PRIMT: Preference-based Reinforcement Learning with Multimodal Feedback and Trajectory Synthesis from Foundation Models

Ruiqi Wang (Purdue University) · Dezhong Zhao (Beijing University of Chemical Technology) · Ziqin Yuan (Purdue University) · Tianyu Shao (Purdue University) · Guohua Chen (Beijing University of Chemical Technology) · Dominic Kao (University of Illinois at Urbana-Champaign) · Sungeun Hong (Inha Univ.) · Byung-Cheol Min (Indiana University Bloomington)

causal auxiliary losscredit assignmentforesight trajectory generationfoundation modelshierarchical neuro-symbolic fusionhindsight trajectory augmentationmultimodal synthetic feedbackpreference-based reinforcement learningprimtquery ambiguityreward engineeringtrajectory buffertrajectory synthesisvision-language models

Preference-based reinforcement learning (PbRL) has emerged as a promising paradigm for teaching robots complex behaviors without reward engineering. However, its effectiveness is often limited by two critical challenges: the reliance on extensive human input and the inherent difficulties in resolving query ambiguity and credit assignment during reward learning. In this paper, we introduce PRIMT, a PbRL framework designed to overcome these challenges by leveraging foundation models (FMs) for multimodal synthetic feedback and trajectory synthesis. Unlike prior approaches that rely on single-modality FM evaluations, PRIMT employs a hierarchical neuro-symbolic fusion strategy, integrating the complementary strengths of vision-language models (VLMs) and large language models (LLMs) in evaluating robot behaviors for more reliable and comprehensive feedback. PRIMT also incorporates foresight trajectory generation to warm-start the trajectory buffer with bootstrapped samples, reducing early-stage query ambiguity, and hindsight trajectory augmentation for counterfactual reasoning with a causal auxiliary loss to improve credit assignment. We evaluate PRIMT on 2 locomotion and 6 manipulation tasks on various benchmarks, demonstrating superior performance over FM-based and scripted baselines. Website at https://primt25.github.io/.