News

우리 연구실의 논문 2편이 AACL-IJCNLP 2026에 게재 승인되었습니다. Main Conference에 1편, Findings에 1편이 각각 채택되었습니다.

Two papers from our lab have been accepted to AACL-IJCNLP 2026: one paper to the Main Conference and one paper to Findings.

Main Conference (1편) Main Conference (1 paper)

• DALSA: Difficulty-Aware Long-to-Short LLM Reasoning with Model Merging Dongwook Min, Jeonhwi Kang, Donghyeon Jeon, Inho Kang, Seung-Hoon Na

Findings (1편) Findings (1 paper)

• Probing Reasoning Length Control in Language Models via Neurons Dongwook Min, Jeonhwi Kang, Cheoneum Park, Seung-Hoon Na

우리 연구실의 논문 7편이 EMNLP 2026에 게재 승인되었습니다. Main Conference에 4편, Findings에 3편이 각각 채택되었습니다.

Seven papers from our lab have been accepted to EMNLP 2026: four papers to the Main Conference and three papers to Findings.

Main Conference (4편) Main Conference (4 papers)

• Signals from the Future: Future-aware Semi-Autoregressive Decoding for Diffusion Language Models Yangryeol Park, Jeonhwi Kang, Kunhui Lee, Hyeonju Kim, Jinyoung Choi, Seung-Hoon Na
• Do Language Models Understand Implicit Logical Meaning? A Case Study of Scope Ambiguity Resolution in Context Jieun Kim, Suhyeon Park, Kunhui Lee, Seung-Hoon Na
• Validator-Guided Decoding for Multi-Hop Knowledge Editing Shanbao Qiao, Xuebing Liu, Seung-Hoon Na
• DuoPRAG: Rank-Disentangled and Gated Knowledge-Reasoning Fusion Nakhyeon Choi, Cheoneum Park, Seung-Hoon Na

Findings (3편) Findings (3 papers)

• When Reflection Helps: Utility-Guided Advantage Rescaling for Efficient Reasoning Jungyu Lee, Kunhui Lee, Dongwook Min, Jeonhwi Kang, Seung-Hoon Na
• Doc2Weight: In-Weights Retrieval-Augmented Generation via Parameter Injection Gyusik Jang, Wonkee Lee, Seung-Hoon Na
• DuCEI: When to Introduce Metadata in Cross-Entity Interaction for Entity Disambiguation Jinwoo Min, Roseline Nyange, Wonkee Lee, Hyun-Je Song, Seung-Hoon Na

우리 연구실의 논문 1편이 CIKM 2026 Short Paper로 게재 승인되었습니다.

One paper from our lab has been accepted as a Short Paper at CIKM 2026.

Short Paper (1편) Short Paper (1 paper)

• Forward-Looking Guidance for Iterative RAG in Multi-Hop Question Answering Jumin Cho, Chanjun Park, Seung-Hoon Na, Hyun-Je Song

우리 연구실의 논문 3편이 ACL 2026 Findings에 게재 승인 되었습니다.

Three papers from our lab have been accepted to Findings of ACL 2026.

우리 연구실의 논문이 Expert Systems with Applications에 게재 승인 되었습니다.

Our lab's paper has been accepted to Expert Systems with Applications.

우리 연구실의 논문이 CVPR 2026에 게재 승인 되었습니다.

Our lab's paper has been accepted to CVPR 2026.

우리 연구실의 논문이 Transactions of the Association for Computational Linguistics (TACL)에 게재 승인 되었습니다.

Our lab's paper has been accepted to Transactions of the Association for Computational Linguistics (TACL).

우리 연구실의 논문이 ICLR 2026에 게재 승인 되었습니다.

Our lab's paper has been accepted to ICLR 2026.

우리 연구실의 "MECA: Modular Editing via Customized Expert Networks and Adaptors in Large Language Models" 논문이 Expert Systems with Applications에 게재 승인 되었습니다.

Our lab's paper "MECA: Modular Editing via Customized Expert Networks and Adaptors in Large Language Models" has been accepted to Expert Systems with Applications.

우리 연구실의 "GenPoE: Generative Passage-level Mixture of Experts for Knowledge Enhancement of LLMs" 논문이 EMNLP 2025 Findings에 게재 승인 되었습니다.

Our lab's paper "GenPoE: Generative Passage-level Mixture of Experts for Knowledge Enhancement of LLMs" has been accepted to Findings of EMNLP 2025.

우리 연구실이 과학기술정보통신부 "AI 최고급 신진연구자 지원(AI 스타펠로우십)"에 선정되었습니다.

Our lab has been selected for the Ministry of Science and ICT’s “AI Star Fellowship” program for top-tier early-career researchers in AI.

우리 연구실이 과학기술정보통신부 "생성AI 선도인재양성 사업"에 선정되었습니다.

Our lab has been selected for the Ministry of Science and ICT’s “Generative AI Leading Talent Training” program.

우리 연구실의 "SeqMMR: Sequential Model Merging and LLM Routing for Enhanced Batched Sequential Knowledge Editing" 논문이 ACL 2025 Findings에 포스터 발표되었습니다.

Our paper, "SeqMMR: Sequential Model Merging and LLM Routing for Enhanced Batched Sequential Knowledge Editing," was presented as a poster at ACL 2025 Findings.

우리 연구실이 재난안전연구원 "생성형 AI기반 안전 제도 진단 지원 시스템 개발" 과제에 주관기관으로 선정되었습니다.

Our lab has been selected as the National Disaster Management Research Institute’s project “Development of a Generative AI-based Safety Regulation Diagnostic Support System.”