Papers by Steven Wu
Multimodal Dual-Path Decoding for Medical Report Generation (2026.findings-acl)
Copied to clipboard
| Challenge: | Current methods for radiology report generation rely on encoder-decoder based frameworks that fail to integrate multimodal clinical evidence with domain-specific knowledge. |
| Approach: | They propose a multimodal dual-path framework that synergistically integrates large vision-language models and large language models for radiology report generation. |
| Outcome: | The proposed framework improves on the public MIMIC-CXR benchmark and shows that it is superior to state-of-the-art models. |
Persona-Augmented Benchmarking: Evaluating LLMs Across Diverse Writing Styles (2025.emnlp-main)
Copied to clipboard
| Challenge: | Current benchmarks for evaluating Large Language Models do not capture the rich variety of communication patterns exhibited by humans. |
| Approach: | They propose a low-cost method to emulate diverse writing styles by rewriting evaluation prompts using persona-based LLM prompting. |
| Outcome: | The proposed method improves the external validity of the benchmarks for Large Language Models (LLMs) based on persona-based prompting. |
Predicting Language Models’ Success at Zero-Shot Probabilistic Prediction (2025.findings-emnlp)
Copied to clipboard
Kevin Ren, Santiago Cortes-Gomez, Carlos Miguel Patiño, Ananya Joshi, Ruiqi Lyu, Jingjing Tang, Alistair Turcan, Khurram Yamin, Steven Wu, Bryan Wilder
| Challenge: | Recent work has investigated the capabilities of large language models (LLMs) as zero-shot models for generating individual-level characteristics. |
| Approach: | They conduct a large-scale empirical study of large language models’ zero-shot predictive capabilities across a wide range of tabular prediction tasks. |
| Outcome: | The results show that LLMs perform well on the base prediction task, and when they perform well, they are more likely to provide high-quality predictions. |
Academics Can Contribute to Domain-Specialized Language Models (2024.emnlp-main)
Copied to clipboard
Mark Dredze, Genta Winata, Prabhanjan Kambadur, Shijie Wu, Ozan Irsoy, Steven Lu, Vadim Dabravolski, David Rosenberg, Sebastian Gehrmann
| Challenge: | Commercially available models dominate academic leaderboards, focusing on creating and adapting general-purpose models . however, general- purpose models often underperform in specialized domains, and domain-specific models yield superior results. |
| Approach: | They advocate for a renewed focus on developing and evaluating domain- and task-specific models . they advocate for an adapted or adapted model that can be used to improve academic leaderboard standings . |
| Outcome: | The proposed model can do well on professional and linguistic examinations, college-level knowledge questions, and collections of reasoning tasks. |
MixCE: Training Autoregressive Language Models by Mixing Forward and Reverse Cross-Entropies (2023.acl-long)
Copied to clipboard
| Challenge: | Autoregressive language models are trained by minimizing the cross-entropy of the model distribution Q relative to the data distribution P. However, these systems still struggle in many openended generation settings, where they are asked to produce a long text following a short prompt. |
| Approach: | They propose to combine forward and reverse cross-entropy to train autoregressive language models by minimizing the cross-Entropy of the model distribution Q relative to the data distribution P. |
| Outcome: | The proposed model overgeneralizes and produces non-human-like text without complex decoding strategies. |
Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning (2026.acl-long)
Copied to clipboard
| Challenge: | Current reasoning large language models (RLMs) are trained on data that is primarily in English, resulting in lower performance when asked the same question in a non-English language. |
| Approach: | They propose a framework for enhancing multilingual reasoning without any data in the target language(s). |
| Outcome: | The proposed framework outperforms Qwen2.5-7B-Instruct on 4 math and non-math tasks with less than 1/8 of the training data (125). |