Nina Panickssery

4 papers · 2024–2026 · 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 👥 Mega-Team · Squad

Venues

NeurIPS (2) EACL (1) ICLR (1)

Top co-authors

Ethan Perez (1) Joe Benton (1) Neel Nanda (1) Fracesco Mosconi (1) Samuel R. Bowman (1) Naomi Bashkansky (1) Deep Ganguli (1) Esin Durmus (1) Christopher Ackerman (1) Frauke Kreuter (1)

Research topics

Artificial Intelligence (4) Core AI (4) Natural Language Processing (1) Resources & Methods (1)

Keywords

large language model (2) model interpretability (1) model alignment (1) latent space (1) jailbreaking attack (1) adversarial prompt (1) red teaming (1) context window (1) steering vector (1) safety fine-tuning (1) causal intervention (1) residual stream (1) weight editing (1) model activation (1) ai safety (1) refusal direction (1)

Papers

Understanding Jailbreak Success: A Study of Latent Space Dynamics in Large Language Models EACL 2026 Inspection and Control of Self-Generated-Text Recognition Ability in Llama3-8b-Instruct ICLR 2025 Many-shot Jailbreaking NeurIPS 2024 Refusal in Language Models Is Mediated by a Single Direction NeurIPS 2024