publications

2026

1. NeurIPS

   

   NavAble: A Large-Scale Dataset and Synthetic Data Generation Pipeline for Blind Navigation

   Hochul Hwang^*, Jahir Sadik Monon^*, Soowan Yang, Shiven Umeshbhai Patel, Anh Nhat Hong Nguyen, Kien Nguyen, Keshav Garg, Dylan Gage, Duretti Hordofaa, Anshu Anjna, Eshed Ohn-Bar, and Donghyun Kim

   In Conference on Neural Information Processing Systems (NeurIPS), Evaluations and Datasets Track, 2026

   Abs Bib PDF Dataset Website

   Reliable recognition of accessibility-critical objects (e.g., audible pedestrian signals, door-activation buttons, and handrails) is essential for assistive navigation technologies that support safe, independent mobility for blind and low-vision (BLV) users. Yet these categories are severely underrepresented in existing large-scale vision datasets, and the few datasets that include them suffer from limited class coverage, inconsistent annotations, and poor diversity, limiting the perception reliability that BLV navigation requires. We introduce NavAble, a large-scale dataset spanning 11 accessibility-critical object classes, combining 41K curated open-source images with 8K newly collected, densely annotated real-world images. To extend scale and diversity, we develop a synthetic data generation pipeline that renders high-fidelity 3D assets across 37 environments under diverse styles and viewpoints, yielding 452K frames with rich ground-truth annotations from 565 assets. The pipeline minimizes manual effort by combining automated web image crawling, VLM-based filtering, SAM 3D-based asset generation, and convenient camera trajectory configuration, enabling large-scale data generation at a diversity unattainable through manual collection. Experiments show that fine-tuning existing segmentation models on the NavAble Dataset achieves reliable accessibility-object segmentation (>80% mIoU on our held-out test set). We also demonstrate the effectiveness of synthetic data augmentation, which yields up to +3.5 mIoU over real-world-only fine-tuning. Beyond benchmark evaluation, validation on egocentric data from two mobility-assistive robots demonstrates improved perception reliability in real-world navigation scenarios. Together, NavAble establishes a scalable foundation for BLV navigation perception, directly addressing the data scarcity and embodiment gaps that have limited progress in this domain. The dataset and pipeline are publicly available at https://huggingface.co/datasets/NavAble/NeurIPS_2026_BLV.

   @inproceedings{hwang2026navable,
     title = {NavAble: A Large-Scale Dataset and Synthetic Data Generation Pipeline for Blind Navigation},
     author = {Hwang, Hochul and Monon, Jahir Sadik and Yang, Soowan and Patel, Shiven Umeshbhai and Nguyen, Anh Nhat Hong and Nguyen, Kien and Garg, Keshav and Gage, Dylan and Hordofaa, Duretti and Anjna, Anshu and Ohn-Bar, Eshed and Kim, Donghyun},
     booktitle = {Conference on Neural Information Processing Systems (NeurIPS), Evaluations and Datasets Track},
     year = {2026},
   }

2. HRI

   

   GuideNav: User-Informed Development of a Vision-Only Robotic Navigation Assistant for Blind Travelers

   Hochul Hwang, Soowan Yang, Jahir Sadik Monon, Nicholas A. Giudice, Sunghoon Ivan Lee, Joydeep Biswas, and Donghyun Kim

   In Proceedings of the 21st ACM/IEEE International Conference on Human-Robot Interaction (HRI), 2026

   Honorable Mention Abs arXiv Bib Dataset Website

   While commendable progress has been made in user-centric research on mobile assistive systems for blind and low-vision (BLV) individuals, references that directly inform robot navigation design remain rare. To bridge this gap, we conducted a comprehensive human study involving interviews with 26 guide dog handlers, four white cane users, nine guide dog trainers, and one O&M trainer, along with 15+ hours of observing guide dog-assisted walking. After de-identification, we open-sourced the dataset to promote human-centered development and informed decision-making for assistive systems for BLV people. Building on insights from this formative study, we developed GuideNav, a vision-only, teach-and-repeat navigation system. Inspired by how guide dogs are trained and assist their handlers, GuideNav autonomously repeats a path demonstrated by a sighted person using a robot. Specifically, the system constructs a topological representation of the taught route, integrates visual place recognition with temporal filtering, and employs a relative pose estimator to compute navigation actions - all without relying on costly, heavy, power-hungry sensors such as LiDAR. In field tests, GuideNav consistently achieved kilometer-scale route following across five outdoor environments, maintaining reliability despite noticeable scene variations between teach and repeat runs. A user study with 3 guide dog handlers and 1 guide dog trainer further confirmed the system’s feasibility, marking (to our knowledge) the first demonstration of a quadruped mobile system retrieving a path in a manner comparable to guide dogs.

