O1 Replication Journey -- Part 3: Inference-time Scaling for Medical Reasoning

Building upon our previous investigations of O1 replication (Part 1: Journey Learning [Qin et al., 2024] and Part 2: Distillation [Huang et al., 2024]), this work explores the potential of inference-time scaling in large language models (LLMs) for medical reasoning tasks, ranging from diagnostic decision-making to treatment planning. Through extensive experiments on medical benchmarks of varying complexity (MedQA, Medbullets, and JAMA Clinical Challenges), our investigation reveals several key insights: (1) Increasing inference time does lead to improved performance. With a modest training set of 500 samples, our model yields substantial performance improvements of 6%-11%. (2) Task complexity directly correlates with the required length of reasoning chains, confirming the necessity of extended thought processes for challenging problems. (3) The differential diagnoses generated by our model adhere to the principles of the hypothetico-deductive method, producing a list of potential conditions that may explain a patient's symptoms and systematically narrowing these possibilities by evaluating the evidence. These findings demonstrate the promising synergy between inference-time scaling and journey learning in advancing LLMs' real-world clinical reasoning capabilities.

Class-based differences in moral judgment: A bayesian approach - Theory and Society

This study employs Bayesian inference to explore class-based differences in moral judgment. Based on the dual-process perspective in interdisciplinary action theory, we estimate in a first step a process model which differentiates parametrically between emotionally driven deontological, deliberatively driven utilitarian, and residual judgmental inclinations. In a second step, our estimates of these parameters are correlated via beta regressions with indicators of social class and thinking dispositions. We find a considerable association between social class, specifically income, and deontological inclinations, whereas consequentialist inclinations correlate with thinking dispositions but not with social class. This research underscores the utility of Bayesian estimation in closing the gap between theoretical and statistical modelling. Employing this approach enhances our understanding of the nuanced interplay between intuitive and deliberative processes in moral judgment and, more generally, offers a promising direction for advancing sociological action theory.

Learn Beyond The Answer: Training Language Models with Reflection for Mathematical Reasoning

Supervised fine-tuning enhances the problem-solving abilities of language models across various mathematical reasoning tasks. To maximize such benefits, existing research focuses on broadening the training set with various data augmentation techniques, which is effective for standard single-round question-answering settings. Our work introduces a novel technique aimed at cultivating a deeper understanding of the training problems at hand, enhancing performance not only in standard settings but also in more complex scenarios that require reflective thinking. Specifically, we propose reflective augmentation, a method that embeds problem reflection into each training instance. It trains the model to consider alternative perspectives and engage with abstractions and analogies, thereby fostering a thorough comprehension through reflective reasoning. Extensive experiments validate the achievement of our aim, underscoring the unique advantages of our method and its complementary nature relative to existing augmentation techniques.

OhioLINK ETD: Seeling, Ashley Diane