Datasets:
Restructure into configs, add dataset card
Browse files- README.md +267 -3
- combined_shots_pool.json +0 -52
- complex_shots_pool.json +0 -44
- listqa_dataset.jsonl → data/all.jsonl +0 -0
- complex_listqa_dataset.jsonl → data/complex.jsonl +0 -0
- simple_listqa_dataset.jsonl → data/simple.jsonl +0 -0
- few_shot/combined.jsonl +12 -0
- few_shot/complex.jsonl +10 -0
- few_shot/simple.jsonl +14 -0
- simple_shots_pool.json +0 -60
README.md
CHANGED
|
@@ -1,10 +1,274 @@
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
| 3 |
-
task_categories:
|
| 4 |
-
- text-generation
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
pretty_name: ListQA
|
| 8 |
size_categories:
|
| 9 |
- 1K<n<10K
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
|
|
|
|
|
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
pretty_name: ListQA
|
| 6 |
size_categories:
|
| 7 |
- 1K<n<10K
|
| 8 |
+
task_categories:
|
| 9 |
+
- question-answering
|
| 10 |
+
- text-generation
|
| 11 |
+
tags:
|
| 12 |
+
- factuality
|
| 13 |
+
- knowledge-retrieval
|
| 14 |
+
- list-generation
|
| 15 |
+
- structured-output
|
| 16 |
+
- hallucination
|
| 17 |
+
- benchmark
|
| 18 |
+
- wikipedia
|
| 19 |
+
configs:
|
| 20 |
+
- config_name: all
|
| 21 |
+
default: true
|
| 22 |
+
data_files:
|
| 23 |
+
- split: test
|
| 24 |
+
path: data/all.jsonl
|
| 25 |
+
- config_name: simple
|
| 26 |
+
data_files:
|
| 27 |
+
- split: test
|
| 28 |
+
path: data/simple.jsonl
|
| 29 |
+
- config_name: complex
|
| 30 |
+
data_files:
|
| 31 |
+
- split: test
|
| 32 |
+
path: data/complex.jsonl
|
| 33 |
+
- config_name: few_shot
|
| 34 |
+
data_files:
|
| 35 |
+
- split: simple
|
| 36 |
+
path: few_shot/simple.jsonl
|
| 37 |
+
- split: complex
|
| 38 |
+
path: few_shot/complex.jsonl
|
| 39 |
+
- split: combined
|
| 40 |
+
path: few_shot/combined.jsonl
|
| 41 |
+
---
|
| 42 |
+
|
| 43 |
+
# ListQA: A Benchmark for Evaluating List-Formatted Factual Knowledge Retrieval in Large Language Models
|
| 44 |
+
|
| 45 |
+
**Project page:** [listqa.github.io](https://listqa.github.io) · **Venue:** NeurIPS 2026, Evaluations and Datasets Track
|
| 46 |
+
|
| 47 |
+
ListQA tests whether an LLM can recall *many* facts from its parameters and arrange them in a list. Most factual QA benchmarks ask for a single answer. ListQA instead has **9,045 human-written, cross-validated questions**, and each answer is a list of **3–10 elements**. Every question and answer is grounded in a Wikipedia table. The questions cover eight topical categories and 60+ countries.
|
| 48 |
+
|
| 49 |
+
There are two question types:
|
| 50 |
+
|
| 51 |
+
- **SimpleListQA** (single-faceted, 5,378 questions): the answer is a flat numbered list of entities.
|
| 52 |
+
- **ComplexListQA** (multi-faceted, 3,667 questions): the answer is a hierarchical list. Each entity carries sub-attributes such as *recipient* or *result*. These questions often require reasoning over several table columns and applying exclusion or negation constraints.
|
| 53 |
+
|
| 54 |
+
The dataset is **evergreen**. Each question either has an explicit time constraint (e.g. *"between 2017 and 2024"*) or asks about facts that cannot change historically. Because of this, the ground truth stays valid without periodic updates. All answers were validated as correct as of **December 31, 2024**.
|
| 55 |
+
|
| 56 |
+
In the paper, the best model under standard (non-thinking) inference scores below **37% LLM-Judge-F1**. When the source Wikipedia page is given as context (oracle RAG), scores rise by 34–65 points. This shows that recalling facts from parameters, not organizing them, is the main bottleneck.