   @inproceedings{hwang2026guidenav,
     title = {GuideNav: User-Informed Development of a Vision-Only Robotic Navigation Assistant for Blind Travelers},
     author = {Hwang, Hochul and Yang, Soowan and Monon, Jahir Sadik and Giudice, Nicholas A. and Lee, Sunghoon Ivan and Biswas, Joydeep and Kim, Donghyun},
     booktitle = {Proceedings of the 21st ACM/IEEE International Conference on Human-Robot Interaction (HRI)},
     year = {2026},
   }

2025

1. arXiv

   

   BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generation

   Fahim Ahmed, Md Mubtasim Ahasan, Jahir Sadik Monon, Muntasir Wahed, M Ashraful Amin, A K M Mahbubur Rahman, and Amin Ahsan Ali

   arXiv preprint arXiv:2511.04153, 2025

   Abs arXiv Bib Code

   Text-to-SQL systems provide a natural language interface that can enable even laymen to access information stored in databases. However, existing Large Language Models (LLM) struggle with SQL generation from natural instructions due to large schema sizes and complex reasoning. Prior work often focuses on complex, somewhat impractical pipelines using flagship models, while smaller, efficient models remain overlooked. In this work, we explore three multi-agent LLM pipelines, with systematic performance benchmarking across a range of small to large open-source models: (1) Multi-agent discussion pipeline, where agents iteratively critique and refine SQL queries, and a judge synthesizes the final answer; (2) Planner-Coder pipeline, where a thinking model planner generates stepwise SQL generation plans and a coder synthesizes queries; and (3) Coder-Aggregator pipeline, where multiple coders independently generate SQL queries, and a reasoning agent selects the best query. Experiments on the Bird-Bench Mini-Dev set reveal that Multi-Agent discussion can improve small model performance, with up to 10.6% increase in Execution Accuracy for Qwen2.5-7b-Instruct seen after three rounds of discussion. Among the pipelines, the LLM Reasoner-Coder pipeline yields the best results, with DeepSeek-R1-32B and QwQ-32B planners boosting Gemma 3 27B IT accuracy from 52.4% to the highest score of 56.4%.

   @article{ahmed2025bappa,
     title = {BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generation},
     author = {Ahmed, Fahim and Ahasan, Md Mubtasim and Monon, Jahir Sadik and Wahed, Muntasir and Amin, M Ashraful and Rahman, A K M Mahbubur and Ali, Amin Ahsan},
     journal = {arXiv preprint arXiv:2511.04153},
     year = {2025},
   }

2. AAMAS

   

   Learning Heterogeneous Agent Collaboration in Decentralized Multi-Agent Systems via Intrinsic Motivation

   Jahir Sadik Monon^*, Deeparghya Dutta Barua^*, and Md. Mosaddek Khan

   In Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2025

   Abs arXiv Bib PDF Code

   Multi-agent Reinforcement Learning (MARL) is emerging as a key framework for various sequential decision-making and control tasks. Unlike their single-agent counterparts, multi-agent systems necessitate successful cooperation among the agents. The real-world deployment of these systems requires decentralized training and execution (DTE), diverse agents, and learning from infrequent environmental rewards. These challenges become more pronounced under partial observability and the lack of prior knowledge about agent heterogeneity. While notable studies use intrinsic motivation (IM) to address reward sparsity or cooperation in decentralized execution settings, those dealing with heterogeneity typically assume centralized training for decentralized execution (CTDE). To overcome these limitations, we propose the CoHet algorithm, which utilizes a novel Graph Neural Network (GNN) based intrinsic motivation to facilitate the learning of heterogeneous agent policies in fully decentralized settings, under the challenges of partial observability and reward sparsity. Evaluation of CoHet in the Multi-agent Particle Environment (MPE) and Vectorized Multi-Agent Simulator (VMAS) benchmarks demonstrates superior performance compared to the state-of-the-art in a range of cooperative multi-agent scenarios.

   @inproceedings{monon2025cohet,
     title = {Learning Heterogeneous Agent Collaboration in Decentralized Multi-Agent Systems via Intrinsic Motivation},
     author = {Monon, Jahir Sadik and Barua, Deeparghya Dutta and Khan, Md. Mosaddek},
     booktitle = {Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS)},
     year = {2025},
   }

2023

1. Dataset

   

   Insights into Dhaka City’s Walkability: A Crowdsourced Dataset

   Jahir Sadik Monon

   2023

   Abs Bib Supp Code Dataset

   A crowdsourced research project with a team of around 70 junior-year CSE students at the University of Dhaka to collect and analyze a dataset on the walkability of areas across Dhaka city. Data on road surface quality, footpath availability and width, traffic congestion, noise pollution, street lighting, and signage were recorded with the Gaia GPS app, cleaned, and analyzed to identify factors that shape pedestrian experience in the city.

   @misc{monon2023walkability,
     title = {Insights into Dhaka City's Walkability: A Crowdsourced Dataset},
     author = {Monon, Jahir Sadik},
     howpublished = {Kaggle dataset},
     year = {2023},
   }