Explanipedia

LM Agents for Coordinating Multi-User Information Gathering Open

Harsh Jhamtani, Jacob Andreas, Benjamin Van Durme · 2025

This paper introduces PeopleJoin, a benchmark for evaluating LM-mediated collaborative problem solving. Given a user request, PeopleJoin agents must identify teammates who might be able to assist, converse with these teammates to gather in…

LLM Agents for Coordinating Multi-User Information Gathering Open

Harsh Jhamtani, Jacob Andreas, Benjamin Van Durme · 2025

Steering Large Language Models between Code Execution and Textual Reasoning Open

Yongchao Chen, Harsh Jhamtani, Srinagesh Sharma, Chuchu Fan, Chi Wang · 2024

While a lot of recent research focuses on enhancing the textual reasoning capabilities of Large Language Models (LLMs) by optimizing the multi-agent framework or reasoning chains, several benchmark tasks can be solved with 100\% success th…

Learning to Retrieve Iteratively for In-Context Learning Open

Yunmo Chen, Tongfei Chen, Harsh Jhamtani, Patrick Xia, Richard Shin , et al. · 2024

We introduce iterative retrieval, a novel framework that empowers retrievers to make iterative decisions through policy optimization. Finding an optimal portfolio of retrieved items is a combinatorial optimization problem, generally consid…

Interpreting User Requests in the Context of Natural Language Standing Instructions Open

Nikita Moghe, Patrick Xia, Jacob Andreas, Jason Eisner, Benjamin Van Durme , et al. · 2023

Users of natural language interfaces, generally powered by Large Language Models (LLMs),often must repeat their preferences each time they make a similar request. We describe an approach to LLM-based dialogue modeling in which persistent u…

SCREWS: A Modular Framework for Reasoning with Revisions Open

Kumar Shridhar, Harsh Jhamtani, Hao Fang, Benjamin Van Durme, Jason Eisner , et al. · 2023

Large language models (LLMs) can improve their accuracy on various tasks through iteratively refining and revising their output based on feedback. We observe that these revisions can introduce errors, in which case it is better to roll bac…

Natural Language Decomposition and Interpretation of Complex Utterances Open

Harsh Jhamtani, Hao Fang, Patrick Xia, Eran Levy, Jacob Andreas , et al. · 2023

Designing natural language interfaces has historically required collecting supervised data to translate user requests into carefully designed intent representations. This requires enumerating and labeling a long tail of user requests, whic…

The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding Open

Hao Fang, Anusha Balakrishnan, Harsh Jhamtani, John Bufe, Jean Crawford , et al. · 2023

Hao Fang, Anusha Balakrishnan, Harsh Jhamtani, John Bufe, Jean Crawford, Jayant Krishnamurthy, Adam Pauls, Jason Eisner, Jacob Andreas, Dan Klein. Findings of the Association for Computational Linguistics: ACL 2023. 2023.

Ontologically Faithful Generation of Non-Player Character Dialogues Open

Nathaniel Weir, Ryan M. Thomas, Randolph d’Amore, Kellie Hill, Benjamin Van Durme , et al. · 2022

We introduce a language generation task grounded in a popular video game environment. KNUDGE (KNowledge Constrained User-NPC Dialogue GEneration) requires models to produce trees of dialogue between video game characters that accurately re…

The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding Open

Hao Fang, Anusha Balakrishnan, Harsh Jhamtani, John Bufe, Jean Crawford , et al. · 2022

In a real-world dialogue system, generated text must be truthful and informative while remaining fluent and adhering to a prescribed style. Satisfying these constraints simultaneously is difficult for the two predominant paradigms in langu…

PINEAPPLE: Personifying INanimate Entities by Acquiring Parallel Personification data for Learning Enhanced generation Open

Sedrick Scott Keh, Kevin Lü, Varun Gangal, Steven Y. Feng, Harsh Jhamtani , et al. · 2022

A personification is a figure of speech that endows inanimate entities with properties and actions typically seen as requiring animacy. In this paper, we explore the task of personification generation. To this end, we propose PINEAPPLE: Pe…

Target-Guided Dialogue Response Generation Using Commonsense and Data Augmentation Open

Prakhar Gupta, Harsh Jhamtani, Jeffrey P. Bigham · 2022

Target-guided response generation enables dialogue systems to smoothly transition a conversation from a dialogue context toward a target sentence. Such control is useful for designing dialogue systems that direct a conversation toward spec…

Achieving Conversational Goals with Unsupervised Post-hoc Knowledge Injection Open

Bodhisattwa Prasad Majumder, Harsh Jhamtani, Taylor Berg-Kirkpatrick, Julian McAuley · 2022

A limitation of current neural dialog models is that they tend to suffer from a lack of specificity and informativeness in generated responses, primarily due to dependence on training data that covers a limited variety of scenarios and con…

Target-Guided Dialogue Response Generation Using Commonsense and Data Augmentation Open

Prakhar Gupta, Harsh Jhamtani, Jeffrey P. Bigham · 2022

Target-guided response generation enables dialogue systems to smoothly transition a conversation from a dialogue context toward a target sentence. Such control is useful for designing dialogue systems that direct a conversation toward spec…

Achieving Conversational Goals with Unsupervised Post-hoc Knowledge Injection Open

Bodhisattwa Prasad Majumder, Harsh Jhamtani, Taylor Berg-Kirkpatrick, Julian McAuley · 2022

