MemEvoBench is presented as the first standardized benchmark for long-horizon memory safety in LLM agents, covering adversarial memory injection, noisy tool outputs, and biased feedback across QA and workflow tasks.
hub Canonical reference
Your agent may misevolve: Emergent risks in self-evolving llm agents
Canonical reference. 100% of citing Pith papers cite this work as background.
12
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 5
citation-polarity summary
years
2026 12roles
background 4polarities
background 4representative citing papers
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
Evolving-RL jointly optimizes experience extraction and utilization in LLM agents via RL with separate evaluation signals, delivering up to 98.7% relative gains on out-of-distribution tasks in ALFWorld and Mind2Web.
AgentForesight introduces an online auditor model that predicts decisive errors in multi-agent trajectories at the earliest step using a coarse-to-fine reinforcement learning recipe on a new curated dataset AFTraj-2K.
BeliefMem is a probabilistic memory architecture for LLM agents that retains multiple candidate conclusions with probabilities updated by Noisy-OR, achieving superior average performance over deterministic baselines on LoCoMo and ALFWorld.
PACE coordinates low-risk prompt evolution with validated higher-risk control-logic updates to improve frozen SLM agents on benchmarks without model retraining.
Priority ranking offers a low-cost direct evaluation for harness optimizers that correlates with their real multi-step optimization performance, supported by the Shor dataset of 182 scenarios.
OEP poisons self-evolving LLM agents by constructing clean edge-case experiences that appear locally valid yet cause harmful over-generalization during reflection, achieving over 50% attack success rate on GPT-4o agents across three domains.
FATE lets LLM agents self-evolve safer behaviors by generating and filtering repairs from their own failure trajectories using verifiers and Pareto optimization.
Self-evolving LLM agents exhibit capability erosion under continual adaptation, which Capability-Preserving Evolution mitigates by raising retained simple-task performance from 41.8% to 52.8% in workflow evolution under GPT-5.1.
Poisoning any single CIK dimension of an AI agent raises average attack success rate from 24.6% to 64-74% across models, and tested defenses leave substantial residual risk.
citing papers explorer
-
MemEvoBench: Benchmarking Safety Risks from Memory Misevolution in LLM Agents
MemEvoBench is presented as the first standardized benchmark for long-horizon memory safety in LLM agents, covering adversarial memory injection, noisy tool outputs, and biased feedback across QA and workflow tasks.
-
What Do Evolutionary Coding Agents Evolve?
Evolutionary coding agents achieve most benchmark gains through a small subset of edit types and by cycling previously deleted code lines rather than developing new algorithmic structures.
-
Evolving-RL: End-to-End Optimization of Experience-Driven Self-Evolving Capability within Agents
Evolving-RL jointly optimizes experience extraction and utilization in LLM agents via RL with separate evaluation signals, delivering up to 98.7% relative gains on out-of-distribution tasks in ALFWorld and Mind2Web.
-
AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems
AgentForesight introduces an online auditor model that predicts decisive errors in multi-agent trajectories at the earliest step using a coarse-to-fine reinforcement learning recipe on a new curated dataset AFTraj-2K.
-
Belief Memory: Agent Memory Under Partial Observability
BeliefMem is a probabilistic memory architecture for LLM agents that retains multiple candidate conclusions with probabilities updated by Noisy-OR, achieving superior average performance over deterministic baselines on LoCoMo and ALFWorld.
-
PACE: Two-Timescale Self-Evolution for Small Language Model Agents
PACE coordinates low-risk prompt evolution with validated higher-risk control-logic updates to improve frozen SLM agents on benchmarks without model retraining.
-
Towards Direct Evaluation of Harness Optimizers via Priority Ranking
Priority ranking offers a low-cost direct evaluation for harness optimizers that correlates with their real multi-step optimization performance, supported by the Shor dataset of 182 scenarios.
-
OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences
OEP poisons self-evolving LLM agents by constructing clean edge-case experiences that appear locally valid yet cause harmful over-generalization during reflection, achieving over 50% attack success rate on GPT-4o agents across three domains.
-
On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
FATE lets LLM agents self-evolve safer behaviors by generating and filtering repairs from their own failure trajectories using verifiers and Pareto optimization.
-
Do Self-Evolving Agents Forget? Capability Degradation and Preservation in Lifelong LLM Agent Adaptation
Self-evolving LLM agents exhibit capability erosion under continual adaptation, which Capability-Preserving Evolution mitigates by raising retained simple-task performance from 41.8% to 52.8% in workflow evolution under GPT-5.1.
-
Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw
Poisoning any single CIK dimension of an AI agent raises average attack success rate from 24.6% to 64-74% across models, and tested defenses leave substantial residual risk.
- Agentic AI for Remote Sensing: Technical Challenges and Research Directions