10 papers
Learning When to Translate for Multilingual Reasoning
Deokhyung Kang, Hyounghun Kim, Gary Geunbae Lee
Reasoning language models (RLMs) achieve strong performance on complex reasoning tasks, but still exhibit substantial multilingual reasoning gaps, largely due to language-understan…
Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models?
Deokhyung Kang, Seonjeong Hwang, Daehui Kim +2
Reasoning language models (RLMs) achieve strong performance on complex reasoning tasks, yet they still exhibit a multilingual reasoning gap, performing better in high-resource lang…
EnSToM: Enhancing Dialogue Systems with Entropy-Scaled Steering Vectors for Topic Maintenance
Heejae Suh, Yejin Jeon, Deokhyung Kang +3
Small large language models (sLLMs) offer the advantage of being lightweight and efficient, which makes them suitable for resource-constrained environments. However, sLLMs often st…
Self-Correcting Code Generation Using Small Language Models
Jeonghun Cho, Deokhyung Kang, Hyounghun Kim +1
Self-correction has demonstrated potential in code generation by allowing language models to revise and improve their outputs through successive refinement. Recent studies have exp…
MiLQ: Benchmarking IR Models for Bilingual Web Search with Mixed Language Queries
Jonghwi Kim, Deokhyung Kang, Seonjeong Hwang +3
Despite bilingual speakers frequently using mixed-language queries in web searches, Information Retrieval (IR) research on them remains scarce. To address this, we introduce MiLQ,…
GuRE:Generative Query REwriter for Legal Passage Retrieval
Daehee Kim, Deokhyung Kang, Jonghwi Kim +2
Legal Passage Retrieval (LPR) systems are crucial as they help practitioners save time when drafting legal arguments. However, it remains an underexplored avenue. One primary reaso…
Retrieval-Augmented Fine-Tuning With Preference Optimization For Visual Program Generation
Deokhyung Kang, Jeonghun Cho, Yejin Jeon +4
Visual programming languages (VPLs) allow users to create programs through graphical interfaces, which results in easier accessibility and their widespread usage in various domains…
Cross-lingual Back-Parsing: Utterance Synthesis from Meaning Representation for Zero-Resource Semantic Parsing
Deokhyung Kang, Seonjeong Hwang, Yunsu Kim +1
Recent efforts have aimed to utilize multilingual pretrained language models (mPLMs) to extend semantic parsing (SP) across multiple languages without requiring extensive annotatio…
Ontology-Free General-Domain Knowledge Graph-to-Text Generation Dataset Synthesis using Large Language Model
Daehee Kim, Deokhyung Kang, Sangwon Ryu +1
Knowledge Graph-to-Text (G2T) generation involves verbalizing structured knowledge graphs into natural language text. Recent advancements in Pretrained Language Models (PLMs) have…
Denoising Table-Text Retrieval for Open-Domain Question Answering
Deokhyung Kang, Baikjin Jung, Yunsu Kim +1
In table-text open-domain question answering, a retriever system retrieves relevant evidence from tables and text to answer questions. Previous studies in table-text open-domain qu…