Publications

CausalPulse: An Industrial-Grade Neurosymbolic Multi-Agent Copilot for Causal Diagnostics in Smart Manufacturing

Shyalika, C., Jaimini, U., Henson, C. and Sheth, A. — AAAI MAKE Spring Symposium, 2026 · Smart Pilot for Drug Manufacturing · ▶ Watch

IndustryAssetEQA: A Neurosymbolic Operational Intelligence System for Embodied Question Answering in Industrial Asset Maintenance

Shyalika, C., Patel, D.C. and Sheth, A. — 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) — Industry Track · Smart Pilot for Drug Manufacturing · ▶ Watch

SEMINAR: Synchronization Engine for Mathematical Illustration, Narration, and Reasoning

Rishov Mondal, Amit Sheth — 6th MATH-AI Workshop at NeurIPS 2026 · Project MEDHA — Mathematical Engine for Domain-specific Higher-order Autoformalisation

Answering Without Arriving: An Evidence Audit for Embodied Question Answering

Shubh Garg, Divyansh Pandey, Niyati Rawal, Vedant Khandelwal, Amit Sheth — NeurIPS 2026 Workshop on Neuro-Symbolic Embodied Intelligence · Long-Range Embodied QA: Benchmarking and Memory-Augmented Navigation in Multi-Room VLM Agents

Bag of Tricks or Bag of Myths? Reducing Modeling Complexity with Task Knowledge in Explainable Suicide Risk Assessment

Shlok Shelat, Shrey Salvi, Souvik Roy, Manas Gaur, Amit Sheth — IEEE BigData 2026 Cup — Explainable Suicide Risk Detection (Competition Report); 1st place, 55 teams · NIDHI: Explainable Suicide Risk Detection

Competition report for the NIDHI team's 1st-place system in the IEEE BigData 2026 Cup. Audits 31 standard NLP techniques from 7 methodological families across ~300 controlled experiments on 1,635 clinician-annotated posts, finding only 5 reliably improve small-data, multi-output suicide-risk assessment. A full paper is in preparation for the IEEE BigData 2026 submission deadline.

AI-enabled water resilience: Applications and future directions

Niyogi, D., Tiwari, A., Patel, A. and Neupane, B. — Global Environmental Issues, Vol. 7: AI and Environment (pp. 51–64), Korea Environment Institute, 2026

Abstention vs. Hallucination: Benchmarking LLM Source Attribution for Scientific Citations

Deepa Tilwani, Yash Saxena, Seyedali Mohammadi, Ankur Padia, Edward Raff, Amit Sheth, Srinivasan Parthasarathy, Manas Gaur — · ▶ Watch

SOFAI-DT: Return-Conditioned Control for Efficient Verified Reasoning

Rahul Maity, Vedant Khandelwal, Amit Sheth — COLM 2026 Workshop on Efficient Reasoning · ▶ Watch

MED-EVAL: Clinical Fact Alignment for Efficient Medical Reasoning

Rishit Agarwal, Vedant Khandelwal, Amit Sheth — COLM 2026 Workshop on Efficient Reasoning · ▶ Watch

Neuro-symbolic PRM: Enhancing Scientific Reasoning via Structured Traces and Symbolic Verification

Zi, Y., Xu, C., Bhattacharya, S., Foltin, M., Sheth, A. — Third Conference on Language Modeling (COLM 2026)

Cognitive Fatigue in Autoregressive Transformers: Formalization and Measurement

R. Marwah, R. Garimella, V. Pallagani, A. Jain, M. Stewart, A. Sheth — 43rd International Conference on Machine Learning (ICML 2026), Seoul, South Korea · ▶ Watch

One Size Does Not Fit All: Revisiting World Models and Neurosymbolic AI

Amit Sheth, Madhur Thareja, Anushka Pawar, Niyati Rawal — ACM AI Leadership Summit (Visionary papers), 2026

LiveSpec: A Tiered Runtime for Planning under Live-Editable Constraints

Vedant Khandelwal, Prateek Biswas, Amit Sheth — ICAPS 2026 · ▶ Watch

Where Should Knowledge Enter? A Layered Framework for Knowledge Infusion in Multimodal Iterative Generative Model

Renjith Prasad, Chathurangi Shyalika, Anushka Pawar, Amit Sheth — 2nd Workshop on Knowledge-Intensive Multimodal Reasoning, CVPR 2026 · ▶ Watch

To be published in IEEE Intelligent Systems.

Toward Neurosymbolic Reinforcement Learning via Editable Specifications

Vedant Khandelwal, Hong Yung Yip, Amit Sheth — AAAI MAKE Spring Symposium, Proceedings of the AAAI Symposium Series, Vol. 8, No. 1, pp. 603–607, 2026 · ▶ Watch

CAMEO: A Black-Box Behavioral Diagnostic for Evaluation-Aware Concealment in Language Models

Vatsalya Soni, Shaivi Malik, Vedant Maheshwari, Venkatesan Nadimuthu, Yash Kulkarni, Amit Sheth — NeurIPS 2026 Workshop on Trustworthy AI for Good (AI4Good)