Papers by Austen Liao
EnDive: A Cross-Dialect Benchmark for Fairness and Performance in Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing benchmarks often overlook intra-language variations, leaving speakers of non-standard dialects underserved. |
| Approach: | EnDive evaluates seven state-of-the-art large language models across tasks . human evaluations confirm high translation quality, with average scores of at least 6.02/7 . |
| Outcome: | EnDive evaluates state-of-the-art large language models across language understanding, reasoning, mathematics, logic tasks. |
TDFlow: Agentic Workflows for Test Driven Development (2026.eacl-long)
Copied to clipboard
| Challenge: | TDFlow decomposes software engineering program repair into four components governed by sub-agents. |
| Approach: | They propose a test-driven agentic workflow that frames repository-scale software engineering as a testing task and decomposes it into four components governed by sub-agents. |
| Outcome: | The proposed workflow achieves 88.8% pass rate on SWE-Bench Lite and 94.3% on Swe-Bech Verified when provided with human-written tests. |