R0027/2026-03-26/Q001/SRC03
Huang et al. — Cross-Lingual-Thought Prompting (XLT)
Source
| Field |
Value |
| Title |
Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting |
| Publisher |
ACL Anthology / EMNLP 2023 Findings |
| Author(s) |
Haoyang Huang, Tianyi Tang, Dongdong Zhang, Wayne Xin Zhao, Ting Song, Yan Xia, Furu Wei |
| Date |
2023-10 |
| URL |
https://aclanthology.org/2023.findings-emnlp.826/ |
| Type |
Research paper (peer-reviewed) |
Summary
| Dimension |
Rating |
| Reliability |
High |
| Relevance |
High |
| Bias: Missing data |
Low risk |
| Bias: Measurement |
Low risk |
| Bias: Selective reporting |
Low risk |
| Bias: Randomization |
N/A — not an RCT |
| Bias: Protocol deviation |
N/A — not an RCT |
| Bias: COI/Funding |
Some concerns |
Rationale
| Dimension |
Rationale |
| Reliability |
Peer-reviewed at top NLP venue (EMNLP). 7 benchmarks, multiple languages. Highly cited in subsequent work. |
| Relevance |
Directly demonstrates that languages are treated unequally and proposes a specific prompting remedy with quantified improvements. |
| Bias flags |
Some concerns about COI — Microsoft Research affiliation may favor approaches compatible with their models. |
| Evidence ID |
Summary |
| SRC03-E01 |
XLT achieves 10+ point average improvement, confirming language inequality in LLMs |