Improve dataset card: add paper link, GitHub link, and task category

README.md

@@ -1,4 +1,9 @@
 ---
+language:
+- ja
+license: apache-2.0
+task_categories:
+- text-retrieval
 dataset_info:
   features:
   - name: query
@@ -19,17 +24,18 @@ configs:
   data_files:
   - split: test
     path: data/test-*
-license: apache-2.0
-language:
-- ja
 ---
 
 # Dataset Card for JF-TE
 
+[**Paper**](https://huggingface.co/papers/2602.01479) | [**GitHub**](https://github.com/The-FinAI/Ebisu)
+
 ### Dataset Summary
 
 **JF-TE (Japanese Financial Term Extraction)** is a benchmark dataset for evaluating hierarchical extraction and ranking of nested financial terminology from Japanese professional disclosures. The dataset consists of 202 note-level instances extracted from 10 professional disclosures, containing 2,412 expert-curated term mentions covering 777 unique finance terms after normalization. The task addresses the challenge of boundary-sensitive grounding of finance terminology in mixed-script text, where nested compounds and script-variant loanwords (kanji, hiragana, and katakana) make term boundaries and semantic scope ambiguous. This dataset provides a linguistically grounded evaluation of financial term extraction that reflects the complexity of real Japanese financial disclosures.
 
+It is part of the **Ebisu** benchmark, presented in the paper [Ebisu: Benchmarking Large Language Models in Japanese Finance](https://huggingface.co/papers/2602.01479).
+
 ### Supported Tasks and Leaderboards
 
 * **Evaluation Metrics:**
@@ -40,7 +46,23 @@ language:
 
 ### Languages
 
-* Japanese (jp)
+* Japanese (ja)
+
+## Usage
+
+You can evaluate models on this dataset using the [lm-evaluation-harness](https://github.com/ASCRX/lm-evaluation-harness) provided by the authors:
+
+```bash
+lm_eval --model vllm \
+    --model_args "pretrained=$MODEL,tensor_parallel_size=4,gpu_memory_utilization=0.95,max_model_len=1024" \
+    --tasks jp \
+    --batch_size auto \
+    --output_path ../results/jp \
+    --hf_hub_log_args "hub_results_org=TheFinAI,details_repo_name=lm-eval-results-jp,push_results_to_hub=True,push_samples_to_hub=True,public_repo=False" \
+    --log_samples \
+    --apply_chat_template \
+    --include_path ../tasks/jp
+```
 
 ## Dataset Structure
 
@@ -101,11 +123,7 @@ Annotations were conducted following a rigorous protocol:
 
 1. **Guideline Development**: Task-specific guidelines were developed tailored to hierarchical financial terminology, with explicit rules governing term boundary decisions and the distinction between financial terminology and general expressions.
 
-2. **Iterative Refinement**: Annotation guidelines were iteratively refined through pre-annotation rounds, with particular attention to:
-   - Term boundary decisions
-   - Distinction between financial terminology and general expressions
-   - Handling of nested and overlapping terms
-   - Mixed-script variant identification
+2. **Iterative Refinement**: Annotation guidelines were iteratively refined through pre-annotation rounds, with particular attention to term boundary decisions, handling of nested terms, and mixed-script variant identification.
 
 3. **Span-Level Annotation**: Annotation was conducted at the span level, allowing multi-word expressions, compound nouns, and nested structures to be marked explicitly.
 
@@ -125,50 +143,28 @@ Annotations were conducted following a rigorous protocol:
 
 All annotations were performed by the same native-level Japanese financial experts as in JF-ICR:
 
-- **Annotator 1**: A doctoral student in Japan with a solid academic background in financial mathematics and several years of study and residence in Japan. Their research focuses on AI applications in finance, and they have previously contributed to financial benchmark annotation. Prior to doctoral studies, they accumulated professional experience as a quantitative researcher in the financial industry.
-
-- **Annotator 2**: A researcher at a Japanese fintech company with nearly two decades of experience in the Japanese financial market. They hold a Ph.D. in Economics and possess deep expertise in Japanese corporate finance, financial reporting, and investor communications. As a native-level Japanese speaker, they are familiar with the linguistic, cultural, and institutional conventions of Japanese financial disclosures.
-
-The annotation process was conducted using the Label Studio platform, ensuring consistency and reproducibility.
+- **Annotator 1**: A doctoral student in Japan with a solid academic background in financial mathematics and several years of study and residence in Japan. Their research focuses on AI applications in finance.
+- **Annotator 2**: A researcher at a Japanese fintech company with nearly two decades of experience in the Japanese financial market. They hold a Ph.D. in Economics and possess deep expertise in Japanese corporate finance.
 
