Scholar

Xinran Zhao

Google Scholar ID: iX71amEAAAAJ

PhD Student, Carnegie Mellon University

Natural Language ProcessingMachine Learning

Citations & Impact

All-time

Citations

2,382

H-index

i10-index

Publications

Co-authors

list available

Contact

Publications

1 items

2026

Cited

Resume (English only)

Academic Achievements

1. Improving Large Language Model Planning with Action Sequence Similarity (ICLR 2025).
2. SPHERE: An Evaluation Card for Human-AI Systems (Findings of ACL 2025).
3. cAST: Enhancing Code Retrieval-Augmented Generation with Structural Chunking via Abstract Syntax Tree (Findings of EMNLP 2025).
4. Strategic Planning and Rationalizing on Trees Make LLMs Better Debaters (In Submission).
5. MoR: Better Handling Diverse Queries with a Mixture of Sparse, Dense, and Human Retrievers (To appear in Proceedings of EMNLP 2025).
6. Revela: Dense Retriever Learning via Language Modeling (In Submission).
7. Pivot-ICL: Adaptive Exemplar Selection for In-Context Learning (In Submission).
8. The Ramon Llull's Thinking Machine for Automated Ideation (To appear in LM4Sci@COLM 2025).
9. Dense X Retrieval: What Retrieval Granularity Should We Use? (EMNLP 2024).
10. MixGR: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity (EMNLP 2024).
11. HiMemFormer: Hierarchical Memory-Aware Transformer for Multi-Agent Action Anticipation (Video-Language Models @ NeurIPS 2024).
12. Beyond Relevance: Evaluate and Improve Retrievers on Perspective Awareness (In Submission).

Research Experience

1. Spent a wonderful summer 2024 at Google DeepMind as a student researcher with Dr. Azade Nova and Dr. Hanie Sedghi.
2. Mentors also include Dr. Shikhar Murty, Dr. Hongming Zhang, and Dr. Esin Durmus.

Education

1. PhD student at CMU LTI, advised by Prof. Sherry Wu.
2. MSCS at Stanford University, worked with Prof. Christopher Manning.
3. Bachelors at HKUST, also exchanged at Cornell, worked with Prof. Yangqiu Song, Prof. Dit-Yan Yeung, and Prof. Claire Cardie.

Background

Research interests include: few-shot learning, commonsense reasoning, coreference resolution, knowledge graph, and argument mining. The goal is to build efficient, effective, and faithful systems to tackle the great ambiguity of natural languages.

Miscellany