Unverified50% confidenceTradeoffExact time
In multi-agent scenarios, as the number of agents and discussion rounds increases, conversation history tokens rapidly approach or exceed the context window limit, leading to token bloat and exponentially growing costs.
1
Sources
50%
Confidence
Long-term
Relevance
8/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMulti-agent systems are prone to falling into an 'infinite conversation' trap, where agents repeatedly exchange messages without converging, causing token consumption and costs to grow exponentially.80% similarUnverified随着对话轮次增加，每轮需携带完整历史上下文，输入token数量持续膨胀，导致推理时间呈现非线性增长78% similarUnverifiedAgent间通信会导致上下文窗口线性增长使成本呈指数级攀升，以及长对话链中的「目标漂移」现象77% similarUnverified当前多Agent系统面临的核心技术挑战是上下文窗口与长程记忆问题，模型对早期内容的关注度会随对话深度增加而下降（注意力稀释现象）73% similarUnverifiedAs multi-turn tool calls and conversations accumulate in Agent tasks, context expansion causes inference costs to rise exponentially and response latency to increase73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/826962API
curl https://kongchang.com/api/v1/knowledge/claims/826962MCP
get_claim(id=826962)