5 papers
Tool-MCoT: Tool Augmented Multimodal Chain-of-Thought for Content Safety Moderation
Shutong Zhang, Dylan Zhou, Yinxiao Liu +3
The growth of online platforms and user content requires strong content moderation systems that can handle complex inputs from various media types. While large language models (LLM…
CoLD: Counterfactually-Guided Length Debiasing for Process Reward Models in Mathematical Reasoning
Congmin Zheng, Jiachen Zhu, Jianghao Lin +6
Process Reward Models (PRMs) play a central role in evaluating and guiding multi-step reasoning in large language models (LLMs), especially for mathematical problem solving. Howeve…
Beyond Markovian: Reflective Exploration via Bayes-Adaptive RL for LLM Reasoning
Shenao Zhang, Yaqing Wang, Yinxiao Liu +5
Large Language Models (LLMs) trained via Reinforcement Learning (RL) have exhibited strong reasoning capabilities and emergent reflective behaviors, such as rethinking and error co…
Dual-Head Reasoning Distillation: Improving Classifier Accuracy with Train-Time-Only Reasoning
Jillian Xu, Dylan Zhou, Vinay Shukla +6
Chain-of-Thought (CoT) prompting often improves classification accuracy, but it introduces a significant throughput penalty with rationale generation (Wei et al., 2022; Cheng and V…
Improve Mathematical Reasoning in Language Models by Automated Process Supervision
Liangchen Luo, Yinxiao Liu, Rosanne Liu +9
Complex multi-step reasoning tasks, such as solving mathematical problems or generating code, remain a significant hurdle for even the most advanced large language models (LLMs). V…