- 12
- 12
- 12
- 12
- 12
- Game Theoretical Approach of Blockchain-Based Spectrum Sharing for 5G-Enabled IoTs in Dense Networks12
- 12
- 11
- 11
- 11
- 10
- 10
- 10
- 10
- 10
- 10
- 10
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-13. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.