Papers by Seungyoon Lee
Semantic Aware Linear Transfer by Recycling Pre-trained Language Models for Cross-lingual Transfer (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly incorporating multilingual capabilities, fueling the demand to transfer them into target language-specific models. |
| Approach: | They propose a novel cross-lingual transfer technique that recycles embeddings from target language Pre-trained Language Models to transmit deep representational strengths to LLMs. |
| Outcome: | The proposed technique outperforms existing methods in cross-lingual understanding setups and achieves faster convergence and lower loss during language adaptation. |
Length-aware Byte Pair Encoding for Mitigating Over-segmentation in Korean Machine Translation (2024.findings-acl)
Copied to clipboard
Jungseob Lee, Hyeonseok Moon, Seungjun Lee, Chanjun Park, Sugyeong Eo, Hyunwoong Ko, Jaehyung Seo, Seungyoon Lee, Heuiseok Lim
| Challenge: | Byte Pair Encoding (BPE) is an effective approach in machine translation across several languages, but it is prone to over-segmentation in Korean, an agglutinative and morphologically rich language. |
| Approach: | They propose a new method that incorporates long words into the Korean vocabulary by strategically preserving morphological information and reducing semantic confusion. |
| Outcome: | The proposed method outperforms BPE and surpasses state-of-the-art morpheme-aware tokenization methods. |
Exploring Inherent Biases in LLMs within Korean Social Context: A Comparative Analysis of ChatGPT and GPT-4 (2024.naacl-srw)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have been criticized for perpetuating stereotypes against diverse groups based on race, sexual orientation, and other attributes. |
| Approach: | They devised a set of prompts that reflect major societal issues in Korea and assign varied personas to both ChatGPT and GPT-4 to assess the toxicity of the generated sentences. |
| Outcome: | The proposed model produces twice the level of toxic content as ChatGPT and GPT-4 under certain conditions. |
FLEX: A Benchmark for Evaluating Robustness of Fairness in Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing safety evaluations may overlook the inherent weaknesses of Large Language Models, despite their benefits. |
| Approach: | They propose a benchmark to evaluate the robustness of Large Language Models under extreme conditions. |
| Outcome: | The proposed approach evaluates the fairness of large language models under extreme conditions. |
Find the Intention of Instruction: Comprehensive Evaluation of Instruction Understanding for Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | LLMs are prone to generate responses to instruction-formatted statements in an instinctive manner, rather than comprehending the underlying user intention within the given instructions. |
| Approach: | They propose to use an instruction-following capability benchmark to evaluate LLMs' instruction understanding capability. |
| Outcome: | The proposed benchmark analyzes the instruction understanding capability of large language models with four instruction candidates and a single candidate. |
Translation of Multifaceted Data without Re-Training of Machine Translation Systems (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a novel MT pipeline that considers the intra-data relation is proposed . previous MT systems have demonstrated relatively low performance, making them hardly utilized as another data source. |
| Approach: | They propose a new MT pipeline that considers the intra-data relation . they propose CS and IT to enhance the intra data relation based on a data point . |
| Outcome: | The proposed pipeline improves translation quality and training data compared with the existing approach . it yields better training data and better translation quality than previous approaches . |
CLEAR: Cross-Lingual Enhancement in Retrieval via Reverse-training (2026.acl-long)
Copied to clipboard
| Challenge: | Existing multilingual embedding models often struggle to capture cross-lingual alignment during training. |
| Approach: | They propose a novel loss function that leverages an English passage as a bridge to strengthen alignments between target language and English. |
| Outcome: | The proposed model improves retrieval performance across cross-lingual scenarios while minimizing performance degradation in English. |
Leveraging Pre-existing Resources for Data-Efficient Counter-Narrative Generation in Korean (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing datasets and methods for detecting hate speech are limited by resource-intensive nature and only focus on the primary language. |
| Approach: | They propose a Korean Hate Speech Counter Punch (KHSCP) method that generates fact-based responses to hate speech in the Korean language and propose to use existing resources to overcome data scarcity. |
| Outcome: | The proposed method can overcome data scarcity in low-resource environments by leveraging existing resources. |
TORSO: Template-Oriented Reasoning Towards General Tasks (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to generate responses using few-shot examples depend on the provided examples, limiting the model’s reasoning capabilities. |
| Approach: | They propose a model that emulates human reasoning during response generation by using curated few-shot prompts instead of manually crafted few-shot examples. |
| Outcome: | The proposed model achieves strong performance on diverse LLMs benchmarks with reasonable rationales. |
MIGRATE: Cross-Lingual Adaptation of Domain-Specific LLMs through Code-Switching and Embedding Transfer (2025.coling-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have advanced in many fields, but focus on English-centric models requires extensive data. |
| Approach: | They propose a method that leverages open-source static embedding models and up to 3 million tokens of code-switching data to facilitate the seamless transfer of embeddables to target languages. |
| Outcome: | The proposed method outperforms baseline and existing cross-lingual transfer methods in target languages. |