|
| 57 |
+
|
| 58 |
+
## Quick start
|
| 59 |
+
|
| 60 |
+
```python
|
| 61 |
+
from datasets import load_dataset
|
| 62 |
+
|
| 63 |
+
full = load_dataset("AmazonScience/ListQA", "all", split="test") # 9,045
|
| 64 |
+
simple = load_dataset("AmazonScience/ListQA", "simple", split="test") # 5,378
|
| 65 |
+
complex_ = load_dataset("AmazonScience/ListQA", "complex", split="test") # 3,667
|
| 66 |
+
|
| 67 |
+
# Few-shot example pools (for 5-shot prompting)
|
| 68 |
+
shots = load_dataset("AmazonScience/ListQA", "few_shot", split="simple") # or "complex" / "combined"
|
| 69 |
+
```
|
| 70 |
+
|
| 71 |
+
## Dataset structure
|
| 72 |
+
|
| 73 |
+
### Configurations
|
| 74 |
+
|
| 75 |
+
| Config | Split | Rows | Description |
|
| 76 |
+
|---|---|---|---|
|
| 77 |
+
| `all` (default) | `test` | 9,045 | Full benchmark (Simple + Complex) |
|
| 78 |
+
| `simple` | `test` | 5,378 | Single-faceted questions with flat-list answers |
|
| 79 |
+
| `complex` | `test` | 3,667 | Multi-faceted questions with hierarchical-list answers |
|
| 80 |
+
| `few_shot` | `simple` / `complex` / `combined` | 14 / 10 / 12 | Pools of in-context examples for few-shot evaluation. These are generic, not ListQA questions. |
|
| 81 |
+
|
| 82 |
+
`simple` and `complex` are disjoint, and together they make up exactly `all`.
|
| 83 |
+
|
| 84 |
+
### Fields (`all`, `simple`, `complex`)
|
| 85 |
+
|
| 86 |
+
| Field | Type | Description |
|
| 87 |
+
|---|---|---|
|
| 88 |
+
| `problem` | string | The question. |
|
| 89 |
+
| `answer` | string | The reference answer: a short optional preamble followed by a numbered list (format below). |
|
| 90 |
+
| `metadata` | string | JSON-encoded string with all metadata fields. This is the format the official evaluation code expects. |
|
| 91 |
+
| `category` | string | Topical category: `Sports`, `Entertainment`, `Politics`, `Movies`, `Lifestyle`, `Places of Interest`, `Food`, `Travel`. |
|
| 92 |
+
| `prompt_type` | string | `single-faceted` (Simple) or `multi-faceted` (Complex). |
|
| 93 |
+
| `ordered` | bool | `true` if the list must follow a specific order (chronological, ranked, sequential). |
|
| 94 |
+
| `year_range` | string | Temporal scope of the question, e.g. `2022-2024`, or `Timeless` for facts that cannot change historically. |
|
| 95 |
+
| `country` | string | Country or region the question concerns. |
|
| 96 |
+
| `wikitable_source` | string | URL of the Wikipedia page that holds the source table. Use it to verify the ground truth independently. |
|
| 97 |
+
|
| 98 |
+
The top-level `category`, `prompt_type`, `ordered`, `year_range`, `country`, and `wikitable_source` columns copy fields from `metadata` so you can filter on them in the Dataset Viewer. The `metadata` string also has two annotation-bookkeeping fields, `data_dump_date` and `data_dump_row`.
|
| 99 |
+
|
| 100 |
+
### Fields (`few_shot`)
|
| 101 |
+
|
| 102 |
+
| Field | Type | Description |
|
| 103 |
+
|---|---|---|
|
| 104 |
+
| `question` | string | Example question. |
|
| 105 |
+
| `answer` | string | Example answer in the canonical list format. |
|
| 106 |
+
|
| 107 |
+
### Answer format
|
| 108 |
+
|
| 109 |
+
Simple (single-faceted) answers are flat numbered lists:
|
| 110 |
+
|
| 111 |
+
```
|
| 112 |
+
1. List Element 1
|
| 113 |
+
2. List Element 2
|
| 114 |
+
3. List Element 3
|
| 115 |
+
```
|
| 116 |
+
|
| 117 |
+
Complex (multi-faceted) answers add `* Category: Value` sub-bullets under each element:
|
| 118 |
+
|
| 119 |
+
```
|
| 120 |
+
1. List Element 1
|
| 121 |
+
* Category 1: Sub-Element 1
|
| 122 |
+
* Category 2: Sub-Element 2
|
| 123 |
+
2. List Element 2
|
| 124 |
+
* Category 1: Sub-Element 1
|
| 125 |
+
* Category 2: Sub-Element 2
|
| 126 |
+
```
|
| 127 |
+
|
| 128 |
+
Because of this consistent format, a simple regex can parse the reference lists into elements: `\n\d+\.\s(.+?)(?=\n\d+\.|\Z)` (DOTALL) finds the top-level elements, and `^\* (.*?): (.*?)$` (MULTILINE) finds the sub-attributes.
