Add authors to dataset card
Browse files
README.md
CHANGED
|
@@ -41,6 +41,8 @@ configs:
|
|
| 41 |
|
| 42 |
**Developed and released by [Kay.ai](https://kay.ai).**
|
| 43 |
|
|
|
|
|
|
|
| 44 |
LongListBench measures **long-list structured extraction** from insurance and commercial trucking PDFs: return every target record without omissions, merged rows, invented extras, or lost fields under complex layout, OCR, and long-range evidence. It contains 32 synthetic PDFs and 29,599 target records, with no real customer PII.
|
| 45 |
|
| 46 |
The task is per-document extraction: give a system one PDF or OCR transcript plus the target contract, then score the complete list. `core_operations` emphasizes scale and output completeness; `claim_multihop` and `policy_packets` require inherited context, heterogeneous schemas, distant supporting sections, and distractors.
|
|
|
|
| 41 |
|
| 42 |
**Developed and released by [Kay.ai](https://kay.ai).**
|
| 43 |
|
| 44 |
+
**Authors:** [Anton Fedoruk](https://orcid.org/0009-0004-0260-1704), [Serhii Shchoholiev](https://orcid.org/0009-0007-2014-4828), and [Akhil Mehta](https://orcid.org/0009-0001-0134-2905)
|
| 45 |
+
|
| 46 |
LongListBench measures **long-list structured extraction** from insurance and commercial trucking PDFs: return every target record without omissions, merged rows, invented extras, or lost fields under complex layout, OCR, and long-range evidence. It contains 32 synthetic PDFs and 29,599 target records, with no real customer PII.
|
| 47 |
|
| 48 |
The task is per-document extraction: give a system one PDF or OCR transcript plus the target contract, then score the complete list. `core_operations` emphasizes scale and output completeness; `claim_multihop` and `policy_packets` require inherited context, heterogeneous schemas, distant supporting sections, and distractors.
|