Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0
Most reported gains from agent-optimization methods are one-shot: an agent is optimized against a fixed benchmark and the resulting improvement is reported as if it were a stable property of the method. This does not test the setting that matters for deployed agents, where optimization is applied recursively as new failures and new tasks appear over time. The central question this raises is whether optimizer-driven gains compound: after an agent has been optimized once, can it be optimized again on newly arrived tasks without eroding the gains the first round produced? We study this question w
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 59%AgentCore-8B →
“Fuzzy title match (0.73): “Do Agent Optimizers Compound? A Continual-Learning Evaluatio” ≈ “AgentCore-8B””
- PossiblePossibly related (embedding) · 52%[2607.07508] Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning →
- FuzzySimilar title/name (fuzzy) · 87%zhayujie/CowAgent →
“Fuzzy title match (0.94): “Do Agent Optimizers Compound? A Continual-Learning Evaluatio” ≈ “zhayujie/CowAgent””
- FuzzyOverlapping authors or contributors · 62%bytedance/deer-flow →
“Shared author/contributor keys: wang”
- FuzzyOverlapping authors or contributors · 62%ray-project/ray →
“Shared author/contributor keys: wang”
- FuzzySimilar title/name (fuzzy) · 59%NousResearch/hermes-agent →
“Fuzzy title match (0.73): “Do Agent Optimizers Compound? A Continual-Learning Evaluatio” ≈ “NousResearch/hermes-agent””
- FuzzySimilar title/name (fuzzy) · 59%2FastLabs/agent-squad →
“Fuzzy title match (0.73): “Do Agent Optimizers Compound? A Continual-Learning Evaluatio” ≈ “2FastLabs/agent-squad””
- LinkedLinked via arxiv author · 85%Wenxiao Wang →
“Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0”