 #### Inter-Annotator Agreement
 
-Inter-annotator agreement results demonstrate strong consistency in identifying financial term spans:
-- **Span-level F1**: [Reported in paper]
-- **Cohen's κ**: [Reported in paper]
-- **Krippendorff's α**: [Reported in paper]
-
-These results indicate that the annotation rules governing term boundaries and financial termhood are operational and reproducible, supporting the reliability of the resulting high-quality JF-TE dataset.
+Inter-annotator agreement results demonstrate strong consistency in identifying financial term spans. Detailed scores for Span-level F1, Cohen's κ, and Krippendorff's α are reported in the official paper.
 
 ### Personal and Sensitive Information
 
-The dataset contains no personal or sensitive information. All data is derived from publicly available professional disclosures (有価証券報告書) that are required to be published by publicly listed companies in Japan. Company names and financial details are preserved as they are essential for understanding the context and meaning of the financial terminology, but no individual personal information is included.
+The dataset contains no personal or sensitive information. All data is derived from publicly available professional disclosures (有価証券報告書).
 
 ## Considerations for Using the Data
 
 ### Social Impact of Dataset
 
-JF-TE contributes to research in Japanese financial NLP by enabling evaluation of financial term extraction capabilities in linguistically complex settings. Accurate extraction of financial terminology is essential for downstream tasks such as financial analysis, regulatory compliance, and automated financial document processing. The dataset supports development of systems that can better understand Japanese financial disclosures, potentially improving accessibility and analysis capabilities for both domestic and international stakeholders.
+JF-TE contributes to research in Japanese financial NLP by enabling evaluation of financial term extraction capabilities in linguistically complex settings. Accurate extraction of financial terminology is essential for downstream tasks such as financial analysis, regulatory compliance, and automated financial document processing.
 
 ### Discussion of Biases
 
-* **Document Type**: The dataset focuses on Annual Securities Reports (有価証券報告書), which may not represent all types of Japanese financial disclosures.
-
-* **Company Representation**: The dataset covers 10 professional disclosures, which may not represent the full diversity of Japanese corporate communication styles across different industries and company sizes.
-
-* **Term Coverage**: The dataset emphasizes financial and accounting terminology, which may underrepresent other types of domain-specific language that appear in financial disclosures.
-
-* **Script Representation**: While the dataset includes mixed-script terms, the distribution may reflect patterns specific to the selected disclosures.
-
-### Other Known Limitations
-
-* **Limited Size**: The dataset contains 202 note-level instances, which may limit coverage of all possible financial term patterns.
-
-* **Note-Level Focus**: The dataset focuses on note-level excerpts rather than full documents, which may miss some contextual information relevant for term extraction.
-
-* **Nested Term Complexity**: The hierarchical nature of nested terms may make evaluation challenging, as different models may prioritize different levels of the hierarchy.
-
-* **Mixed-Script Handling**: The complexity of mixed-script Japanese text (kanji, hiragana, katakana) may pose challenges for models not specifically trained on Japanese text.
+* **Document Type**: The dataset focuses on Annual Securities Reports (有価証券報告書).
+* **Company Representation**: The dataset covers 10 professional disclosures, which may not represent the full diversity of Japanese corporate communication styles.
+* **Term Coverage**: The dataset emphasizes financial and accounting terminology.
 
 ## Additional Information
 
@@ -185,15 +181,13 @@ JF-TE contributes to research in Japanese financial NLP by enabling evaluation o
 If you use this dataset, please cite:
 
 ```bibtex
-@misc{ebisu2025,
-      title={EBISU: Benchmarking Large Language Models in Japanese Finance}, 
-      author={[Authors]},
-      year={2025},
-      eprint={[arXiv number]},
+@misc{peng2026ebisu,
+      title={Ebisu: Benchmarking Large Language Models in Japanese Finance}, 
+      author={Xueqing Peng and Ruoyu Xiang and Fan Zhang and Mingzi Song and Mingyang Jiang and Yan Wang and Lingfei Qian and Taiki Hara and Yuqing Guo and Jimin Huang and Junichi Tsujii and Sophia Ananiadou},
+      year={2026},
+      eprint={2602.01479},
       archivePrefix={arXiv},
       primaryClass={cs.CL},
-      url={https://arxiv.org/abs/[number]}, 
+      url={https://huggingface.co/papers/2602.01479}, 
 }
-```
-
-**Note:** Please check the official paper for the complete citation once published.
+```