🧑‍🔬 Daniel D. Kang

🏛️ Affiliations

🎓University of Illinois Urbana-ChampaignUrbana · United States2022–2026 · 39 papersEmployment since 2023 · Assistant Professor, Computer Sciencecurrent🎓Stanford UniversityStanford · United States2016–2024 · 19 papersEducation 2016–2023🎓Massachusetts Institute of TechnologyCambridge · United States2011–2017 · 5 papersEducation 2011–2015🔬MIT Computer Science and Artificial Intelligence LaboratoryCambridge · United States2016–2017 · 2 papers🎓University of California, BerkeleyBerkeley · United States2022–2023Employment 2022–2023 · Postdoc, Computer Science🎓University of CambridgeCambridge · United Kingdom2015–2016Education 2015–2016

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 4 more institutions on single bylines are not listed.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents (2024 · 193 citations)
   LLM agents that call tools often read outside content such as emails, reviews, notes or web pages. An attacker can plant instructions in that content to steer the agent against its own user. This is called indirect prompt injection (IPI). InjecAgent is presented as the first benchmark that measures how vulnerable…
2. Improved Natural Language Generation via Loss Truncation (2020 · 130 citations)
   Conditional text generators are usually trained with log loss (KL divergence). This objective forces the model to cover every reference in the training set, so a small share of noisy or invalid targets can distort what it learns. Examples include titles containing facts absent from the source. Kang and Hashimoto…
3. Teams of LLM Agents can Exploit Zero-Day Vulnerabilities (2026 · 18 citations)
   The paper asks whether LLM-based agents can exploit real-world zero-day vulnerabilities, meaning flaws whose description the attacker does not have.
4. SafeSearch: Do Not Trade Safety for Utility in LLM Search Agents (2026 · 1 citation)
   The paper looks at the safety of LLM search agents, which repeatedly write queries, retrieve documents, and reason over them.

All works

17 works of theirs in our library, the most cited first; 4 with a page here, the others link their original until we summarise them.

1. Exploiting Programmatic Behavior of LLMs: Dual-Use Through Standard Security Attacks ↗IEEE Security and Privacy Workshops (SPW) · no summary yetDaniel D. Kang et al.2024 · 206 citations
2. InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model AgentsFindings of the Association for Computational LinguisticsQiusi Zhan, Zhixiang Liang, Zifan Ying, et al.2024 · 193 citations
3. Improved Natural Language Generation via Loss TruncationAnnual Meeting of the Association for Computational Linguistics (ACL)Daniel Kang, Tatsunori B. Hashimoto2020 · 130 citations
4. Testing Robustness Against Unforeseen Adversaries ↗arXiv (Cornell University) · no summary yetKaufmann, Max et al.2019 · 80 citations
5. Identifying and Mitigating the Security Risks of Generative AI ↗Foundations and Trends® in Privacy and Security · no summary yetClark Barrett et al.2023 · 79 citations
6. ZKML: An Optimizing System for ML Inference in Zero-Knowledge Proofs ↗no summary yetBing-Jyue Chen et al.2024 · 73 citations
7. Scaling up Trustless DNN Inference with Zero-Knowledge Proofs ↗arXiv (Cornell University) · no summary yetDaniel D. Kang et al.2022 · 27 citations
8. A Safe Harbor for AI Evaluation and Red Teaming ↗arXiv (Cornell University) · no summary yetShayne Longpre et al.2024 · 22 citations
9. Transfer of Adversarial Robustness Between Perturbation Types ↗arXiv (Cornell University) · no summary yetDaniel D. Kang et al.2019 · 22 citations
10. Teams of LLM Agents can Exploit Zero-Day VulnerabilitiesConference of the European Chapter of the Association for Computational Linguistics (EACL)Yuxuan Zhu, Antony Kellermann, Akul Gupta, et al.2026 · 18 citations
11. Holistic Agent Leaderboard: The Missing Infrastructure for AI Agent Evaluation ↗arXiv:2510.11977 · no summary yetSayash Kapoor et al.2025 · 11 citations
12. Dias: Dynamic Rewriting of Pandas Code ↗arXiv:2303.16146 · no summary yetStefanos Baziotis, Daniel Kang, Charith Mendis2024 · 9 citations
13. Establishing Best Practices for Building Rigorous Agentic Benchmarks ↗arXiv:2507.02825 · no summary yetYuxuan Zhu et al.2025 · 7 citations
14. Establishing Best Practices in Building Rigorous Agentic Benchmarks ↗neural information processing systems · no summary yetYuxuan Zhu et al.2025 · 4 citations
15. ZK-IMG: Attested Images via Zero-Knowledge Proofs to Fight Disinformation ↗arXiv (Cornell University) · no summary yetDaniel D. Kang et al.2022 · 4 citations
16. Trustless Audits without Revealing Data or Models ↗arXiv (Cornell University) · no summary yetSuppakit Waiwitlikhit et al.2024 · 2 citations
17. SafeSearch: Do Not Trade Safety for Utility in LLM Search AgentsFindings of the Association for Computational LinguisticsQiusi Zhan, Angeline Budiman-Chan, Abdelrahman Zayed, et al.2026 · 1 citation

🤝 Co-authors here

5 co-authors with a page, the most papers together first.

• Qiusi Zhan ×2
• Joo-Kyung Kim
• Tatsunori Hashimoto
• Xingzhi Guo
• Yuxuan Zhu

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• Findings of the Association for Computational Linguistics ×2: Findings of the Association for Computational Linguistics 2026, Findings of the Association for Computational Linguistics 2024
• Annual Meeting of the Association for Computational Linguistics (ACL): ACL 2020
• Conference of the European Chapter of the Association for Computational Linguistics (EACL): EACL 2026
• Foundations and Trends® in Privacy and Security
• IEEE Security and Privacy Workshops (SPW)
• neural information processing systems

🔗 Cited by

Papers of our library cite their works 320 times (of the 149,439 papers whose bibliographies we read); 77 citing papers with a page here →.

Updated 31 October 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/daniel-d-kang/