Bodhisattwa Prasad Majumder, Harsh Jhamtani, Taylor Berg-Kirkpatrick, Julian McAuley. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Truth-Conditional Captioning of Time Series Data Open

Harsh Jhamtani, Taylor Berg-Kirkpatrick · 2021

In this paper, we explore the task of automatically generating natural language descriptions of salient patterns in a time series, such as stock prices of a company over a week. A model for this task should be able to extract high-level pa…

Investigating Robustness of Dialog Models to Popular Figurative Language Constructs Open

Harsh Jhamtani, Varun Gangal, Eduard Hovy, Taylor Berg-Kirkpatrick · 2021

Humans often employ figurative language use in communication, including during interactions with dialog systems. Thus, it is important for real-world dialog systems to be able to handle popular figurative language constructs like metaphor …

Investigating Robustness of Dialog Models to Popular Figurative Language\n Constructs Open

Harsh Jhamtani, Varun Gangal, Eduard Hovy, Taylor Berg-Kirkpatrick · 2021

Humans often employ figurative language use in communication, including\nduring interactions with dialog systems. Thus, it is important for real-world\ndialog systems to be able to handle popular figurative language constructs like\nmetaph…

Unsupervised Enrichment of Persona-grounded Dialog with Background\n Stories Open

Bodhisattwa Prasad Majumder, Taylor Berg-Kirkpatrick, Julian McAuley, Harsh Jhamtani · 2021

Humans often refer to personal narratives, life experiences, and events to\nmake a conversation more engaging and rich. While persona-grounded dialog\nmodels are able to generate responses that follow a given persona, they often\nmiss out …

Improving Automated Evaluation of Open Domain Dialog via Diverse\n Reference Augmentation Open

Varun Gangal, Harsh Jhamtani, Eduard Hovy, Taylor Berg-Kirkpatrick · 2021

Multiple different responses are often plausible for a given open domain\ndialog context. Prior work has shown the importance of having multiple valid\nreference responses for meaningful and robust automated evaluations. In such\ncases, co…

The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics Open

Sebastian Gehrmann, Tosin Adewumi, Karmanya Aggarwal, Pawan Sasanka Ammanamanchi, Anuoluwapo Aremu , et al. · 2021

We introduce GEM, a living benchmark for natural language Generation (NLG), its Evaluation, and Metrics. Measuring progress in NLG relies on a constantly evolving ecosystem of automated metrics, datasets, and human evaluation standards. Du…

Improving Automated Evaluation of Open Domain Dialog via Diverse Reference Augmentation Open

Varun Gangal, Harsh Jhamtani, Eduard Hovy, Taylor Berg-Kirkpatrick · 2021

Multiple different responses are often plausible for a given open domain dialog context. Prior work has shown the importance of having multiple valid reference responses for meaningful and robust automated evaluations. In such cases, commo…

The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics Open

Sebastian Gehrmann, Tosin Adewumi, Karmanya Aggarwal, Pawan Sasanka Ammanamanchi, Anuoluwapo Aremu , et al. · 2021

International audience

Formulating Neural Sentence Ordering as the Asymmetric Traveling Salesman Problem Open

Vishal Keswani, Harsh Jhamtani · 2021

The task of Sentence Ordering refers to rearranging a set of given sentences in a coherent ordering. Prior work (Prabhumoye et al., 2020) models this as an optimal graph traversal (with sentences as nodes, and edges as local constraints) u…

Truth-Conditional Captions for Time Series Data Open

Harsh Jhamtani, Taylor Berg-Kirkpatrick · 2021

In this paper, we explore the task of automatically generating natural language descriptions of salient patterns in a time series, such as stock prices of a company over a week. A model for this task should be able to extract high-level pa…

Unsupervised Enrichment of Persona-grounded Dialog with Background Stories Open

Bodhisattwa Prasad Majumder, Taylor Berg-Kirkpatrick, Julian McAuley, Harsh Jhamtani · 2021

Humans often refer to personal narratives, life experiences, and events to make a conversation more engaging and rich. While persona-grounded dialog models are able to generate responses that follow a given persona, they often miss out on …

Investigating Robustness of Dialog Models to Popular Figurative Language Constructs Open

Harsh Jhamtani, Varun Gangal, Eduard Hovy, Taylor Berg-Kirkpatrick · 2021

Humans often employ figurative language use in communication, including during interactions with dialog systems. Thus, it is important for real-world dialog systems to be able to handle popular figurative language constructs like metaphor …

Like hiking? You probably enjoy nature: Persona-grounded Dialog with\n Commonsense Expansions Open

Bodhisattwa Prasad Majumder, Harsh Jhamtani, Taylor Berg-Kirkpatrick, Julian McAuley · 2020

Existing persona-grounded dialog models often fail to capture simple\nimplications of given persona descriptions, something which humans are able to\ndo seamlessly. For example, state-of-the-art models cannot infer that interest\nin hiking…

Learning to Explain: Datasets and Models for Identifying Valid Reasoning\n Chains in Multihop Question-Answering Open

Harsh Jhamtani, Peter Clark · 2020

Despite the rapid progress in multihop question-answering (QA), models still\nhave trouble explaining why an answer is correct, with limited explanation\ntraining data available to learn from. To address this, we introduce three\nexplanati…

Domain Adaptation via Context Prediction for Engineering Diagram Search Open

Harsh Jhamtani, Taylor Berg-Kirkpatrick · 2020

Harsh Jhamtani YOU? Author Swipe