Papers by Seanie Lee
Optimized Speculative Sampling for GPU Hardware Accelerators (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large foundational speech and language models require more memory and computational resources to generate long sequences. |
| Approach: | They propose to optimize speculative sampling for parallel hardware accelerators by combining multiple GPU threads to reduce profiling time. |
| Outcome: | The proposed approach improves profiling time from 6% to 13% without compromising accuracy. |
Effective and Efficient Conversation Retrieval for Dialogue State Tracking with Implicit Text Summaries (2024.naacl-long)
Copied to clipboard
| Challenge: | Recent studies use in-context learning with large language models (LLM) to find similar dialogue exemplars for prompt learning. |
| Approach: | They propose to use a conversation retriever to find similar in-context examples for prompt learning. |
| Outcome: | The proposed approach improves on multiWOZ datasets with GPT-Neo-2.7B and LLaMA-7B/30B . |
SafeRoute: Adaptive Model Selection for Efficient and Accurate Safety Guardrails in Large Language Models (2025.findings-acl)
Copied to clipboard
Seanie Lee, Dong Bok Lee, Dominik Wagner, Minki Kang, Haebin Seong, Tobias Bocklet, Juho Lee, Sung Ju Hwang
| Challenge: | Deploying large language models (LLMs) requires robust safety guard models to detect and block harmful user prompts. |
| Approach: | They propose a binary router that selectively applies a larger safety guard model to the data that the router considers hard. |
| Outcome: | The proposed method outperforms baselines on multiple benchmark datasets on hard and hard examples. |
Domain-agnostic Question-Answering with Adversarial Training (D19-58)
Copied to clipboard
| Challenge: | Adapting models to new domain without finetuning is a challenging problem in deep learning. |
| Approach: | They propose an adversarial training framework for domain generalization in Question Answering task using a conventional QA model and a discriminator. |
| Outcome: | The proposed model outperforms the baseline model on Question Answering (QA) task. |
Learning to Perturb Word Embeddings for Out-of-distribution QA (2021.acl-long)
Copied to clipboard
| Challenge: | QA models that are pretraining with unlabeled data can overfit and may not generalize well to unseen data that falls outside the training distribution. |
| Approach: | They propose a method which perturbs word embedding without changing their semantics. |
| Outcome: | The proposed method outperforms baseline methods on five target domains on a single source dataset on five different target domain domains. |
Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs (2020.acl-main)
Copied to clipboard
| Challenge: | Existing models for question answering are limited in the availability of labeled data. |
| Approach: | They propose a hierarchical conditional variational autoencoder for generating QA pairs given unstructured texts as contexts while maximizing mutual information between generated QA pair to ensure consistency. |
| Outcome: | The proposed framework achieves impressive performance gains over baseline models on both tasks, using only a fraction of data for training. |