Research map / Large language models / Retrieval and knowledge
GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation
Yifan Wang, Mingxuan Jiang, Zhihao Sun, Yixin Cao, Yicun Liu, Keyang Chen, Guangnan Ye, Hongfeng Chai
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 gain-adaptive sentence memory to the retrieval graph | GAM-RAG · MuSiQue | GPT-Acc (%): 35.30 -> 43.30 | Improved |
| add 2 turns of memorization | GAM-RAG · 2Wiki | GPT-Acc (%): 64.20 -> 67.80 | Improved |
| add 1 turn of memorization | GAM-RAG · HotpotQA | GPT-Acc (%): 74.90 -> 74.90 | No clear effect |
| add 5 turns of memorization | GAM-RAG · 2Wiki same query | GPT-Acc (%): 65.30 -> 71.80 | Improved |
| add 5 turns of memorization | GAM-RAG · 2Wiki same query | retrieval time (s): 7.43 -> 3.11 | Improved |
| use GAM-RAG instead of event-graph DyG-RAG | GAM-RAG · TimeQA | GPT-Acc (%): 51.26 -> 49.45 | Hurt |
| remove task-memory vector (ablation) | GAM-RAG · four datasets (10-turn) | accuracy | Improved |
| remove time-memory vector (ablation) | GAM-RAG · TimeQA | accuracy | Improved |
| add uncertainty-scaled propagation (ablation) | GAM-RAG · long-term memorization | accuracy | Improved |
| add Kalman gain scheduling instead of fixed small learning rate (ablation) | GAM-RAG · long-term memorization | accuracy | Improved |
| inject 20% noisy feedback (ablation) | GAM-RAG · 2Wiki | GPT-Acc (%): 72.9 -> 71.8 at round 10 | Hurt |
| increase activated sentences/entities per iteration too far (ablation) | GAM-RAG · 2Wiki | GPT-Acc | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- LinearRAG — LinearRAG: Linear Graph Retrieval Augmented Generation on Large-scale Corpora (ICLR 2026) · better on 4 of 4 shared results
- HippoRAG2 — HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models (NeurIPS 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- From RAG to Memory: Non-Parametric Continual Learning for Large Language Models (ICML 2025 · similarity 0.89)
- Adapting to Non-Stationary Environments: Multi-Armed Bandit Enhanced Retrieval-Augmented Generation on Knowledge Graphs (AAAI 2025 · similarity 0.89)
- Long-Context LLMs Meet RAG: Overcoming Challenges for Long Inputs in RAG (ICLR 2025 · similarity 0.89)
- Speculative RAG: Enhancing Retrieval Augmented Generation through Drafting (ICLR 2025 · similarity 0.88)
- Inference Scaling for Long-Context Retrieval Augmented Generation (ICLR 2025 · similarity 0.88)
- RAG-DDR: Optimizing Retrieval-Augmented Generation Using Differentiable Data Rewards (ICLR 2025 · 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.
