arXiv subjects

Yangmin Huang

Publications and source records attributed to Yangmin Huang.

5 recordsLinked to original sources

arXiv2026

Dense Clinical Contrasts Enhance Medical Knowledge Updating in Large Language Models

Medical knowledge changes continually, making large language models vulnerable to relying on outdated yet clinically plausible information. We study whether the format of supervision affects medical knowledge updating under a matched training-budget setting. We introduce SEER-Bench, a temporally anchored oncology-staging benchmark curated from the latest versioned SEER Research Data release, and render identical medical update events from NCCN oncology guidelines into four supervision formats: EMQ, MSQ, FITB, and SAQ. Across SEER-Bench and HealthBench Professional, EMQ gives the most stable external transfer and retention among same-budget SFT variants. With EMQ supervision, the updated 4B model produces competitive results on temporally anchored oncology staging, reaching 64.8% answer accuracy and 59.6% rationale accuracy on SEER-Bench. Diagnostic analyses suggest that EMQ exposes denser clinical contrast signals while preserving discriminative representations with smaller movement from the base model. These results show that medical knowledge updating depends not only on the update algorithm, but also on how knowledge is structured as supervision.

cs.AI↗

arXiv2026

ProMedical: Hierarchical Fine-Grained Criteria Modeling for Medical LLM Alignment via Explicit Injection

Aligning Large Language Models (LLMs) with high-stakes medical standards remains a significant challenge, primarily due to the dissonance between coarse-grained preference signals and the complex, multi-dimensional nature of clinical protocols. To bridge this gap, we introduce ProMedical, a unified alignment framework grounded in fine-grained clinical criteria. We first construct ProMedical-Preference-50k, a dataset generated via a human-in-the-loop pipeline that augments medical instructions with rigorous, physician-derived rubrics. Leveraging this corpus, we propose the Explicit Criteria Injection paradigm to train a multi-dimensional reward model. Unlike traditional scalar reward models, our approach explicitly disentangles safety constraints from general proficiency, enabling precise guidance during reinforcement learning. To rigorously validate this framework, we establish ProMedical-Bench, a held-out evaluation suite anchored by double-blind expert adjudication. Empirical evaluations demonstrate that optimizing the Qwen3-8B base model via ProMedical-RM-guided GRPO yields substantial gains, improving overall accuracy by 22.3% and safety compliance by 21.7%, effectively rivaling proprietary frontier models. Furthermore, the aligned policy generalizes robustly to external benchmarks, demonstrating performance comparable to state-of-the-art models on UltraMedical. We publicly release our datasets, reward models, and benchmarks to facilitate reproducible research in safety-aware medical alignment.

cs.AI↗

arXiv2025

MedFact: Benchmarking the Fact-Checking Capabilities of Large Language Models on Chinese Medical Texts

Deploying Large Language Models (LLMs) in medical applications requires fact-checking capabilities to ensure patient safety and regulatory compliance. We introduce MedFact, a challenging Chinese medical fact-checking benchmark with 2,116 expert-annotated instances from diverse real-world texts, spanning 13 specialties, 8 error types, 4 writing styles, and 5 difficulty levels. Construction uses a hybrid AI-human framework where iterative expert feedback refines AI-driven, multi-criteria filtering to ensure high quality and difficulty. We evaluate 20 leading LLMs on veracity classification and error localization, and results show models often determine if text contains errors but struggle to localize them precisely, with top performers falling short of human performance. Our analysis reveals the "over-criticism" phenomenon, a tendency for models to misidentify correct information as erroneous, which can be exacerbated by advanced reasoning techniques such as multi-agent collaboration and inference-time scaling. MedFact highlights the challenges of deploying medical LLMs and provides resources to develop factually reliable medical AI systems.

cs.CL↗

arXiv2023

Transformer and Snowball Graph Convolution Learning for Brain functional network Classification

Advanced deep learning methods, especially graph neural networks (GNNs), are increasingly expected to learn from brain functional network data and predict brain disorders. In this paper, we proposed a novel Transformer and snowball encoding networks (TSEN) for brain functional network classification, which introduced Transformer architecture with graph snowball connection into GNNs for learning whole-graph representation. TSEN combined graph snowball connection with graph Transformer by snowball encoding layers, which enhanced the power to capture multi-scale information and global patterns of brain functional networks. TSEN also introduced snowball graph convolution as position embedding in Transformer structure, which was a simple yet effective method for capturing local patterns naturally. We evaluated the proposed model by two large-scale brain functional network datasets from autism spectrum disorder and major depressive disorder respectively, and the results demonstrated that TSEN outperformed the state-of-the-art GNN models and the graph-transformer based GNN models.

cs.LG↗

arXiv2023

BrainNPT: Pre-training of Transformer networks for brain network classification

Deep learning methods have advanced quickly in brain imaging analysis over the past few years, but they are usually restricted by the limited labeled data. Pre-trained model on unlabeled data has presented promising improvement in feature learning in many domains, including natural language processing and computer vision. However, this technique is under-explored in brain network analysis. In this paper, we focused on pre-training methods with Transformer networks to leverage existing unlabeled data for brain functional network classification. First, we proposed a Transformer-based neural network, named as BrainNPT, for brain functional network classification. The proposed method leveraged token as a classification embedding vector for the Transformer model to effectively capture the representation of brain network. Second, we proposed a pre-training framework for BrainNPT model to leverage unlabeled brain network data to learn the structure information of brain networks. The results of classification experiments demonstrated the BrainNPT model without pre-training achieved the best performance with the state-of-the-art models, and the BrainNPT model with pre-training strongly outperformed the state-of-the-art models. The pre-training BrainNPT model improved 8.75% of accuracy compared with the model without pre-training. We further compared the pre-training strategies, analyzed the influence of the parameters of the model, and interpreted the trained model.

q-bio.NC↗