ACL Findings2026
Context Attribution with Multi-Armed Bandit Optimization
Deng Pan, Keerthiram Murugesan, Ting Hua, Nuno Moniz, Nitesh V. Chawla
Abstract
Understanding which parts of the retrieved context contribute to a large language model's generated answer is essential for building interpretable and trustworthy retrieval-augmented generation. We propose a novel framework that formulates context attribution as a combinatorial multi-armed bandit problem. We utilize Linear Thompson Sampling to efficiently identify the most influential context segments while minimizing the number of model queries. Our reward function leverages token log-probabilities to measure how well a subset of segments supports the original response, making it applicable to both open-source and black-box API-based models. Unlike SHAP and other perturbation-based methods that sample subsets uniformly, our approach adaptively prioritizes informative subsets based on posterior estimates of segment relevance, reducing computational costs. Experiments on multiple QA benchmarks demonstrate that our method achieves up to 30% reduction in model queries while matching or exceeding the attribution quality of existing approaches. Our code is publicly available at https://github.com/ pd90506/camab.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Multi-Level Explanations for Generative Language Models cites Generating Hierarchical Explanations on Text Classification via Feature Interaction Detection
- Multi-Level Explanations for Generative Language Models cites ContextCite: Attributing Model Generation to Context
Builds on4
- ContextCite: Attributing Model Generation to ContextBenjamin Cohen-Wang, Harshay Shah, Kristian Georgiev, Aleksander MadryNeurIPS 2024 · 118 citations
- Generating Hierarchical Explanations on Text Classification via Feature Interaction DetectionHanjie Chen, Guangtao Zheng, Yangfeng JiACL 2020 · 85 citations
- Multi-Level Explanations for Generative Language ModelsLucas Monteiro Paes, Dennis Wei, Hyo Jin Do, Hendrik Strobelt et al.ACL 2025 · 16 citations
- Training Language Models to Generate Text with Citations via Fine-grained RewardsChengyu Huang, Zeqiu Wu, Yushi Hu, Wenya WangACL 2024 · 4 citations
Related papers
- TextGenSHAP: Scalable Post-Hoc Explanations in Text Generation with Long DocumentsJames Enouen, Hootan Nakhost, Sayna Ebrahimi, Sercan Ö. Arik et al.ACL Findings 2024 · 3 citations
- TokenShapley: Token Level Context Attribution with Shapley ValueYingtai Xiao, Yuqing Zhu, Sirat Samyoun, Wanrong Zhang et al.ACL Findings 2025 · 2 citations
- PRCA: Fitting Black-Box Large Language Models for Retrieval Question Answering via Pluggable Reward-Driven Contextual AdapterHaoyan Yang, Zhitao Li, Yong Zhang, Jianzong Wang et al.EMNLP 2023 · 16 citations
- Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented GenerationJirui Qi, Gabriele Sarti, Raquel Fernández, Arianna BisazzaEMNLP 2024 · 6 citations
- Why Retrieval-Augmented Generation Fails: A Graph PerspectiveKai Guo, Xinnan Dai, Zhibo Zhang, Nuohan Lin et al.KDD 2026
