Datasets:
update readme
Browse files
README.md
CHANGED
|
@@ -35,9 +35,20 @@ size_categories:
|
|
| 35 |
|
| 36 |
# DTLBench
|
| 37 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
DTLBench is a benchmark for **deployment-time learning** of large language model agents. It collects diverse task streams spanning medical diagnosis, legal analysis, operational reasoning, financial prediction, text-to-SQL, embodied decision making, tabular reasoning on EHRs, deep search, etc.
|
| 39 |
|
| 40 |
-
The benchmark is used by **CASCADE**, whose codebase is released at [GitHub Repo](https://github.com/guosyjlu/CASCADE).
|
| 41 |
|
| 42 |
## Benchmark Overview
|
| 43 |
|
|
|
|
| 35 |
|
| 36 |
# DTLBench
|
| 37 |
|
| 38 |
+
<p align="center">
|
| 39 |
+
<a href="https://github.com/guosyjlu/CASCADE">
|
| 40 |
+
💻 GitHub Repo
|
| 41 |
+
</a> |
|
| 42 |
+
<a href="https://physionet.org/content/mimic-iv-ext-dtlbench/1.0.0/">
|
| 43 |
+
🫀 DTLBench PhysioNet (To be released)
|
| 44 |
+
</a> |
|
| 45 |
+
<a href="https://arxiv.org/abs/xxxx">
|
| 46 |
+
📄 Paper (To be released)
|
| 47 |
+
</a>
|
| 48 |
+
</p>
|
| 49 |
+
|
| 50 |
DTLBench is a benchmark for **deployment-time learning** of large language model agents. It collects diverse task streams spanning medical diagnosis, legal analysis, operational reasoning, financial prediction, text-to-SQL, embodied decision making, tabular reasoning on EHRs, deep search, etc.
|
| 51 |
|
|
|
|
| 52 |
|
| 53 |
## Benchmark Overview
|
| 54 |
|