Research map / Reinforcement learning / RL from feedback
MIRA: Memory-Integrated Reinforcement Learning Agent with Limited LLM Guidance
Narjes Nourzad, Carlee Joe-Wong
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add memory-graph utility shaping of advantages | PPO · MiniGrid/BabyAI (RedBall, LavaCrossing, DoorKey, RedBlueDoor) | return / success rate | Improved |
| add memory-graph utility shaping | PPO · FrozenLake-8x8 | mean return | Mixed |
| use MIRA instead of per-step LLM supervision (LLM4Teach) | MIRA · Distracted DoorKey | return per LLM query: about 30 queries vs more than 500 | Improved |
| use MIRA instead of LLM-RS reward shaping | MIRA · Distracted DoorKey | return | Improved |
| add online LLM queries to offline-initialized memory (ablation) | MIRA · DoorKey | learning speed: 0 -> 10 -> 20 online queries | Improved |
| add offline-only memory graph (ablation) | PPO · DoorKey | return | Improved |
| swap to degraded LLM (Gemini Pro) and disable screening late in training (ablation) | MIRA · Distracted DoorKey | return | No clear effect |
| change LLM used to build memory (Gemma3, Claude, Gemini, o4-mini) (ablation) | MIRA · DoorKey variants | asymptotic return | Mixed |
Closest papers by meaning
- LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Models (NeurIPS 2025 · similarity 0.89)
- AEM: Adaptive Entropy Modulation for Multi-Turn Agentic Reinforcement Learning (NeurIPS 2026 · similarity 0.88)
- Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents (NeurIPS 2026 · similarity 0.88)
- From Player to Master: Enhancing Test-Time Learning of LLM Agents via Reinforcement Learning over Memory (ICML 2026 · similarity 0.88)
- ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering (ICML 2026 · similarity 0.88)
- Learning Agentic Policy from Action Guidance (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
