Yuxin Liang

unsupervised learning (1) contrastive learning (1) reinforcement learning (1) question answering (1) information retrieval (1) knowledge probing (1) semantic similarity (1) language model (1) hallucination mitigation (1) sentence embedding (1) negative sampling (1) semantic text similarity (1) momentum contrast (1) large language model (1) long-context question answering (1) position-agnostic training (1) multi-document qa (1) reinforcement learning from knowledge feedback (1) negative sample queue (1)