Papers by Heidar Davoudi
Generating Spatial Knowledge Graphs from Automotive Diagrams for Question Answering (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Useful answers require obvious landmarks as a reference point . a decomposed pipeline is the most effective strategy for generating a high-quality SKG . |
| Approach: | They propose to generate a spatial knowledge graph from a vehicle dashboard diagram . they use large vision-language models to generate the graph using a decomposed pipeline . |
| Outcome: | The proposed method identifies landmarks with 71.3% agreement with human annotators on a new vehicle dataset. |
Question Generation Using Sequence-to-Sequence Model with Semantic Role Labels (2023.eacl-main)
Copied to clipboard
| Challenge: | Existing question generation methods that generate multiple questions from text are labor-intensive and do not capture the complexity of ways a human asks questions. |
| Approach: | They propose a question generation method that combines the benefits of rule-based and neural sequence-to-sequence (Seq2Sequen) models. |
| Outcome: | The proposed method significantly improves the state-of-the-art neural question generation approaches on three real-world data sets. |
GEAR: A Scalable and Interpretable Evaluation Framework for RAG-Based Car Assistant Systems (2025.emnlp-industry)
Copied to clipboard
Niloufar Beyranvand, Hamidreza Dastmalchi, Aijun An, Heidar Davoudi, Winston Chan, Ron DiCarlantonio
| Challenge: | Large language models (LLMs) increasingly power car assistants, but evaluating response quality remains a challenge. |
| Approach: | They propose a framework that uses large language models as evaluators to compare assistant responses against ground-truth counterparts. |
| Outcome: | The proposed framework compares assistant responses against ground-truth counterparts, assessing coverage, correctness, and other dimensions of answer quality. |
Generating Vehicular Icon Descriptions and Indications Using Large Vision-Language Models (2024.emnlp-industry)
Copied to clipboard
James Fletcher, Nicholas Dehnen, Seyed Nima Tayarani Bathaie, Aijun An, Heidar Davoudi, Ron DiCarlantonio, Gary Farmaner
| Challenge: | Existing image description systems are trained mainly on natural images, whereas icon images are drawings. |
| Approach: | They propose to use a dataset to generate both visual and functional icon descriptions based on the icon image and its context information in the car manual. |
| Outcome: | The proposed model performs well on the dashboard icon description task while the third model perform poorly. |
Neural Document Segmentation Using Weighted Sliding Windows with Transformer Encoders (2025.coling-industry)
Copied to clipboard
| Challenge: | Using overlapping text sequences and position-aware weighting, we achieve up to a 10% increase in segmentation F1 score compared to existing methods. |
| Approach: | They propose a Transformer-based method for document segmentation that utilizes overlapping text sequences with a unique position-aware weighting mechanism to enhance segmentation accuracy. |
| Outcome: | The proposed method achieves up to 10% increase in segmentation F1 score compared to existing methods and improves quality of generated responses by 5% while achieving four times greater efficiency. |
Content-based Dwell Time Engagement Prediction Model for News Articles (N19-2)
Copied to clipboard
| Challenge: | Existing studies on article dwell time prediction are noisy and may not show the actual user engagement or satisfaction. |
| Approach: | They propose a deep neural network architecture to extract emotion, event and entity features from an article and learn interactions among them. |
| Outcome: | The proposed model outperforms state-of-the-art models on a real newspaper dataset. |
Affective and Contextual Embedding for Sarcasm Detection (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods to detect sarcasm from text lack vocal intonation or facial gestures in textual data. |
| Approach: | They propose two deep neural network models for sarcasm detection that extend the architecture of BERT by incorporating both affective and contextual features. |
| Outcome: | The proposed models outperform state-of-the-art models on different datasets with significant margins. |