9 papers
Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval
Linhai Ma, Ethan F. Wei, Xueqing Peng +3
Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose me…
PVminerLLM2: Improving Structured Extraction of Patient Voice via Preference Optimization
Samah Fodeh, Linhai Ma, Ganesh Puthiaraju +7
Motivation: Patient-generated text contains critical information on patients' lived experiences, social context, and care engagement, but remains largely unstructured, limiting its…
STaR-DRO: Stateful Tsallis Reweighting for Group-Robust Structured Prediction
Samah Fodeh, Ganesh Puthiaraju, Elyas Irankhah +5
Structured prediction with large language models requires outputs that are label-accurate, ontology-constrained, structurally valid, and evidence-grounded under label imbalance and…
TAB-PO: Preference Optimization with a Token-Level Adaptive Barrier for Token-Critical Structured Generation
Samah Fodeh, Linhai Ma, Ganesh Puthiaraju +8
Direct Preference Optimization (DPO) is effective for offline alignment but poorly matched to ontology-driven structured prediction, where preferred and rejected JSON often differ…
Herculean: An Agentic Benchmark for Financial Intelligence
Xueqing Peng, Zhuohan Xie, Yupeng Cao +60
As AI agents improve, the central question is no longer whether they can solve isolated well-defined financial tasks, but whether they can reliably carry out financial professional…
EPPC-OASIS: Ontology-Aware Adaptation and Structured Inference Refinement for Electronic Patient-Provider Communication Mining in Secure Messages
Samah Fodeh, Sreeraj Ramachandran, Elyas Irankhah +7
Secure patient-provider messages contain clinically important communication behaviors that are difficult to characterize manually at scale. The Electronic Patient-Provider Communic…