Toma Suzuki

3 papers · 2025–2025 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (11)

Venues

ACL (1) COLING (1) NAACL (1)

Top co-authors

Taro Watanabe (3) Yusuke Sakai (3) Hidetaka Kamigaito (3) Kazuki Hayashi (2) Seiji Gobara (2) Shigeki Saito (1) Kazuma Onishi (1) Yusuke Ide (1) Hibiki Nakatani (1) Justin Vasselli (1)

Research topics

Artificial Intelligence (3) Machine Learning (3) Core AI (3) Optimization & Theory (2) Deep Learning (1) Natural Language Processing (1) Application Areas (1) Learning Types · Machine Learning (1) Learning Types · Deep Learning (1) Resources & Methods (1)

Keywords

model robustness (1) instruction tuning (1) evaluation framework (1) rational reasoning (1) language model robustness (1) vision-language model (1) vulnerability detection (1) multimodal evaluation (1) vulnerability analysis (1) distribution prediction (1) large language model (1) task-specific vulnerability (1) model vulnerability (1) output reliability (1) counter-intuitive reasoning (1) template overfitting (1) instruction template (1) human interpretation alignment (1) critic review (1) response distribution (1)

Papers

IRR: Image Review Ranking Framework for Evaluating Vision-Language Models COLING 2025 Reliability of Distribution Predictions by LLMs: Insights from Counterintuitive Pseudo-Distributions NAACL 2025 Superfluous Instruction: Vulnerabilities Stemming from Task-Specific Superficial Expressions in Instruction Templates ACL 2025