# Chunran Zhang

Source: https://ln-one.github.io/about/

Studying computer science, with interests in cognition and philosophy.

I read Dostoevsky, occasionally. Most inspired by Gödel, Escher, Bach — two chapters in.

## Education

Southwest Jiaotong University

Bachelor’s degree; Sep 2023 - Jul 2027 (expected completion: July 2027).

School of Computing and Artificial Intelligence. Chengdu, China.

## Research interests

Artificial Intelligence; Information Retrieval; Data Mining

## Contact and profiles

- Email: chronis@my.swjtu.edu.cn

- google scholar: https://scholar.google.com/citations?user=nxuJYO0AAAAJ&hl=en

- github: https://github.com/ln-one

- orcid: https://orcid.org/0009-0005-8865-2090

- researchgate: https://www.researchgate.net/profile/Chunran-Zhang

- [CV (PDF)](https://ln-one.github.io/papers/Chunran_Zhang_CV.pdf)

## Research works

Complete list; the homepage displays at most four selected works. Status is recorded per work. Source: https://ln-one.github.io/publications/

### Opacity Is Not Just Opacity

Authors: Chunran Zhang

Year: 2026

Status: arXiv preprint

Extending alpha compositing to real-valued coefficients so fixed-source web graphics adapt contrast to their backgrounds, with measured browser behavior and clipping trade-offs.

- [Website record](https://ln-one.github.io/publications/#pub-zhang2026opacity)
- [arXiv](https://arxiv.org/abs/2609.14971)
- [DOI](https://doi.org/10.48550/arXiv.2609.14971)
- [Code](https://github.com/ln-one/opacity-is-not-just-opacity)

### Top-K Is Not a Budget for Hybrid Retrieval

Authors: Chunran Zhang

Year: 2026

Status: arXiv preprint

Certifying an exact prefix of the full-list RRF ranking within a specified access budget.

- [Website record](https://ln-one.github.io/publications/#pub-zhang2026dibud)
- [arXiv](https://arxiv.org/abs/2609.15143)
- [DOI](https://doi.org/10.48550/arXiv.2609.15143)
- [Code](https://github.com/ln-one/top-k-is-not-a-budget)

### Query Expansion Should Be Coordinated: Dense Expands, Sparse Anchors

Authors: Chunran Zhang

Year: 2026

Status: arXiv preprint

Coordinating generated references across dense and sparse channels: semantic expansion for dense retrieval and anchoring for sparse retrieval.

- [Website record](https://ln-one.github.io/publications/#pub-zhang2026desa)
- [arXiv](https://arxiv.org/abs/2608.15851)
- [DOI](https://doi.org/10.48550/arXiv.2608.15851)
- [Code](https://github.com/ln-one/dense-expands-sparse-anchors)

### Exact Adaptive Hybrid Retrieval Without Fixed Top-L Cutoffs

Authors: Chunran Zhang

Year: 2026

Status: arXiv preprint

Recovering exact complete-list fused rankings with request-specific retrieval depth instead of a fixed top-L cutoff.

- [Website record](https://ln-one.github.io/publications/#pub-zhang2026eahr)
- [arXiv](https://arxiv.org/abs/2608.07152)
- [DOI](https://doi.org/10.48550/arXiv.2608.07152)
- [Code](https://github.com/ln-one/exact-adaptive-hybrid-retrieval)

## News

[Full news archive](https://ln-one.github.io/allnews.html)