Papers by Bugeun Kim

8 papers
Inspecting Soundness of AMR Similarity Metrics in terms of Equivalence and Inequivalence (2024.starsem-1)

Copied to clipboard

Challenge: Existing Abstract Meaning Representation (AMR) similarity metrics have less investigated their soundness .
Approach: They propose a new experimental method to evaluate soundness of AMR similarity metrics in terms of equivalence and inequivalentity.
Outcome: The proposed method satisfies the soundness criteria of existing AMR similarity metrics and improves them by proposing a revised metric, SMATCH .
DART: An AIGT Detector using AMR of Rephrased Text (2025.naacl-short)

Copied to clipboard

Challenge: Existing methods for detecting AIGTs focus on probabilistic features, causing problems . performance of black-box detectors is low, and it is difficult to detect black- box models .
Approach: They propose a detector that can discriminate multiple black-box LLMs without probabilistic features and the origin of AIGT.
Outcome: The proposed method can discriminate multiple black-box LLMs without probabilistic features and the origin of AIGT.
EPT-X: An Expression-Pointer Transformer model that generates eXplanations for numbers (2022.acl-long)

Copied to clipboard

Challenge: Existing neural approaches to solve algebraic word problems have a plausible answer, but this belief has less been verified due to Q.
Approach: They propose a neural model EPT-X which utilizes natural language explanations to solve an algebraic word problem.
Outcome: The proposed model achieves an average performance of 69.59% on a PEN dataset and produces explanations with quality comparable to human output.
VoiceBBQ: Investigating Effect of Content and Acoustics in Social Bias of Spoken Language Model (2025.emnlp-main)

Copied to clipboard

Challenge: Due to the nature of speech modality, social bias in Spoken Language Models (SLMs) can emerge from two distinct sources: 1) content aspect and 2) acoustic aspect.
Approach: They propose a dataset that measures social bias by presenting ambiguous or disambiguated contexts followed by questions that may elicit stereotypical responses.
Outcome: The proposed dataset converts every BBQ context into controlled voice conditions, enabling per-axis accuracy, bias, and consistency scores comparable to the original text benchmark.
Generating Equation by Utilizing Operators : GEO model (2020.coling-main)

Copied to clipboard

Challenge: Existing neural models that use hand-crafted features are expensive and lack domain-specific knowledge.
Approach: They propose a GEO model that uses operator-based features to generate equations using natural language sentences.
Outcome: The proposed model outperforms state-of-the-art models on two datasets and 82.1% in ALG514.
Point to the Expression: Solving Algebraic Word Problems using the Expression-Pointer Transformer Model (2020.emnlp-main)

Copied to clipboard

Challenge: Existing models that generate solution equations using ‘Op (operator/operand) tokens suffered expression fragmentation and operand-context separation.
Approach: They propose a pure neural model, Expression-Pointer Transformer, which uses (1) ‘Expression’ token and (2) operand-context pointers when generating solution equations.
Outcome: The proposed model achieves comparable performance accuracy to state-of-the-art models and achieves better performance than existing models by at most 40%.
Belief in Authority: Impact of Authority in Multi-Agent Evaluation Framework (2026.findings-acl)

Copied to clipboard

Challenge: Multi-agent systems utilizing large language models assign authoritative roles to improve performance, yet the impact of authority bias on agent interactions remains underexplored.
Approach: They propose to classify authoritative roles into legitimate, referent, and expert types and analyze their influence across 12-turn conversations using French and Raven’s power-based theory.
Outcome: The proposed model enables agents to perform better in multi-agent evaluations.
People will agree what I think: Investigating LLM’s False Consensus Effect (2025.findings-naacl)

Copied to clipboard

Challenge: Recent studies have focused on the False Consensus Effect (FCE) where individuals overestimate the extent to which others share their beliefs or behaviors.
Approach: They conduct two studies to examine the FCE phenomenon in Large Language Models (LLMs) they find that popular LLMs have FCE and that they have different prompting styles.
Outcome: The proposed model is popular among LLM users and specifies the conditions when FCE becomes more or less prevalent compared to normal usage.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations