Copertina

Multi-Step Reasoning in Korean and the Emergent Mirage

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Son, Guijin, Ko, Hyunwoo, Choi, Dasol
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Understand, Solve and Translate: Bridging the Multilingual Mathematical Reasoning Gap
di: Ko, Hyunwoo, et al.
Pubblicazione: (2025)

Ko-PIQA: A Korean Physical Commonsense Reasoning Dataset with Cultural Context
di: Choi, Dasol, et al.
Pubblicazione: (2025)

Won: Establishing Best Practices for Korean Financial NLP
di: Son, Guijin, et al.
Pubblicazione: (2025)

KAIO: A Collection of More Challenging Korean Questions
di: Lee, Nahyun, et al.
Pubblicazione: (2025)

Linguistic Generalizability of Test-Time Scaling in Mathematical Reasoning
di: Son, Guijin, et al.
Pubblicazione: (2025)

KMMMU: Evaluation of Massive Multi-discipline Multimodal Understanding in Korean Language and Context
di: Lee, Nahyun, et al.
Pubblicazione: (2026)

Controlling Language Confusion in Multilingual LLMs
di: Lee, Nahyun, et al.
Pubblicazione: (2025)

LLM-as-a-Judge & Reward Model: What They Can and Cannot Do
di: Son, Guijin, et al.
Pubblicazione: (2024)

Improving Fine-grained Visual Understanding in VLMs through Text-Only Training
di: Choi, Dasol, et al.
Pubblicazione: (2024)

Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thought
di: Son, Guijin, et al.
Pubblicazione: (2025)

Redefining Evaluation Standards: A Unified Framework for Evaluating the Korean Capabilities of Language Models
di: Lee, Hanwool, et al.
Pubblicazione: (2025)

ResearchMath-14K: Scaling Research-Level Mathematics via Agents
di: Son, Guijin, et al.
Pubblicazione: (2026)

Revisiting the UID Hypothesis in LLM Reasoning Traces
di: Gwak, Minju, et al.
Pubblicazione: (2025)

Revisiting the Uniform Information Density Hypothesis in LLM Reasoning
di: Gwak, Minju, et al.
Pubblicazione: (2025)

KMMLU: Measuring Massive Multitask Language Understanding in Korean
di: Son, Guijin, et al.
Pubblicazione: (2024)

Pushing the Boundaries of Multiple Choice Evaluation to One Hundred Options
di: Lee, Nahyun, et al.
Pubblicazione: (2026)

From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluation
di: Hong, Seokhee, et al.
Pubblicazione: (2025)

Judging What We Cannot Solve: A Consequence-Based Approach for Oracle-Free Evaluation of Research-Level Math
di: Son, Guijin, et al.
Pubblicazione: (2026)

When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research
di: Son, Guijin, et al.
Pubblicazione: (2025)

HAE-RAE Bench: Evaluation of Korean Knowledge in Language Models
di: Son, Guijin, et al.
Pubblicazione: (2023)

K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts
di: Lee, Nahyun, et al.
Pubblicazione: (2026)

Multi-Task Inference: Can Large Language Models Follow Multiple Instructions at Once?
di: Son, Guijin, et al.
Pubblicazione: (2024)

No Language Data Left Behind: A Comparative Study of CJK Language Datasets in the Hugging Face Ecosystem
di: Choi, Dasol, et al.
Pubblicazione: (2025)

What Users Leave Unsaid: Under-Specified Queries Limit Vision-Language Models
di: Choi, Dasol, et al.
Pubblicazione: (2026)

Graph Elicitation for Guiding Multi-Step Reasoning in Large Language Models
di: Park, Jinyoung, et al.
Pubblicazione: (2023)

KORMo: Korean Open Reasoning Model for Everyone
di: Kim, Minjun, et al.
Pubblicazione: (2025)

KoSimpleQA: A Korean Factuality Benchmark with an Analysis of Reasoning LLMs
di: Ko, Donghyeon, et al.
Pubblicazione: (2025)

Better Safe Than Sorry? Overreaction Problem of Vision Language Models in Visual Emergency Recognition
di: Choi, Dasol, et al.
Pubblicazione: (2025)

Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback
di: Son, Guijin, et al.
Pubblicazione: (2026)

ESG Classification by Implicit Rule Learning via GPT-4
di: Yun, Hyo Jeong, et al.
Pubblicazione: (2024)

Unveiling Causal Reasoning in Large Language Models: Reality or Mirage?
di: Chi, Haoang, et al.
Pubblicazione: (2025)

The Mirage of Model Editing: Revisiting Evaluation in the Wild
di: Yang, Wanli, et al.
Pubblicazione: (2025)

Does Thinking More always Help? Mirage of Test-Time Scaling in Reasoning Models
di: Ghosal, Soumya Suvra, et al.
Pubblicazione: (2025)

First-Step Advantage: Importance of Starting Right in Multi-Step Math Reasoning
di: Jain, Kushal, et al.
Pubblicazione: (2023)

Relevance-aware Multi-context Contrastive Decoding for Retrieval-augmented Visual Question Answering
di: Kim, Jongha, et al.
Pubblicazione: (2026)

IndicGEC: Powerful Models, or a Measurement Mirage?
di: Vajjala, Sowmya
Pubblicazione: (2025)

Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens
di: Zhao, Chengshuai, et al.
Pubblicazione: (2025)

From Tokens to Steps: Verification-Aware Speculative Decoding for Efficient Multi-Step Reasoning
di: Purohit, Kiran, et al.
Pubblicazione: (2026)

KorMedMCQA: Multi-Choice Question Answering Benchmark for Korean Healthcare Professional Licensing Examinations
di: Kweon, Sunjun, et al.
Pubblicazione: (2024)

Correct Prediction, Wrong Steps? Consensus Reasoning Knowledge Graph for Robust Chain-of-Thought Synthesis
di: Ling, Zipeng, et al.
Pubblicazione: (2026)