frett commited on
Commit
d10d360
·
verified ·
1 Parent(s): c46f95d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -7
README.md CHANGED
@@ -304,23 +304,30 @@ dataset_info:
304
  dataset_size: 1114583
305
  ---
306
 
 
307
  **Text2DistBench** is a reading comprehension benchmark for evaluating whether large language models can infer distributional knowledge from natural language evidence.
308
- Given metadata and a set of user comments about an entity (e.g., a movie or song), models must understand and response statistics such as:
309
  stance/topic distribution, and most/second- frequent labels.
310
 
 
 
 
311
  The dataset is constructed from movie and music entities released between 2025-12-01 and 2026-03-01.
312
  It includes two types of configurations:
313
 
314
- **Posterior configurations** include both metadata and user comments as evidence.
315
- These configurations differ by sample size (50 or 100 comments) and task type (estimation, most_freq, second_freq):
316
- `posterior_sampled_<comment_num>_<task>`
 
317
 
318
- **Prior configurations** contain only metadata without comments: `prior_<task>`
 
 
319
 
320
 
321
  ---
322
 
323
- ## Dataset Format
324
  Each benchmark instance corresponds to a distributional reading comprehension question.
325
  ```
326
  {
@@ -339,7 +346,7 @@ Each benchmark instance corresponds to a distributional reading comprehension qu
339
 
340
  ---
341
 
342
- ## Load the Dataset
343
  ```bash
344
  from datasets import load_dataset
345
  ds = load_dataset("frett/Text2DistBench", "posterior_sampled_50_estimation", split="test")
 
304
  dataset_size: 1114583
305
  ---
306
 
307
+ ## 📌 Overview
308
  **Text2DistBench** is a reading comprehension benchmark for evaluating whether large language models can infer distributional knowledge from natural language evidence.
309
+ Given metadata and a set of user comments about an entity (e.g., a movie or song), models must infer statistics such as:
310
  stance/topic distribution, and most/second- frequent labels.
311
 
312
+ ---
313
+
314
+ ## ⚙️ Configurations
315
  The dataset is constructed from movie and music entities released between 2025-12-01 and 2026-03-01.
316
  It includes two types of configurations:
317
 
318
+ ### 🟦 Posterior Configurations
319
+ Include both metadata and user comments as evidence.
320
+ These configurations differ by sample size (50 or 100 comments) and task type (estimation, most_freq, second_freq).
321
+ **Format:** `posterior_sampled_<comment_num>_<task>`
322
 
323
+ ### 🟨 Prior Configurations
324
+ Include **only metadata (no comments)**.
325
+ **Format:** `prior_<task>`
326
 
327
 
328
  ---
329
 
330
+ ## ⚙️ Dataset Format
331
  Each benchmark instance corresponds to a distributional reading comprehension question.
332
  ```
333
  {
 
346
 
347
  ---
348
 
349
+ ## ⚙️ Load the Dataset
350
  ```bash
351
  from datasets import load_dataset
352
  ds = load_dataset("frett/Text2DistBench", "posterior_sampled_50_estimation", split="test")