Research map / Large language models / Retrieval and knowledge
Accommodate Knowledge Conflicts in Retrieval-augmented LLMs: Towards Robust Response Generation in the Wild
Jiatai Wang, Zhiwei Xu, Di Jin, Xuewen Yang, Tao Li
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 Swin-VIB window filtering to in-context RAG | Llama2-7B · ConflictQA (multiple-choice) | ACC: 79.68 -> 84.04 | Improved |
| add Swin-VIB window filtering to in-context RAG | Llama2-7B · DRUID (multiple-choice) | ACC: 50.08 -> 63.24 | Improved |
| add Swin-VIB to Astute RAG | Astute RAG · TruthfulQA (open-ended QA) | EM: 53.02 -> 64.16 | Improved |
| add Swin-VIB to Self-RAG | Self-RAG · TruthfulQA (open-ended QA) | EM: 39.39 -> 58.32 | Improved |
| add Swin-VIB to Naive RAG | Naive RAG · TruthfulQA (open-ended QA) | Faithfulness: 73.03 -> 66.58 | Improved |
| add Swin-VIB to in-context RAG (uncertain answer rate) | Llama2-7B · ConflictQA (multiple-choice) | UAR: 2.96 -> 0.21 | Improved |
| use Swin-VIB instead of CK-PLUG | DeepSeek-7B · DRUID (multiple-choice) | Resistance Rate: 12.24 -> 16.55 | Improved |
| mix conflicting and supplementary context 2:2 (ablation) | Llama2 and other LLMs · ConflictQA (UAR vs proportion) | UAR | Hurt |
Closest papers by meaning
- Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts (ICLR 2024 · similarity 0.91)
- $\texttt{ConflictBank}$: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLMs (NeurIPS 2024 · similarity 0.90)
- WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia (NeurIPS 2024 · similarity 0.90)
- ClashEval: Quantifying the tug-of-war between an LLM’s internal prior and external evidence (NeurIPS 2024 · similarity 0.89)
- Boosting the Potential of Large Language Models with an Intelligent Information Assistant (NeurIPS 2024 · similarity 0.88)
- Can LLMs Take Retrieved Information with a Grain of Salt? (NeurIPS 2026 · similarity 0.87)
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.
