Simplify base dataset attribution

Remove the base benchmark BibTeX block while retaining a concise provenance link.

README.md

@@ -182,20 +182,7 @@ The extension and its 1.14B-token evaluation are described in:
 
 The 618M-token base corpus and the 500 evaluation questions come from
 [EnterpriseRAG-Bench](https://huggingface.co/datasets/onyx-dot-app/EnterpriseRAG-Bench).
-They are redistributed without modifying their records. Users of the combined
-collection should also cite the source benchmark:
-
-```bibtex
-@misc{sun2026enterpriseragbench,
-  title         = {EnterpriseRAG-Bench: A RAG Benchmark for Company Internal Knowledge},
-  author        = {Sun, Yuhong and Rahmfeld, Joachim and Weaver, Chris and Desai, Roshan and Huang, Wenxi and Butler, Mark H.},
-  year          = {2026},
-  eprint        = {2605.05253},
-  archivePrefix = {arXiv},
-  primaryClass  = {cs.IR},
-  url           = {https://arxiv.org/abs/2605.05253}
-}
-```
+They are redistributed without modifying their records.
 
 ---