Papers by Vishakh Padmakumar
Intent-aware Schema Generation and Refinement for Literature Review Tables (2025.findings-emnlp)
Copied to clipboard
| Challenge: | ambiguity in reference-based evaluations and lack of editing/refinement methods have slow progress on schema generation. |
| Approach: | They propose a method for augmenting unannotated table corpora with synthesized intents . they propose prompted workflows and fine-tuned models to improve schema generation . |
| Outcome: | The proposed approach significantly improves baseline performance in reconstructing reference schemas. |
Exploring the Role of Task Transferability in Large-Scale Multi-Task Learning (2022.naacl-main)
Copied to clipboard
| Challenge: | Recent work has found that multi-task training with a large number of diverse tasks can uniformly improve downstream performance on unseen target tasks. |
| Approach: | They aim to disentangle the effect of scale and relatedness of tasks in multi-task representation learning by increasing the number of tasks and incorporating smaller sets of related tasks. |
| Outcome: | The proposed model improves on unseen target tasks by increasing the scale of multi-task learning to incorporate more tasks and developing similarity metrics to incorporate tasks related to the target task. |
Principled Content Selection to Generate Diverse and Personalized Multi-Document Summaries (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models exhibit the _”lost in the middle” phenomenon when they are unevenly attending to different parts of the provided context. |
| Approach: | They propose principled content selection as a way to increase source coverage . they use determinantal point processes to prioritize diverse content . |
| Outcome: | The proposed method improves source coverage on the DiverseSumm benchmark. |
Help me write a Poem: Instruction Tuning as a Vehicle for Collaborative Poetry Writing (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent work in training large language models to follow natural language instructions has opened up exciting opportunities for natural language interface design. |
| Approach: | They propose to train large language models to follow natural language instructions and to test whether LLMs improve the quality of the generated content. |
| Outcome: | The proposed system is competitive to publicly available LLMs trained on instructions and can satisfy unseen compositional instructions. |
Creative Natural Language Generation (2023.emnlp-tutorial)
Copied to clipboard
| Challenge: | This tutorial aims to bring awareness of the important and emerging research area of open-domain creative generation. |
| Approach: | They will review recent studies on creative language generation at sentence level as well as longer forms of text. |
| Outcome: | This paper reviews recent studies on creative language generation at sentence level as well as longer forms of text. |
Whose Boat Does it Float? Improving Personalization in Preference Tuning via Inferred User Personas (2025.acl-long)
Copied to clipboard
| Challenge: | a recent study shows that LLMs can't tailor outputs to users with uncommon preferences . despite the success of persona inference, we may need debiasing and abstention. |
| Approach: | They propose to use preference data to infer needs and interests of users who prefer either output . they argue that training on preference data augmented with PI boosts personalization . |
| Outcome: | The proposed method can be used to improve personalization with less privacy concerns. |
Machine-in-the-Loop Rewriting for Creative Image Captioning (2022.naacl-main)
Copied to clipboard
| Challenge: | Prior work has shown that providing users with a machine-written draft or sentence-level continuations has limited success since the generated text tends to deviate from users’ intention. |
| Approach: | They propose to train a rewriting model that modifies specified spans of text within the user’s original draft to introduce descriptive and figurative elements in the text. |
| Outcome: | The proposed model is rated more helpful by users than a baseline infilling language model on a user study through Amazon Mechanical Turk. |
QuALITY: Question Answering with Long Input Texts, Yes! (2022.naacl-main)
Copied to clipboard
Richard Yuanzhe Pang, Alicia Parrish, Nitish Joshi, Nikita Nangia, Jason Phang, Angelica Chen, Vishakh Padmakumar, Johnny Ma, Jana Thompson, He He, Samuel Bowman
| Challenge: | Existing models for natural language understanding are limited to processing only a few hundred words at a time. |
| Approach: | They propose a dataset with context passages in English that have an average length of 5,000 tokens. |
| Outcome: | a new dataset with long-text comprehension questions is used to test models on long-document comprehension . the questions are validated by contributors who have read the entire passage, not just excerpts . only half of the questions can be answered by annotators working under tight time constraints . |
Unsupervised Extractive Summarization using Pointwise Mutual Information (2021.eacl-main)
Copied to clipboard
| Challenge: | Unsupervised approaches to extractive summarization rely on notion of sentence importance defined by semantic similarity between a sentence and the document. |
| Approach: | They propose a method to measure relevance and redundancy using PMI between sentences. |
| Outcome: | The proposed method outperforms similarity-based methods on news, medical journal articles, and personal anecdotes. |
BBQ: A hand-built bias benchmark for question answering (2022.findings-acl)
Copied to clipboard
Alicia Parrish, Angelica Chen, Nikita Nangia, Vishakh Padmakumar, Jason Phang, Jana Thompson, Phu Mon Htut, Samuel Bowman
| Challenge: | NLP models learn social biases, but little work has been done on how these biase manifest in outputs for applied tasks like question answering (QA). |
| Approach: | They propose a dataset that highlights attested social biases against people belonging to protected classes along nine social dimensions relevant for U.S. English-speaking contexts. |
| Outcome: | The proposed dataset highlights attested social biases against people belonging to protected classes along nine social dimensions relevant for U.S. English-speaking contexts. |
Reward Gaming in Conditional Text Generation (2023.acl-long)
Copied to clipboard
| Challenge: | Recent work has used reward functions learned from human annotations to align conditional text generation models with desired behaviors. |
| Approach: | They propose to use reinforcement learning to train conditional text generation models with reward functions learned from human annotations to align outputs with desired behaviors. |
| Outcome: | The proposed framework improves the quality of generated summaries by using saliency and faithfulness metrics. |