@inproceedings{Bhosale2026FairLLaVA,author={Bhosale, Mahesh and Wasi, Abdul and Shrivastva, Shantam and Latif, Shifa and Luan, Tianyu and Gao, Mingchen and Doermann, David and Gong, Xuan},title={FairLLaVA: Fairness-Aware Parameter-Efficient Fine-Tuning for Large Vision-Language Assistants},booktitle={IEEE/CVF Conference on Computer Vision and Pattern Recognition},year={2026},keywords={published},}
@inproceedings{Shrivastava2026Radiology,author={Shrivastava, Shantam and Bhosale, Mahesh and Doermann, David and Gao, Mingchen},title={Category-wise Structured Radiology Report Generation with Contrastive Decoding},booktitle={Medical Imaging with Deep Learning},year={2026},keywords={published},note={Oral presentation (≤ 8% of submitted papers)},}
@inproceedings{Bhosale2026CRAFT,author={Bhosale, Mahesh and Wasi, Abdul and Trivedi, Vishvesh and Yan, Pengyu and Gorugantu, Akhil V S S and Doermann, David},title={CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering},booktitle={ACL Multimodal Augmented Generation via MultimodAl Retrieval Workshop},year={2026},keywords={published},}
@inproceedings{Yan2026TRACE,author={Yan, Pengyu and Gorugantu, Akhil V S S and Bhosale, Mahesh and Wasi, Abdul and Trivedi, Vishvesh and Doermann, David},title={TRACE: Evidence Grounding-Guided Multi-Video Event Understanding and Claim Generation},booktitle={ACL Multimodal Augmented Generation via MultimodAl Retrieval Workshop},year={2026},keywords={published},}
@inproceedings{Bhosale2025PathDiff,author={Bhosale, Mahesh and Wasi, Abdul and Zhai, Yuanhao and Tian, Yunjie and Border, Samuel and Xi, Nan and Sarder, Pinaki and Yuan, Junsong and Doermann, David and Gong, Xuan},title={PathDiff: Histopathology Image Synthesis with Unpaired Text and Mask Conditions},booktitle={IEEE/CVF International Conference on Computer Vision},year={2025},keywords={published},}
@inproceedings{Pham2025AutoEdit,author={Pham, Chau and Dao, Quan and Bhosale, Mahesh and Tian, Yunjie and Metaxas, Dimitris N. and Doermann, David},title={AutoEdit: Automatic Hyperparameter Tuning for Image Editing},booktitle={Neural Information Processing Systems},year={2025},keywords={published},}
Chart visualizations are essential for data interpretation and communication; however, most charts are only accessible in image format and lack the corresponding data tables and supplementary information, making it difficult to alter their appearance for different scenarios of application. To eliminate the need for original underlying data and information to perform chart editing, we propose ChartReformer, a natural language-driven chart image editing solution that directly edits the charts from the input images with the given instruction prompts. Instead of predicting the plotting code, the key in this method is that we allow the model to comprehend the chart and reason over the prompt to generate the corresponding underlying data table and visual attributes for new charts, enabling a precise and stable editing result.
@inproceedings{10.1007/978-3-031-70533-5_26,author={Yan, Pengyu and Bhosale, Mahesh and Lal, Jay and Adhikari, Bikhyat and Doermann, David},title={ChartReformer: Natural Language-Driven Chart Image Editing},booktitle={Document Analysis and Recognition - ICDAR 2024},year={2024},publisher={Springer Nature Switzerland},address={Cham},pages={453--469},isbn={978-3-031-70533-5},keywords={published},url={https://link.springer.com/chapter/10.1007/978-3-031-70533-5_26},}
Data extraction from line-chart images is an essential component of the automated document understanding process, as line charts are a ubiquitous data visualization format. However, the amount of visual and structural variations in multi-line graphs makes them particularly challenging for automated parsing. Existing works, however, are not robust to all these variations, either taking an all-chart unified approach or relying on auxiliary information such as legends for line data extraction. In this work, we propose LineFormer, a robust approach to line data extraction using instance segmentation. We achieve state-of-the-art performance on several benchmark synthetic and real chart datasets.
@inproceedings{10.1007/978-3-031-41734-4_24,author={Lal, Jay and Bhosale, Mahesh and Mitkari, Aditya and Doermann, David},title={LineFormer: Line Chart Data Extraction Using Instance Segmentation},booktitle={Document Analysis and Recognition - ICDAR 2023},year={2023},publisher={Springer Nature Switzerland},address={Cham},pages={387--400},isbn={978-3-031-41734-4},keywords={published},url={https://link.springer.com/chapter/10.1007/978-3-031-41734-4_24},}
We propose a neural network architecture that learns body part appearances for soccer player re-identification. Our model consists of a two-stream network and a bilinear-pooling layer that generates and spatially pools the body part map.
@article{bhosale2023player,title={Player Re-Identification Using Body Part Appearences},author={Bhosale, Mahesh and Kumar, Abhishek and Doermann, David},journal={arXiv preprint arXiv:2310.14469},year={2023},keywords={published},}
Under Review
2026
Variance-Guided Score Regularization for Hallucination Mitigation in Diffusion Models
Mahesh Bhosale, Naresh Kumar Devulapalli, Abdul Wasi, and 3 more authors
@misc{Bhosale2026VarianceGuided,author={Bhosale, Mahesh and Devulapalli, Naresh Kumar and Wasi, Abdul and Pham, Chau and Lokhande, Vishnu and Doermann, David},title={Variance-Guided Score Regularization for Hallucination Mitigation in Diffusion Models},year={2026},keywords={under_review},}
RADI3N: Training-Free GeometRic ImAge EDiting vIa Parametric 3D CoNtrol
Abdul Wasi, Akhil V S S Gorugantu, Mahesh Bhosale, and 4 more authors
@misc{Wasi2026RADI3N,author={Wasi, Abdul and Gorugantu, Akhil V S S and Bhosale, Mahesh and Nag, Sauradip and Dutta, Anjan and Yuan, Junsong and Doermann, David},title={RADI3N: Training-Free GeometRic ImAge EDiting vIa Parametric 3D CoNtrol},year={2026},keywords={under_review},}
