Abhinav Narayan Harish — accepted AI-conference papers

2026

Why DDIM Hallucinates More than DDPM: A Theoretical Analysis of Reverse Dynamics

Muhammad Ashiq, Samanyu Arora, Abhinav Narayan Harish, Ishaan Kharbanda, Hung Yun Tseng, Grigorios Chrysos

ICML 2026poster

We theoretically study the hallucination phenomena in two canonical diffusion samplers: the stochastic Denoising Diffusion Probabilistic Model (DDPM) and the deterministic Denoising Diffusion Implicit Model (DDIM). We analyze the reverse ODE (DDIM) and SDE (DDPM) for a Gaussian mixture target, provi…

Cited by 0 Source ↗Scholar ↗

2025

Reinforcement Learning Within the Classical Robotics Stack: A Case Study in Robot Soccer

Adam Labiosa, Zhihan Wang, Siddhant Agarwal, William Cong, Geethika Hemkumar, Abhinav Narayan Harish +7

ICRA 2025

Robot decision-making in partially observable, real-time, dynamic, and multi-agent environments remains a difficult and unsolved challenge. Model-free reinforcement learning (RL) is a promising approach to learning decisionmaking in such domains, however, end-to-end RL in complex environments is oft

Cited by 6 Source ↗Scholar ↗