All Updates

Thrilled to share that our paper ACPBench-Hard has been accepted to ICLR 2026! 🎉 It’s incredibly rewarding to see this work is already gaining traction in the research community — with 10,000+ downloads on Hugging Face.

Jan '26

Will be attending NeurIPS 2025 in San Diego. Presenting a Tutorial on Planning in the Era of Language Models and showcasing Query Gym at IBM Booth.

Dec '25

The work on Black-Box Uncertainty Quantification for Large Language Models via Ensemble-of-Ensembles lead by Wang Ma in the Summer of 2025 is now accepted at Assessing and Improving Reliability of Foundation Models in the Real World (AIR-FM) Workshop at AAAI 2026.

Nov '25

Gave an invited talk at NxtAI Conference in San Francisco on LLMs for AI Planning.

Nov '25

Will be presenting LLMs as MoldMakers; not 3D Printers at BayLearn - Machine Learning Symposium (2025) at Santa Clara University, CA.

Oct '25

Gave an Invited talk at the AI Institute at UofSC on LLMs for AI Planning
slide | video

Sep '25

Excited to share that PLAN-FM Bridge Program is accepted to AAAI 2026.

Sep '25

Will be presenting a Tutorial on Planning in the Era of Language Models at NeurIPS 2025 in San Diego, CA.

Aug '25

Presented my perspective on using LLMs as Mold Makers; not 3D Printers to AWING
slide

Jun '25

Gave a brief tutorial on LLMs for AI Planning at the 1st DAAIG workshop on Modern LLMs and Applications.
slides

Mar '25

Presented a tutorial on Benchmarks for Planning in Natural Language at the PLAN-FM Bridge at AAAI.
slides | video

Feb '25

ACPBench: Reasoning about Action, Change, and Planning was accepted to AAAI 2025.
video

Feb '25

TabSketchFM paper got accepted at ICDE 2025. This work will also be presented at the NeurIPS TRL workshop on Dec 14th, in Vancouver.

Dec '24

Co-organizing PLAN-FM Bridge at AAAI.

Dec '24

We released a question-answering style dataset that evaluates AI-model’s ability to reason about action, change, and planning, ACP Bench. It is available on the GitHub as well as on the 🤗 HuggingFace Datasets.

Oct '24

Co-organizing the PRL workshop at AAAI 2025 in Philadelphia, Pennsylvania. The CFP is out, and the submission site is open. Check out the PRL website for details.

Oct '24

Extremely happy to share that our work Thought of Search: Planning with Language Models Through The Lens of Efficiency got accepted at NeurIPS 2024. Awesome work led by Michael Katz. Congratulations to Michael Katz, Shirin Sohrabi and Kavitha Srinivas.

Sep '24

Katz et al. Thought of Search (ToS) and Cao et al. Automating ToS got featured on the venture beat. Thanks Ben Dickson for the coverage.

Sep '24

Can LLMs automatically produce sound and complete functions for search? Our work Cao et al. Automating Thought of Search: A Journey Towards Soundness and Completeness shows that we can. Dive in to find how! This was also featured in Top 3 researches of this week by The Turing Post on X.

Aug '24

Should LLMs be used for Planning? We look at this question through the lens of efficiency in our latest work, now available on ArXiv.
Katz et al. Thought of Search: Planning with Language Models Through The Lens of Efficiency.

Jun '24

Our Sketch-based tabular representation learning approach, TabSketchFM is now available on ArXiv.

Jun '24

Our work on using Large Language Models as Planning Domain Generators was presented at in the main track at ICAPS 2024 in Banff and two other works, Guiding Hiearchical Reinforcement Learning in Partially Observable Environments with AI Planning and Planning with Language Models Through The Lens of Efficiency, were presented at the PRL Workshop.

Jun '24

Pleased to announce the arrival of our son; Kush Kokel Singh. The most intricate neural network I’ve had the pleasure to create.

May '24

My Ph.D. Thesis was honored with the David Daniel Thesis Award by the faculty in the Jonsson School of Engineering and Computer Science at UT Dallas.

Mar '24

Two abstracts were accepted at AAAI 2024 Student Abstract Program:
Oswald et al. Large Language Models as Planning Domain Generators, and
Rozek et al. Partially Observable Hierarchical Reinforcement Learning with AI Planning.

Oct '23

Attended IJCAI in Macao, S.A.R, where I presented our work on Action Space Reduction for Planning Domains and also co-organized the sixth edition of the Planning and RL Workshop (PRL)

Aug '23

Presented my work on Integrated Planning and Reinforcement Learning to Almaden Women’s Interest and Networking Group (AWING) at IBM
slide

Jun '23

Gave an Invited talk at the AI Seminar in Oregon State University on Integrated Planning and Reinforcement Learning for Compositional Domains
slide | video

May '23

Got formally hooded as Doctor of Philosophy in Computer Science at UT Dallas.

May '23

IJCAI accepted our work on Action Space Reduction for Planning Domains.

Apr '23

Joined as a Research Scientist at the AI Planning and Applications group in IBM Research.

Mar '23

Successfully defended my Ph.D. thesis, Beyond Data: Efficient Knowledge-guided Learning for Sparse and Structured Domains, under the guidance of Prof. Sriraam Natarajan

Feb '23

Gave an Invited talk at the Robotics Lab in the Brown University on Dynamic probabilistic logic models for effective task-specific abstractions in RL.
slide | video

Sep '22

Presented our work on Hybrid Deep RePReL: Integrating Relational Planning and Reinforcement Learning for Information Fusion at IEEE 25th International Conference on Information Fusion (FUSION), Linköping, Sweden.
slide | video

Jul '22

Gave an invited talk at Eleventh RBCDSAI Workshop on Recent progress in Data Science and AI, IIT Madras, India on RePReL: Integrating Relational Planning and Reinforcement Learning for Effective Abstractions.
slide | video

Nov '21

Addressed Ph.D. mixer at UT Dallas. Presented our lab’s work on Human Allied Artificial Intelligence.
slide

Sep '21

Gave an invited talk at AAIR Lab Spring Meeting, Arizon State University on A Unified Framework for Knowledge Intensive Gradient Boosting: Leveraging Human Experts for Noisy Sparse Domain.

Jan '21