publications
Publications in reverse chronological order. Equal-contribution information is shown below the relevant entries.
2026
- LLM, Reliable AI
A Latent-Load Framework for Reliability Analysis and Intervention Design in LLM PipelinesIn NeurIPS, 2026 - FL, CP
Valid and Efficient Uncertainty Quantification for Federated Joint ShiftIn UAI, 2026 - LLM, CP
Inference-Time Conformal Reasoning with Valid Factuality Control for Large Language ModelsIn ICML, 2026Ting Wang and Yuanjie Shi contributed equally. - LLM, Alignment
Keep the Best, Forget the Rest: Reliable Alignment with Order-Aware Preference OptimizationIn ICLR, 2026Jiahui Zhu and Yuanjie Shi contributed equally. - Robust ML, CP
Conformal Margin Risk Minimization: An Envelope Framework for Robust Learning under Label NoiseIn AISTATS, 2026Yuanjie Shi and Peihong Li contributed equally. - ML, CP
Cost-Sensitive Conformal Training with Provably Controllable Learning BoundsIn AAAI, 2026Xuesong Jia and Yuanjie Shi contributed equally. - CP
Minimum-Length Conformal Prediction Sets for Ordinal ClassificationIn AAAI, 2026 - CPConformal Classification on Conformity Spectrum: A Probabilistic Geometric FrameworkUnder review, 2026
2025
- ML, CP
Direct Prediction Set Minimization via Bilevel Conformal Classifier TrainingIn ICML, 2025Yuanjie Shi and Hooman Shahrokhi contributed equally. - FL, FairnessFederated Rényi Fair Inference in Federated Heterogeneous SystemIn UAI, 2025Zhiyong Ma and Yuanjie Shi contributed equally.
- FL, NLPFedSum: Data-Efficient Federated Learning under Data Scarcity Scenario for Text SummarizationIn AAAI, 2025
- CP, Reliable AIReliable Uncertainty Quantification in Machine Learning via Conformal PredictionIn AAAI/SIGAI Doctoral Consortium, 2025
2024
- CPConformal Prediction for Class-wise Coverage via Augmented Label Rank CalibrationIn NeurIPS, 2024
2023
- Robust ML, CPProbabilistically Robust Conformal PredictionIn UAI, 2023Subhankar Ghosh and Yuanjie Shi contributed equally.