Papers by Michael Lu
Modality Matching Matters: Calibrating Language Distances for Cross-Lingual Transfer in URIEL+ (2026.eacl-srw)
Copied to clipboard
York Hay Ng, Aditya Khan, Xiang Lu, Matteo Salloum, Michael Zhou, Phuong Hanh Hoang, A. Seza Doğruöz, En-Shiun Annie Lee
| Challenge: | Existing linguistic knowledge bases such as URIEL+ lack a principled method for aggregating these signals into a single, comprehensive score. |
| Approach: | They propose a framework for type-matched language distances that unifies these signals into a robust, task-agnostic composite distance. |
| Outcome: | The proposed representations improve transfer performance when the distance type is relevant to the task, while yielding gains in most tasks. |
Not All Errors are Equal: Learning Text Generation Metrics using Stratified Error Synthesis (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing learning metrics are limited to tasks where large human ratings are available. |
| Approach: | They propose a model-based natural language generation (NLG) evaluation metric that is highly correlated with human judgements without requiring human annotation. |
| Outcome: | The proposed metric outperforms all prior unsupervised metrics on multiple NLG tasks including translation, image captioning, and WebNLG text generation. |
Let’s Think Frame by Frame with VIP: A Video Infilling and Prediction Dataset for Evaluating Video Chain-of-Thought (2023.emnlp-main)
Copied to clipboard
Vaishnavi Himakunthala, Andy Ouyang, Daniel Rose, Ryan He, Alex Mei, Yujie Lu, Chinmay Sonar, Michael Saxon, William Wang
| Challenge: | Existing studies show vision-language systems can reason about images using natural language, but their capacity for video reasoning remains underexplored. |
| Approach: | They propose to frame video reasoning as the sequential understanding of a small number of keyframes, thereby leveraging the power and robustness of vision-language systems' capacity to reason about images using natural language. |
| Outcome: | The proposed models can generate multiple intermediate keyframes and predict future keyframe, and they perform poorly on GPT-4, GPT-3, and VICUNA. |
i-Code V2: An Autoregressive Generation Framework over Vision, Language, and Speech Data (2024.findings-naacl)
Copied to clipboard
Ziyi Yang, Mahmoud Khademi, Yichong Xu, Reid Pryzant, Yuwei Fang, Chenguang Zhu, Dongdong Chen, Yao Qian, Xuemei Gao, Yi-Ling Chen, Robert Gmyr, Naoyuki Kanda, Noel Codella, Bin Xiao, Yu Shi, Lu Yuan, Takuya Yoshioka, Michael Zeng, Xuedong Huang
| Challenge: | i-Code V2 is one of the first models capable of generating natural language from any combination of Vision, Language, and Speech data. |
| Approach: | They propose to create a model that can generate natural language from any combination of Vision, Language, and Speech data. |
| Outcome: | i-Code V2 matches or outperforms state-of-the-art single- and dual-modality baselines on 7 multimodal tasks. |
NewsInterview: a Dataset and a Playground to Evaluate LLMs’ Grounding Gap via Informational Interviews (2025.acl-long)
Copied to clipboard
Alexander Spangher, Michael Lu, Sriya Kalyan, Hyundong Justin Cho, Tenghao Huang, Weiyan Shi, Jonathan May
| Challenge: | Existing large datasets (1k-10k transcripts) are generated via crowdsourcing and are inherently unnatural. |
| Approach: | They curate a dataset of 40,000 two-person informational interviews from NPR and CNN . they find that LLMs are significantly less likely than human interviewers to use acknowledgements and pivot to higher-level questions. |
| Outcome: | The proposed model is based on 40,000 interviews with journalists and CNN . |
BinaryBERT: Pushing the Limit of BERT Quantization (2021.acl-long)
Copied to clipboard
| Challenge: | Recent pre-trained language models have achieved remarkable performance improvement in various tasks, but the improvement generally comes at the cost of increasing model size and computation. |
| Approach: | They propose a binary quantization technique which initializes binaryBERT by splitting from a ternary network. |
| Outcome: | The proposed model achieves state-of-the-art performance on the GLUE and SQUAD benchmarks while being 24x smaller. |
TVWorld: Foundations for Remote-Control TV Agents (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing work on large vision–language models focuses on point-and-click interaction, while remote-control interaction is underexplored. |
| Approach: | They propose a topology-aware training framework that injects topology awareness into LVLMs. |
