Papers by Hansaem Kim
AI Knows Where You Are: Exposure, Bias, and Inference in Multimodal Geolocation with KoreaGEO (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing benchmarks show coarse granularity, linguistic bias, and a neglect of multimodal privacy risks. |
| Approach: | They propose a benchmark for visual-language models that analyzes social photos to assess location privacy risks. |
| Outcome: | The proposed benchmarks show coarse granularity, linguistic bias, and neglect of privacy risks. |
Read the Room, Read the Image: Understanding Indirect Speech Acts in Multimodal Visual Contexts (2026.findings-acl)
Copied to clipboard
Jaehee Kim, Ji Hoon Chung, Seoyoon Park, Unsol Kim, Kyungwon Park, JiHak Kim, Yi-Jun Chen, Hansaem Kim
| Challenge: | Existing benchmarks focus on explicit context, but do not address context-dependent pragmatic understanding. |
| Approach: | They propose a benchmark for evaluating ISA understanding through integrated reasoning over visual context and dialogue. |
| Outcome: | Experiments show that state-of-the-art models struggle with visually grounded indirect speech acts . linguistic meaning emerges through the relationship between an utterance and situational context . |
Can LLMs Truly Plan? A Comprehensive Evaluation of Planning Capabilities (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing assessments of planning capabilities of large language models are limited to single-language or specific representation formats. |
| Approach: | a new benchmark is developed to assess the planning capabilities of large language models. |
| Outcome: | The Multi-Plan benchmark highlights performance disparities among models . language differences showed minimal impact, while mathematically structured representations improved accuracy . |
Subject-level Inference for Realistic Text Anonymization Evaluation (2026.acl-long)
Copied to clipboard
Myeong Seok Oh, Dong-Yun Kim, Hanseok Oh, Chaean Kang, Joeun Kang, Xiaonan Wang, Hyunjung Park, Young Cheol Jung, Hansaem Kim
| Challenge: | Existing text anonymization evaluations assume only a single data subject, ignoring multi-subject scenarios. |
| Approach: | They propose a benchmark that shifts the unit of evaluation from text spans to individuals . they show that subject-level inference protection drops as low as 33% when masked . |
| Outcome: | The proposed benchmark reduces the amount of protection available when PII spans are masked. |
FLUID QA: A Multilingual Benchmark for Figurative Language Usage in Dialogue across English, Chinese, and Korean (2025.emnlp-main)
Copied to clipboard
| Challenge: | Figurative language is a core component of everyday communication . existing benchmarks focus on sentence-level classification or inference tasks . |
| Approach: | They propose a multilingual benchmark that evaluates figurative usage in dialogue . they use a sentence-level diagnostic task to embed figurativ choices into multi-turn contexts . |
| Outcome: | The benchmark evaluates large language models' ability to use figurative expressions coherently in conversation. |
Crowdsourcing in the Development of a Multilingual FrameNet: A Case Study of Korean FrameNet (2020.lrec-1)
Copied to clipboard
| Challenge: | Using current methods, the construction of multilingual FrameNets is expensive and complex. |
| Approach: | They evaluated whether crowdsourcing approaches captured cross-cultural and cross-linguistic meanings . they found that crowd workers made intuitive choices comparable to trained FrameNet experts . |
| Outcome: | The results are now available in Korean FrameNet 1.1. |
Optimizing Language Augmentation for Multilingual Large Language Models: A Case Study on Korean (2024.lrec-main)
Copied to clipboard
ChangSu Choi, Yongbin Jeong, Seoyoon Park, Inho Won, HyeonSeok Lim, SangMin Kim, Yejee Kang, Chanhyuk Yoon, Jaewan Park, Yiseul Lee, HyeJin Lee, Younggyun Hahm, Hansaem Kim, KyungTae Lim
| Challenge: | Large language models (LLMs) use pretraining to predict the subsequent word, but less-resourced languages are being overlooked. |
| Approach: | They propose to expand the MLLM vocabularies to enhance expressiveness and use bilingual data for pretraining to align the high- and less-resourced languages. |
| Outcome: | The proposed model outperforms existing models in qualitative analyses compared to Korean monolingual models. |