SEABench: Benchmarking Endogenous Misalignment In Self-Evolving Agents
cs.CR, cs.AI, cs.CL
Submitted: 2026-09-28
Updated: 2026-09-29
Code: https://github.com/SEABench-Endogenous-Misalignment/SEABench
Terminology
Sources
- GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
- On Randomness in Agentic Evals
- Beyond Jailbreaking: Auditing Contextual Privacy in LLM Agents
- AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
- Memory Injection Attacks on LLM Agents via Query-Only Interaction
- Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models
- On the Reliability of Computer Use Agents
- Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection
- Reinforcement Learning via Self-Distillation
- Eliciting Harmful Capabilities by Fine-Tuning On Safeguarded Outputs
- ReasoningFlow: Discourse Structures for Understanding LLM Reasoning Traces
- "Someone Hid It": Query-Agnostic Black-Box Attacks on LLM-Based Retrieval
- Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
- Self-Distillation Enables Continual Learning
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing
- Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models
- Zombie Agents: Persistent Control of Self-Evolving LLM Agents via Self-Reinforcing Injections
- MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
- Memento-Skills: Let Agents Design Agents
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs