Papers by Jiho Kim
A Korean Knowledge Extraction System for Enriching a KBox (C18-2)
Copied to clipboard
| Challenge: | Existing systems for knowledge extraction from natural language sentences are lacking for all languages. |
| Approach: | They propose a Korean knowledge extraction system and web interface for enriching a KBox knowledge base based on the Korean DBpedia. |
| Outcome: | The proposed system can extract factual knowledge from natural language sentences . the endpoint can be used to add knowledge to a KBox knowledge base anytime and anywhere . |
KG-GPT: A General Framework for Reasoning on Knowledge Graphs Using Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Using large language models for complex reasoning tasks on knowledge graphs remains unexplored. |
| Approach: | They propose a multi-purpose framework leveraging large language models for complex reasoning tasks on knowledge graphs. |
| Outcome: | The proposed framework outperforms fully-supervised models in KG-based fact verification and KGQA benchmarks. |
Speculative Verification: Exploiting Information Gain for Speculative Decoding (2026.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are used for many applications but their size and computational cost make inference serving a significant challenge. |
| Approach: | They propose an efficient augmentation to Speculative Decoding (SD) that predicts speculation accuracy and dynamically adapts the verification length to maximize throughput. |
| Outcome: | The proposed model reduces wasted verification on rejected tokens and improves decoding efficiency. |
Utilizing Graph Measure to Deduce Omitted Entities in Paragraphs (C18-2)
Copied to clipboard
| Challenge: | Existing studies on relation extraction only take into account intrasentence relationships that contain pairs of entities. |
| Approach: | They propose to capture omitted arguments in relation extraction given a proper knowledge base for entities of interest. |
| Outcome: | The proposed method improves relation extraction quality by capturing omitted arguments in sentences. |
FactKG: Fact Verification via Reasoning on Knowledge Graphs (2023.acl-long)
Copied to clipboard
| Challenge: | knowledge graphs (KGs) have not been fully utilized as a knowledge source for fact verification. |
| Approach: | They propose a dataset to enable the community to better use knowledge graphs . they propose 108k natural language claims with five types of reasoning . |
| Outcome: | The proposed dataset consists of 108k natural language claims with five types of reasoning . authors believe the proposed method can advance reliability and practicality . |
Exploring Cross-Cultural Differences in English Hate Speech Annotations: From Dataset Construction to Analysis (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing datasets for hate speech detection neglect the cultural diversity within a single language. |
| Approach: | They propose a CR**oss-cultural **E**nglish **Hate* speech dataset that uses culturally hateful keywords to identify posts from four countries plus the United States. |
| Outcome: | The proposed dataset shows that only 56.2% of the posts in CREHate achieve consensus among all countries, with the highest pairwise label difference rate of 26%. |
3D-Aware Vision-Language Models Fine-Tuning with Geometric Distillation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Vision-Language Models (VLMs) have shown remarkable performance on diverse visual and linguistic tasks, yet they remain limited in their understanding of 3D spatial structures. |
| Approach: | They propose a framework that injects human-inspired geometric cues into pretrained VLMs . they use sparse correspondences, relative depth relations and dense cost volumes . |
| Outcome: | The proposed framework outperforms existing methods on vision-language reasoning and 3D perception benchmarks. |
TRUEBench: Can LLM Response Meet Real-world Constraints as Productivity Assistant? (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing benchmarks fail to evaluate large language models' instruction-following capabilities . current benchmarks lack multilinguality, implicit constraints and multi-turn dialogue . |
| Approach: | a new benchmark is designed to evaluate large language models' instruction-following capabilities . the benchmark features input prompts across 12 languages and includes inter-instance multilingual instructions . |
| Outcome: | a new benchmark for large language models (LLMs) is designed to assess their performance in real-world settings. |
Two-Step Question Retrieval for Open-Domain QA (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing question retrieval models have shown a significant increase in inference speed but at the cost of lower QA performance compared to the retriever-reader pipeline. |
| Approach: | They propose a two-step question retrieval model with distant supervision to improve inference speed. |
| Outcome: | The proposed model significantly increases the performance of existing question retrieval models with a negligible loss on inference speed. |
Perceptions to Beliefs: Exploring Precursory Inferences for Theory of Mind in Large Language Models (2024.emnlp-main)
Copied to clipboard
| Challenge: | While theory of mind (ToM) is naturally developed for humans in childhood, large language models (LLMs) exhibit inconsistency in ToM tasks, despite early reports of successful cases. |
| Approach: | They propose to evaluate human ToM precursors-perception inference and perception-to-belief inference-in large language models (LLMs) by annotating characters’ perceptions on ToMi and FANToM. |
| Outcome: | The proposed method significantly improves LLMs’ performance in false belief scenarios. |
Detection of Adversarial Examples in Text Classification: Benchmark and Baseline via Robust Density Estimation (2022.findings-acl)
Copied to clipboard
| Challenge: | Word-level adversarial attacks have shown success in NLP, decreasing performance of transformer-based models with smaller perturbation rate. |
| Approach: | They propose a dataset for four popular attack methods on four datasets and four models to encourage further research in this field. |
| Outcome: | The proposed baseline has the highest auc on 29 out of 30 dataset-attack-model combinations. |
MUG-Eval: A Proxy Evaluation Framework for Multilingual Generation Capabilities in Any Language (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Evaluating text generation capabilities of large language models (LLMs) is challenging, especially for low-resource languages where methods for direct assessment are scarce. |
| Approach: | They propose a framework that transforms existing benchmarks into conversational tasks and measures LLMs’ accuracies on those tasks. |
| Outcome: | The proposed framework correlates strongly with established benchmarks while enabling standardized comparisons across languages and models. |