|
| 129 |
+
|
| 130 |
+
### Examples
|
| 131 |
+
|
| 132 |
+
**Simple:**
|
| 133 |
+
|
| 134 |
+
```json
|
| 135 |
+
{
|
| 136 |
+
"problem": "Name the legislations introduced and supported in 2017 by the United States Chamber of Commerce.",
|
| 137 |
+
"answer": "...\n\n1. Ozone Implementation Act of 2017\n2. Furthering Asbestos Claim Transparency Act\n3. Fairness in Class Action Litigation Act\n4. SAFE Act\n5. Reauthorization Act\n6. Self Drive Act\n7. Tribal Tax and Investment Reform Act of 2017",
|
| 138 |
+
"category": "Politics", "prompt_type": "single-faceted", "ordered": false,
|
| 139 |
+
"year_range": "2017-2018", "country": "USA",
|
| 140 |
+
"wikitable_source": "https://en.wikipedia.org/wiki/United_States_Chamber_of_Commerce#"
|
| 141 |
+
}
|
| 142 |
+
```
|
| 143 |
+
|
| 144 |
+
**Complex:**
|
| 145 |
+
|
| 146 |
+
```json
|
| 147 |
+
{
|
| 148 |
+
"problem": "Which years between 2017 and 2024 were works performed by Dua Lipa nominated for the MTV Video Music Award for Best Choreography, and what was the name of the work performed by her?",
|
| 149 |
+
"answer": "...\n\n1. 2018\n* Name of the work: \"IDGAF\"\n\n2. 2020\n* Name of the work: \"Physical\"\n\n3. 2023\n* Name of the work: \"Dance the Night\"\n\n4. 2024\n* Name of the work: \"Houdini\"",
|
| 150 |
+
"category": "Entertainment", "prompt_type": "multi-faceted", "ordered": false,
|
| 151 |
+
"year_range": "2017-2024", "country": "USA",
|
| 152 |
+
"wikitable_source": "https://en.wikipedia.org/wiki/MTV_Video_Music_Award_for_Best_Choreography"
|
| 153 |
+
}
|
| 154 |
+
```
|
| 155 |
+
|
| 156 |
+
## Dataset statistics
|
| 157 |
+
|
| 158 |
+
| Property | Value |
|
| 159 |
+
|---|---|
|
| 160 |
+
| Total samples | 9,045 |
|
| 161 |
+
| SimpleListQA / ComplexListQA | 5,378 (59.5%) / 3,667 (40.5%) |
|
| 162 |
+
| Unique Wikipedia source pages | ~3,765 |
|
| 163 |
+
| Elements per answer | 3–8 (mean 4.65, median 4) |
|
| 164 |
+
| Year-constrained / Timeless | 70.1% / 29.9% |
|
| 165 |
+
| Ordered / Unordered | 14.6% / 85.4% |
|
| 166 |
+
| US-centric samples | ~78% |
|
| 167 |
+
| Temporal coverage | 2000–2024, emphasis on 2017–2024 |
|
| 168 |
+
|
| 169 |
+
| Category | Simple | Complex | Total |
|
| 170 |
+
|---|---|---|---|
|
| 171 |
+
| Sports | 1,100 | 722 | 1,822 |
|
| 172 |
+
| Entertainment (incl. Music) | 1,033 | 731 | 1,764 |
|
| 173 |
+
| Politics | 950 | 789 | 1,739 |
|
| 174 |
+
| Movies | 927 | 772 | 1,699 |
|
| 175 |
+
| Lifestyle | 279 | 275 | 554 |
|
| 176 |
+
| Places of Interest | 360 | 139 | 499 |
|
| 177 |
+
| Food | 341 | 153 | 494 |
|
| 178 |
+
| Travel | 388 | 86 | 474 |
|
| 179 |
+
|
| 180 |
+
## Evaluation
|
| 181 |
+
|
| 182 |
+
Each question is preceded by a system instruction that describes the two list formats. The benchmark supports two prompting modes:
|
| 183 |
+
|
| 184 |
+
- **zero-shot:** the instruction plus the question.
|
| 185 |
+
- **few-shot:** 5 in-context examples drawn from the matching `few_shot` pool (`simple` for SimpleListQA, `complex` for ComplexListQA, `combined` for the full set).
