Comparative Analysis of Reasoning Accuracy in Multimodal Large Language Models and Diffusion-Based Trajectory Policies on

Large Language Models (LLM) with reasoning capabilities offer a promising path for improving candidate evaluation in planning frameworks, but their relative performance against traditional non-reasoning models remains largely underexplored. In this study, we benchmark a distilled 1.5B parameter reasoning model (DeepSeek-R1) against several state-of-the-art non-reasoning LLMs within a generator-discriminator LLM planning framework for the text-to-SQL task. For this, we introduce a novel method for extracting soft scores from the chain-of-thought (CoT) outputs from reasoning that enables fine-gr

Research goal: How does the reasoning accuracy of multimodal large language models compare to diffusion-based trajectory policies in dynamic task planning environments when evaluated on the RoboBench benchmark with varying levels of environmental noise?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 7.6/10.