Research map / Multi-agent systems and games / LLM agents
Large Language Models Are Semi-Parametric Reinforcement Learning Agents
Danyang Zhang, Lu Chen, Situo Zhang, Hongshen Xu, Zihan Zhao, Kai Yu
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 n-step bootstrapping to Q updates (ablation) | Rememberer · WikiHow | success rate: 0.89 -> 0.93 | Improved |
| add n-step bootstrapping to Q updates (ablation) | Rememberer · WebShop | success rate: 0.36 -> 0.37 | No clear effect |
| add n-step bootstrapping to Q updates (ablation) | Rememberer · WebShop | abs error of reward estimation: 0.22 -> 0.02 | Improved |
| add random advice when no suitable candidates (ablation) | Rememberer · WikiHow | success rate: 0.90 -> 0.93 | Improved |
| add random advice when no suitable candidates (ablation) | Rememberer · WebShop | success rate: 0.37 -> 0.37 | No clear effect |
| include discouraged actions in action advice (ablation) | Rememberer · WikiHow | success rate: 0.81 -> 0.93 | Improved |
| add task similarity f_g to retrieval (ablation) | Rememberer · WikiHow | success rate: 0.94 -> 0.93 | No clear effect |
| add observation similarity f_o to retrieval (ablation) | Rememberer · WikiHow | success rate: 0.87 -> 0.93 | Improved |
| use Double Q-Learning instead of plain Q-Learning | Rememberer · WikiHow | abs error of reward estimation: 0.12 -> 0.43 | Hurt |
| add experience memory with RL updates (Rememberer vs fixed exemplars) | text-davinci-003 agent · WebShop | success rate: 0.29 -> 0.39 | Improved |
Closest papers by meaning
- Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions (ICLR 2026 · similarity 0.92)
- A-Mem: Agentic Memory for LLM Agents (NeurIPS 2025 · similarity 0.90)
- Exploratory Memory-Augmented LLM Agent via Hybrid On- and Off-Policy Optimization (ICLR 2026 · similarity 0.90)
- EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective (NeurIPS 2026 · similarity 0.90)
- From Player to Master: Enhancing Test-Time Learning of LLM Agents via Reinforcement Learning over Memory (ICML 2026 · similarity 0.89)
- PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents (ICML 2026 · similarity 0.89)
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.