|
| 186 |
+
|
| 187 |
+
<details>
|
| 188 |
+
<summary>Zero-shot instruction prompt</summary>
|
| 189 |
+
|
| 190 |
+
```
|
| 191 |
+
Answer the question below by producing a list-formatted output,
|
| 192 |
+
lists can be formatted one of two ways:
|
| 193 |
+
For simple questions with few constraints requesting only main
|
| 194 |
+
elements and not requesting sub-elements, use the following format:
|
| 195 |
+
1. List Element 1
|
| 196 |
+
2. List Element 2
|
| 197 |
+
For complex questions asking for multiple factoids in addition to
|
| 198 |
+
the primary question, use the following format:
|
| 199 |
+
1. List Element 1
|
| 200 |
+
* Category 1: Sub-Element 1
|
| 201 |
+
* Category 2: Sub-Element 2
|
| 202 |
+
2. List Element 2
|
| 203 |
+
* Category 1: Sub-Element 1
|
| 204 |
+
* Category 2: Sub-Element 2
|
| 205 |
+
All answers are bounded by 3 to 10 main elements, do not generate
|
| 206 |
+
more than 10 main elements in the list. Do not add any information
|
| 207 |
+
that is not asked for in the question.
|
| 208 |
+
<QUESTION>
|
| 209 |
+
```
|
| 210 |
+
|
| 211 |
+
</details>
|
| 212 |
+
|
| 213 |
+
The paper reports two headline metrics:
|
| 214 |
+
|
| 215 |
+
- **LLM-Judge-F1** (factual accuracy): parse the reference and predicted lists into elements. Align them one-to-one by solving a linear sum assignment over pairwise ROUGE-1 scores (Jonker-Volgenant). An LLM judge (GPT-OSS-120B by default) then grades each matched pair CORRECT or INCORRECT. Unmatched predictions count as INCORRECT. Precision and recall are computed over CORRECT elements. This metric correlates with human F1 at Pearson r = 0.965.
|
| 216 |
+
- **Format-Correctness** (instruction following): the model must produce the right list structure, flat for single-faceted questions and hierarchical for multi-faceted ones, with at least 3 elements.
|
| 217 |
+
|
| 218 |
+
The paper also reports Relaxed-Precision/Recall (string match), ROUGE-F1, and **LLM-Judge-F1-Ordered**, which uses positional pairing and applies only to samples with `ordered = true`. The evaluation code is linked from [listqa.github.io](https://listqa.github.io).
|
| 219 |
+
|
| 220 |
+
## Dataset creation
|
| 221 |
+
|
| 222 |
+
### Source data
|
| 223 |
+
|
| 224 |
+
Every question and its reference answer come from a table on English Wikipedia. Wikipedia appears in most LLM training corpora, which makes it a fair shared knowledge source. Tables also lend themselves to list answers, because several rows can satisfy one query. Each sample records its source URL in `wikitable_source`.
|
| 225 |
+
|
| 226 |
+
### Annotation
|
| 227 |
+
|
| 228 |
+
Annotators were given these instructions:
|
| 229 |
+
|
| 230 |
+
1. Find Wikipedia tables on popular entities in the eight categories.
|
| 231 |
+
2. Write close-ended, objective questions that draw on one or more table columns.
|
| 232 |
+
3. Write list answers with 3–10 elements.
|
| 233 |
+
4. Add explicit time constraints where facts could change.
|
| 234 |
+
5. Tag whether the list must be ordered.
|
| 235 |
+
|
| 236 |
+
Subjective questions (e.g. *"top 10 vegan restaurants in NYC"*) were excluded.
|
| 237 |
+
|
| 238 |
+
### Quality control
|
| 239 |
+
|
| 240 |
+
Quality control had four stages:
|
| 241 |
+
|
| 242 |
+
1. Two pilot rounds calibrated the annotators. Pilot data was excluded from the dataset.
|
| 243 |
+
2. Each QA pair was written by one of 44 annotators and validated by a second annotator who did not write it. Every pair was validated.
|
| 244 |
+
3. Pairs that needed major rework were discarded (10%). The rest were corrected.
|
| 245 |
+
4. At least one more reviewer verified every correction.
|
| 246 |
+
|
| 247 |
+
In total, at least three people reviewed every sample. Of the annotations sent to cross-validation, 39% passed without flags and 51% needed minor fixes. An audit of 481 stratified questions found an error rate of **1.7%** (95% Wilson CI [0.8%, 3.2%]). Most of those errors came from ambiguous wording, not factual mistakes.
|
| 248 |
+
|
| 249 |
+
### Personal and sensitive information
|
| 250 |
+
|
| 251 |
+
The dataset is built entirely from public Wikipedia content. It contains no personally identifiable information, offensive content, or sensitive data. Annotators were paid in line with applicable labor standards.
