Chan Hsu

Chan Hsu, Wei-Chun Huang, Jun-Ting Wu et al.

In this study, we address the interpretability issue in complex, black-box Machine Learning models applied to sequence data. We introduce the Model-Based tree Hidden Semi-Markov Model (MOB-HSMM), an inherently interpretable model aimed at detecting high mortality risk events and discovering hidden patterns associated with the mortality risk in Intensive Care Units (ICU). This model leverages knowledge distilled from Deep Neural Networks (DNN) to enhance predictive performance while offering clear explanations. Our experimental results indicate the improved performance of Model-Based trees (MOB trees) via employing LSTM for learning sequential patterns, which are then transferred to MOB trees. Integrating MOB trees with the Hidden Semi-Markov Model (HSMM) in the MOB-HSMM enables uncovering potential and explainable sequences using available information.

I-Ling Cheng, Chan Hsu, Chantung Ku et al.

Machine learning models are often criticized for their black-box nature, raising concerns about their applicability in critical decision-making scenarios. Consequently, there is a growing demand for interpretable models in such contexts. In this study, we introduce Model-based Deep Rule Forests (mobDRF), an interpretable representation learning algorithm designed to extract transparent models from data. By leveraging IF-THEN rules with multi-level logic expressions, mobDRF enhances the interpretability of existing models without compromising accuracy. We apply mobDRF to identify key risk factors for cognitive decline in an elderly population, demonstrating its effectiveness in subgroup analysis and local model optimization. Our method offers a promising solution for developing trustworthy and interpretable machine learning models, particularly valuable in fields like healthcare, where understanding differential effects across patient subgroups can lead to more personalized and effective treatments.

Chan Hsu, Jun-Ting Wu, Yihuang Kang

Understanding and inferencing Heterogeneous Treatment Effects (HTE) and Conditional Average Treatment Effects (CATE) are vital for developing personalized treatment recommendations. Many state-of-the-art approaches achieve inspiring performance in estimating HTE on benchmark datasets or simulation studies. However, the indirect predicting manner and complex model architecture reduce the interpretability of these approaches. To mitigate the gap between predictive performance and heterogeneity interpretability, we introduce the Causal Rule Forest (CRF), a novel approach to learning hidden patterns from data and transforming the patterns into interpretable multi-level Boolean rules. By training the other interpretable causal inference models with data representation learned by CRF, we can reduce the predictive errors of these models in estimating HTE and CATE, while keeping their interpretability for identifying subgroups that a treatment is more effective. Our experiments underscore the potential of CRF to advance personalized interventions and policies, paving the way for future research to enhance its scalability and application across complex causal inference challenges.

Chan-Tung Ku, Chan Hsu, Pei-Cing Huang et al.

Silicon sampling-using large language models (LLMs) to simulate human survey respondents-has emerged as a promising approach for augmenting traditional survey research. However, most evaluations rely on distributional comparisons rather than individual-level prediction, which risks conflating pattern matching with coherent respondent-level prediction. We propose cross-survey transfer, a more rigorous evaluation framework in which an LLM is given a respondent's answers to one set of questions and must predict their answers to entirely different questions from the same survey. Using data from the Taiwan Election and Democratization Study (TEDS) 2024, three open-weight LLMs (27B-120B parameters), and supervised machine learning baselines, we find that: (1) zero-shot LLMs achieve 52% accuracy on genuinely unseen items, closing to within 6 percentage points (pp) of a supervised random forest trained on same-population data; (2) a stable construct predictability hierarchy emerges, from 67% for partisan attitudes to 23% for sovereignty; and (3) variance collapse and safety alignment effects-two commonly cited LLM limitations-turn out to be more nuanced than previously reported, with variance collapse affecting supervised models as well and alignment effects varying dramatically across model families. These findings clarify both the promise and boundaries of silicon sampling.

Tingting Yu, Pei-Cing Huang, Chan Hsu et al.

