Dhruvesh Patel

2025

1. NeurIPS

   

   Insertion Language Models: Sequence Generation with Arbitrary-Position Insertions

   Dhruvesh Patel, Aishwarya Sahoo, Avinash Amballa, Tahira Naseem, Tim Rudner, and 1 more author

   In Neurips (SPIGM Workshop), 2025

   HTML Code

2024

1. NAACL

   Language Guided Exploration for RL Agents in Text Environments

   Hitesh Golchha, Sahil Yerawar, Dhruvesh Patel, Soham Dan, and Keerthiram Murugesan

   In Findings of the Association for Computational Linguistics: NAACL 2024, Jun 2024

   Code

2023

1. ICCV

   

   Pretrained Language Models as Visual Planners for Human Assistance

   Dhruvesh Patel, Hamid Eghbalzadeh, Nitin Kamra, Michael Louis Iuzzolino, Unnat Jain, and 1 more author

   Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Jun 2023

   arXiv HTML Code

2022

1. ICLR

   Modeling Label Space Interactions in Multi-label Classification using Box Embeddings

   Dhruvesh Patel, Pavitra Dangati, Jay-Yoon Lee, Michael Boratko, and Andrew McCallum

   In International Conference on Learning Representations, Jun 2022

   Abs Bib

   Multi-label classification is a challenging structured prediction task in which a set of output class labels are predicted for each input. Real-world datasets often have natural or latent taxonomic relationships between labels, making it desirable for models to employ label representations capable of capturing such taxonomies. Most existing multi-label classification methods do not do so, resulting in label predictions that are inconsistent with the taxonomic constraints, thus failing to accurately represent the fundamentals of problem setting. In this work, we introduce the multi-label box model (MBM), a multi-label classification method that combines the encoding power of neural networks with the inductive bias and probabilistic semantics of box embeddings (Vilnis, et al 2018). Box embeddings can be understood as trainable Venn-diagrams based on hyper-rectangles. Representing labels by boxes rather than vectors, MBM is able to capture taxonomic relations among labels. Furthermore, since box embeddings allow these relations to be learned by stochastic gradient descent from data, and to be read as calibrated conditional probabilities, our model is endowed with a high degree of interpretability. This interpretability also facilitates the injection of partial information about label-label relationships into model training, to further improve its consistency. We provide theoretical grounding for our method and show experimentally the model’s ability to learn the true latent taxonomic structure from data. Through extensive empirical evaluations on both small and large-scale multi-label classification datasets, we show that BBM can significantly improve taxonomic consistency while preserving or surpassing the state-of-the-art predictive performance.

   @inproceedings{patel2022modeling,
     title = {Modeling Label Space Interactions in Multi-label Classification using Box Embeddings},
     author = {Patel, Dhruvesh and Dangati, Pavitra and Lee, Jay-Yoon and Boratko, Michael and McCallum, Andrew},
     booktitle = {International Conference on Learning Representations},
     year = {2022},
     url = {https://openreview.net/forum?id=tyTH9kOxcvh},
   }

2. NeurIPS

   Structured Energy Network As a Loss

   Jay Yoon Lee, Dhruvesh Patel, Purujit Goyal, Wenlong Zhao, Zhiyang Xu, and 1 more author

   In Advances in Neural Information Processing Systems, Jun 2022

   Abs

   Belanger & McCallum (2016) and Gygli et al. (2017) have shown that an energy network can capture arbitrary dependencies amongst the output variables in structured prediction; however, their reliance on gradient-based inference (GBI) makes the inference slow and unstable. In this work, we propose Structured Energy As Loss (SEAL) to take advantage of the expressivity of energy networks without incurring the high inference cost. This is a novel learning framework that uses an energy network as a trainable loss function (loss-net) to train a separate neural network (task-net), which is then used to perform the inference through a forward pass. We establish SEAL as a general framework wherein various learning strategies like margin-based, regression, and noise-contrastive, could be employed to learn the parameters of loss-net. Through extensive evaluation on multi-label classification, semantic role labeling, and image segmentation, we demonstrate that SEAL provides various useful design choices, is faster at inference than GBI, and leads to significant performance gains over the baselines.

2020

1. ACL

   Weakly Supervised Medication Regimen Extraction from Medical Conversations

   Dhruvesh Patel, Sandeep Konam, and Sai Prabhakar

   In Proceedings of the 3rd Clinical Natural Language Processing Workshop, Nov 2020

   Abs Bib

   Automated Medication Regimen (MR) extraction from medical conversations can not only improve recall and help patients follow through with their care plan, but also reduce the documentation burden for doctors. In this paper, we focus on extracting spans for frequency, route and change, corresponding to medications discussed in the conversation. We first describe a unique dataset of annotated doctor-patient conversations and then present a weakly supervised model architecture that can perform span extraction using noisy classification data. The model utilizes an attention bottleneck inside a classification model to perform the extraction. We experiment with several variants of attention scoring and projection functions and propose a novel transformer-based attention scoring function (TAScore). The proposed combination of TAScore and Fusedmax projection achieves a 10 point increase in Longest Common Substring F1 compared to the baseline of additive scoring plus softmax projection.

   @inproceedings{patel-etal-2020-weakly,
     title = {Weakly Supervised Medication Regimen Extraction from Medical Conversations},
     author = {Patel, Dhruvesh and Konam, Sandeep and Prabhakar, Sai},
     booktitle = {Proceedings of the 3rd Clinical Natural Language Processing Workshop},
     month = nov,
     year = {2020},
     address = {Online},
     publisher = {Association for Computational Linguistics},
     url = {https://aclanthology.org/2020.clinicalnlp-1.20},
     doi = {10.18653/v1/2020.clinicalnlp-1.20},
     pages = {178--193},
   }

2. AKBC

   Representing Joint Hierarchies with Box Embeddings

   *Dhruvesh Patel, *Shib Sankar Dasgupta, Michael Boratko, Xiang Li, Luke Vilnis, and 1 more author

   In Automated Knowledge Base Construction (AKBC), Nov 2020

   Abs Bib Code Slides

   Learning representations for hierarchical and multi-relational knowledge has emerged as an active area of research. Box Embeddings [Vilnis et al., 2018, Li et al., 2019] represent concepts with hyperrectangles in n-dimensional space and are shown to be capable of modeling tree-like structures efficiently by training on a large subset of the transitive closure of the WordNet hypernym graph. In this work, we evaluate the capability of box embeddings to learn the transitive closure of a tree-like hierarchical relation graph with far fewer edges from the transitive closure. Box embeddings are not restricted to tree-like structures, however, and we demonstrate this by modeling the WordNet meronym graph, where nodes may have multiple parents. We further propose a method for modeling multiple relations jointly in a single embedding space using box embeddings. In all cases, our proposed method outperforms or is at par with all other embedding methods.

   @inproceedings{dhruveshbox2020,
     title = {Representing Joint Hierarchies with Box Embeddings},
     author = {Patel, *Dhruvesh and Dasgupta, *Shib Sankar and Boratko, Michael and Li, Xiang and Vilnis, Luke and McCallum, Andrew},
     booktitle = {Automated Knowledge Base Construction (AKBC)},
     year = {2020},
     url = {https://openreview.net/forum?id=J246NSqR_l},
     video = {https://youtu.be/yqP8wjMocAs},
   }