Shubhashis Roy Dipta, PhD Researcher in NLP and Multimodal AI
Upcoming Travel
- NeurIPS 2026 in Atlanta, GA (Dec 9-13)
- ICLR 2027 in San Francisco, CA (Apr 26-30)
- NAACL 2027 in San Francisco, CA (Jun 1-5)
I’m a final-year CS Ph.D. researcher at the University of Maryland, Baltimore County (UMBC), advised by Dr. Frank Ferraro, with research internships at Amazon (Alexa AI; Summer 2025 + 2026) and Scale AI (Summer 2024). I make models expose the structure behind an answer, events, questions, policies, graphs, then train and score against that structure with GRPO or OPD. My work on agentic LLMs earned a $20K Google Cloud Gemini Academic Program Award (2026).
- Agentic LLM & Tool Use AgentCollabBench (NeurIPS 2026) TRIAGE (NeurIPS 2026) PA3 (EMNLP 2026)
- Post-Training (RL, OPD) Rock Tokens (NeurIPS 2026) DAGGER (EMNLP 2026) GanitLLM (ACL 2026)
- Multimodal Learning & Vision VC-Inspector (ACL 2026) Q2E (AACL 2025)
- Reasoning & Decomposition De-Presuppose (*SEM 2025) SHEM (*SEM 2023)
Graduating Spring 2027 · No visa sponsorship needed · actively seeking Research Scientist roles in NLP / Multimodal AI. Please reach out if you have an opening.
Recent News (See All)
| Oct 28, 2026 | 🧑⚖️ Invited as Area Chair for the ARR October 2026 cycle (NAACL 2027), and reviewed 6 papers for ICLR 2027. |
|---|---|
| Oct 22, 2026 | 🌍 New preprint - CulturalRiddles: human-written riddles from 62 cultures and 51 languages. The best models reach only 22% in some cultures. News coverage by Fast Company and Cohere Labs (X, LinkedIn). |
| Sep 26, 2026 | 🎉 Three papers accepted at NeurIPS 2026 - Rock Tokens to the main conference, plus TRIAGE and AgentCollabBench to the E&D track. |
| Sep 6, 2026 | 🚀 New preprint - OracleZoom: recursive super-resolution to 256× with ground truth only at 4×. A VLM judge prefers it over Chain-of-Zoom 78% of the time. Demo. |
Beyond Research
I’ve competed internationally in algorithms and robotics, ranking 8th out of 300+ teams at the 2018 ACM ICPC Asia Dhaka Regional with multiple regional and national placements, reaching the top 70 on Kaggle 🥉 in the Birdcall Identification competition, and placing 9th at the University Rover Challenge 2015 (Utah, USA) and 22nd at the European Rover Challenge 2016 (Poland).
Before the PhD, I also founded UniShopr (2019-2021), a cross-border e-commerce platform serving consumers in Bangladesh.
Featured Publications
A few papers that best represent my work. See all publications
-
Under reviewOracleZoom: On-Policy Self-Distillation Inspired Reference-Constrained Recursive Image Super Resolution† Equal contribution -
Under reviewDecomposeRL: Learning to Ask Useful, Informative, and Diverse Questions for Semi-Supervised, Traceable Claim Verification -
NeurIPS 2026Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation† Equal contribution -
NeurIPS 2026TRIAGE: Evaluating Prospective Metacognitive Control in LLMs under Resource Constraints -
EMNLP 2026PA3: Policy-Aware Agent Alignment through Chain-of-ThoughtWork done during internship at Amazon Alexa AI
Patents
- AI-Based System and Method for Code Localization Using Dependency Graph Retrieval for Bug Identification in Software Repositories
- System and Method for Anchor-Based Memory Indexing and Policy-Driven Retrieval Across Heterogeneous Sources
Research Mentorship
- MeLLM @ ACL 2026BLP @ AACL-IJCNLP 2025
- GenAI4Health @ NeurIPS 2025
- LoResLM @ EACL 2026BLP @ AACL-IJCNLP 2025
- Rakib Hossan
Peer Review
Area chair and reviewer, 47+ papers across top venues since 2023.