Predicate invention (PI), the creation of new predicates to extend the hypothesis space, remains a critical bottleneck in Inductive Logic Programming (ILP). Existing methods rely on domain expertise and produce semantically opaque predicates, hindering adaptation to unfamiliar domains and cross-task reuse. We present ADVENT, an LLM-driven PI mechanism for ILP. ADVENT pairs LLM abductive generation with Prolog deductive verification, forming an iterative loop in which concrete execution results guide the LLM to refine candidate predicates. The mechanism leverages Large Language Models to identify implicit patterns in structured relational data and invent auxiliary predicates with meaningful names and definitions. Invented predicates and learned rules accumulate in a knowledge pool for cross-task reuse. Experiments on nine poker-hand concepts across seven LLMs show that LLM-driven PI achieves 58% success rate where ILP alone fails entirely, formal verification raises this to 80%, and the knowledge pool yields gains up to +31 percentage points, while producing human-interpretable rules. These results suggest that ADVENT offers a promising direction for automating predicate invention and enabling cross-task knowledge reuse in ILP.

Ting-Yu Chen, Tingting Yu, Pei-Cing Huang et al.

Confidence calibration in large language models (LLMs) is commonly evaluated by comparing predicted confidence with observed accuracy. However, such approaches do not model item difficulty, making it difficult to interpret discrepancies and to determine whether model confidence reflects genuine self-assessment or is merely a byproduct of the response generation process. To address this, we adopt a Rasch model-based latent ability framework and a metacognitive perspective, and propose Latent Confidence Alignment Error (LCAE) to measure the consistency between model self-assessment and the latent error probability implied by model ability and item difficulty. We further incorporate item difficulty as an external signal with a reasoning mechanism. Experiments on a medical-domain dataset with 20 models show that the proposed approach improves self-assessment quality without affecting model ability, and reveals an association between reliability and inference cost.

Pei-Cing Huang, Chienyu Liu, Chan Hsu et al.

Current evaluations of Large Language Models (LLMs) on logical fallacy detection focus on predicted labels, but do not establish whether those labels are supported by the reasoning the models provide. We propose ForEx (Formal Verification for Explainable Reasoning), a framework that translates LLM-generated explanations into Lean4 and verifies whether the translated rationale is derivable under encoded premises, not the logical validity of the original natural language argument. To distinguish prediction outcomes from the formal status of the supporting reasoning, we introduce the LLM Argument Verification Matrix, which separates label consistency from formal verification status. Experiments on LOGIC-Climate show that over 90% of LLM outputs can be translated into formal reasoning chains that pass verification, while agreement with human annotations remains around 20%. These results expose a systematic gap between formal derivability and label agreement, a distinction invisible to prediction-based metrics. ForEx moves LLM evaluation beyond label correctness toward machine-checkable analysis of formalized reasoning chains.

En-Qi Tseng, Pei-Cing Huang, Chan Hsu et al.

Grading programming assignments is crucial for guiding students to improve their programming skills and coding styles. This study presents an automated grading framework, CodEv, which leverages Large Language Models (LLMs) to provide consistent and constructive feedback. We incorporate Chain of Thought (CoT) prompting techniques to enhance the reasoning capabilities of LLMs and ensure that the grading is aligned with human evaluation. Our framework also integrates LLM ensembles to improve the accuracy and consistency of scores, along with agreement tests to deliver reliable feedback and code review comments. The results demonstrate that the framework can yield grading results comparable to human evaluators, by using smaller LLMs. Evaluation and consistency tests of the LLMs further validate our approach, confirming the reliability of the generated scores and feedback.

Po-Han Lee, Yu-Cheng Lin, Chan-Tung Ku et al.

Estimating individualized treatment effects from observational data presents a persistent challenge due to unmeasured confounding and structural bias. Causal Machine Learning (causal ML) methods, such as causal trees and doubly robust estimators, provide tools for estimating conditional average treatment effects. These methods have limited effectiveness in complex real-world environments due to the presence of latent confounders or those described in unstructured formats. Moreover, reliance on domain experts for confounder identification and rule interpretation introduces high annotation cost and scalability concerns. In this work, we proposed Large Language Model-based agents for automated confounder discovery and subgroup analysis that integrate agents into the causal ML pipeline to simulate domain expertise. Our framework systematically performs subgroup identification and confounding structure discovery by leveraging the reasoning capabilities of LLM-based agents, which reduces human dependency while preserving interpretability. Experiments on real-world medical datasets show that our proposed approach enhances treatment effect estimation robustness by narrowing confidence intervals and uncovering unrecognized confounding biases. Our findings suggest that LLM-based agents offer a promising path toward scalable, trustworthy, and semantically aware causal inference.