More Memory, Worse Agents: Error Reproduction and Anti-Persistence in LLM Agents

Key points

• Retaining these abstractions allows agents to reuse strategies without repeatedly deriving them from raw trajectories.1
• We term this feedback mechanism error reproduction: errors are regenerated through a cycle of abstraction, execution, and re-abstraction, extending their persistence beyond direct reuse of the original faulty knowledge.1
• We propose Anti-Persistence, a memory design that accumulates trajectories and outcomes while reconstructing task-specific guidance on demand.1
• These results support separating experience accumulation from abstraction persistence in memorybased self-improvement.1

Sources (2)

• [1]More Memory, Worse Agents: Error Reproduction and Anti-Persistence in LLM Agents
  Hugging Face Daily Papers · Oct 19, 12:00 AM

  LLM agents can improve across tasks without updating model parameters by storing experience from their own executions in external memory and reusing it to guide later decisions.

  Retaining these abstractions allows agents to reuse strategies without repeatedly deriving them from raw trajectories.

• [2]More Memory, Worse Agents: Error Reproduction and Anti-Persistence in LLM Agents
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 19, 11:58 PM · same content

Extractive summary: sentences quoted from the sources.