Adam Khoja

EnigmaEval: A Benchmark of Long Multimodal Reasoning Challenges

2025 · arXiv (Cornell University)

As language models master existing reasoning benchmarks, we need new challenges to evaluate their cognitive frontiers. Puzzle-solving events are rich repositories of challenging multimodal problems that test a wide range of advanced reasoning and knowledge …

Multi-Agent Inverse Q-Learning from Demonstrations

2025

When reward functions are hand-designed, deep reinforcement learning algorithms often suffer from reward misspecification, causing them to learn suboptimal policies in terms of the intended task objectives. In the single-agent case, inverse reinforcement learning (IRL) …