| Outcome: | The proposed model achieves 68.3% success rate on TVWorld-N, surpassing closed-source benchmarks and state-of-the-art (SOTA) benchmarks show that existing agents lack topology awareness for focus-based, long-horizon TV navigation. |
Modeling the Relationship between User Comments and Edits in Document Revision (D19-1)
Copied to clipboard
| Challenge: | Managing collaborative documents can be difficult due to the profusion of edits and comments that multiple authors make during a document’s evolution. |
| Approach: | They propose a hierarchical multi-layer deep neural network to model the relationship between edits and comments by encoding specific edit actions such as additions and deletions while accounting for document context. |
| Outcome: | The proposed model outperforms baselines in a number of evaluation settings and achieves a precision@1 of 71.0% and precision@3 of 94.4% for Comment Ranking while achieving 74.4% accuracy on Edit Anchoring. |
Adapting LLM Agents with Universal Communication Feedback (2025.findings-naacl)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) have demonstrated potential for LLM agents. |
| Approach: | They propose a universal buffer and iterative pipeline to store feedback and itersative pipelines to enable LLM agents to explore and update their policy in an environment. |
| Outcome: | The proposed approach outperforms supervised instruction fine-tuning baselines on four datasets. |
NovelHopQA: Diagnosing Multi-Hop Reasoning Failures in Long Narrative Contexts (2025.emnlp-main)
Copied to clipboard
| Challenge: | Current large language models struggle to answer questions that span tens of thousands of tokens. |
| Approach: | They evaluate 1–4 hop QA over 64k–128k-token excerpts from 83 novels . they find consistent accuracy drops with increased hops and context length . |
| Outcome: | The novelhopqa benchmark evaluates 1–4 hop QA over 64k–128k-token excerpts from 83 public-domain novels. |
Rosetta-PL: Propositional Logic as a Benchmark for Large Language Model Reasoning (2025.naacl-srw)
Copied to clipboard
Shaun Lee Baek, Shaun Esua-Mensah, Cyrus Tsui, Sejan Vigneswaralingam, Abdullah Alali, Michael Lu, Vasu Sharma, Kevin Zhu
| Challenge: | Large Language Models (LLMs) are primarily trained on high-resource natural languages, limiting their effectiveness in low-resourced settings and in tasks requiring deep logical reasoning. |
| Approach: | They propose to use a dataset of logical propositions from Lean into a custom logical language to evaluate LLMs' logical reasoning and generalization capabilities in a controlled environment. |
| Outcome: | The proposed model improves accuracy and accuracy beyond 20,000 training samples. |
JobFair: A Framework for Benchmarking Gender Hiring Bias in Large Language Models (2024.findings-emnlp)
Copied to clipboard
Ze Wang, Zekun Wu, Xin Guan, Michael Thaler, Adriano Koshiyama, Skylar Lu, Sachin Beepath, Ediz Ertekin, Maria Perez-Ortiz
| Challenge: | a framework for benchmarking hierarchical gender hiring bias in Large Language Models (LLMs) is developed to protect vulnerable demographic groups. |
| Approach: | They propose a framework for benchmarking hierarchical gender hiring bias in Large Language Models for resume scoring. |
| Outcome: | The proposed framework reveals significant issues of reverse gender hiring bias and overdebiasing in ten state-of-the-art LLMs. |
i-Code Studio: A Configurable and Composable Framework for Integrative AI (2024.emnlp-demo)
Copied to clipboard
Yuwei Fang, Mahmoud Khademi, Chenguang Zhu, Ziyi Yang, Reid Pryzant, Yichong Xu, Yao Qian, Takuya Yoshioka, Lu Yuan, Michael Zeng, Xuedong Huang
| Challenge: | Existing frameworks for Integrative AI lack flexibility and composability to handle multimodal tasks. |
| Approach: | They propose a configurable framework for Integrative AI that orchestrates multiple pre-trained models to conduct complex multimodal tasks. |
| Outcome: | The proposed framework achieves impressive results on zero-shot multimodal tasks . it can communicate and personalize for users, and it can be used in a multimodal agent . |
Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Speculative decoding uses a small draft model to generate a single input token, instead of sequentially generating tokens until completion. |
| Approach: | They propose a method that generates varying draft models adapted to the input context using simple rules. |
| Outcome: | The proposed method is competitive with the current SOTA for self-speculative decoding while being a truly plug-and-play method. |