|
| 252 |
+
|
| 253 |
+
## Limitations
|
| 254 |
+
|
| 255 |
+
- **English-only and US-centric.** About 78% of samples concern US entities, so overall scores mostly reflect US-centric knowledge.
|
| 256 |
+
- **Ground-truth completeness.** Each reference comes from a single Wikipedia table. A correct element that is missing from that table would be scored as a hallucination. Explicit constraints narrow the answer set, and source URLs are included so references can be audited.
|
| 257 |
+
- **No abstention reward.** Hallucinations are penalized, but the benchmark does not reward a model for declining to answer.
|
| 258 |
+
- **LLM-judge cost and noise.** The primary metric uses an LLM judge. The paper shows that smaller judges (down to 4B) give near-identical rankings (Spearman ρ ≥ 0.96).
|
| 259 |
+
- **`country` labels** use mixed granularity. Some are plain country names (`Spain`) and some include a region prefix (`Europe - Spain`, `Africa - General`). Normalize them before running per-country analyses.
|
| 260 |
+
|
| 261 |
+
## License
|
| 262 |
+
|
| 263 |
+
Released under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/). The questions and answers reference facts from Wikipedia, whose text is available under CC BY-SA.
|
| 264 |
+
|
| 265 |
+
## Citation
|
| 266 |
+
|
| 267 |
+
```bibtex
|
| 268 |
+
@inproceedings{kulkarni2026listqa,
|
| 269 |
+
title = {ListQA: A Benchmark for Evaluating List-Formatted Factual Knowledge Retrieval in Large Language Models},
|
| 270 |
+
author = {Kulkarni, Mayank and Vaideeswaran, Rakesh},
|
| 271 |
+
booktitle = {Advances in Neural Information Processing Systems (NeurIPS), Evaluations and Datasets Track},
|
| 272 |
+
year = {2026}
|
| 273 |
+
}
|
| 274 |
+
```
|
combined_shots_pool.json
DELETED
|
@@ -1,52 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"shots": [
|
| 3 |
-
{
|
| 4 |
-
"question": "What are the primary colors?",
|
| 5 |
-
"answer": "The primary colors are the fundamental colors that cannot be created by mixing other colors.\n\n1. Red\n2. Blue\n3. Yellow"
|
| 6 |
-
},
|
| 7 |
-
{
|
| 8 |
-
"question": "What are the main branches of government in the United States?",
|
| 9 |
-
"answer": "The U.S. government is structured with three separate branches to ensure checks and balances.\n\n1. Executive branch\n2. Legislative branch\n3. Judicial branch"
|
| 10 |
-
},
|
| 11 |
-
{
|
| 12 |
-
"question": "What are the basic states of matter?",
|
| 13 |
-
"answer": "Matter exists in several distinct physical states depending on temperature and pressure conditions.\n\n1. Solid\n2. Liquid\n3. Gas\n4. Plasma"
|
| 14 |
-
},
|
| 15 |
-
{
|
| 16 |
-
"question": "What are the continents of the world?",
|
| 17 |
-
"answer": "Earth's landmasses are divided into seven major continental regions.\n\n1. Asia\n2. Africa\n3. North America\n4. South America\n5. Antarctica\n6. Europe\n7. Australia/Oceania"
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"question": "What are the major programming languages and their primary use cases?",
|
| 21 |
-
"answer": "Modern software development relies on various programming languages, each optimized for specific domains and applications.\n\n1. Python\n* Primary Use: Data science and web development\n* Paradigm: Multi-paradigm\n* Performance: Interpreted, slower execution\n\n2. JavaScript\n* Primary Use: Web development and frontend\n* Paradigm: Multi-paradigm, event-driven\n* Performance: Just-in-time compiled\n\n3. Java\n* Primary Use: Enterprise applications\n* Paradigm: Object-oriented\n* Performance: Compiled to bytecode"
|
| 22 |
-
},
|
| 23 |
-
{
|
| 24 |
-
"question": "What are the main cloud service providers and their key offerings?",
|
| 25 |
-
"answer": "The cloud computing market is dominated by several major providers, each offering distinct services and specializations.\n\n1. Amazon Web Services (AWS)\n* Market Share: ~32% of cloud market\n* Key Services: EC2, S3, Lambda, RDS\n* Specialty: Comprehensive service portfolio\n\n2. Microsoft Azure\n* Market Share: ~20% of cloud market\n* Key Services: Virtual Machines, Blob Storage, Functions\n* Specialty: Enterprise integration and hybrid cloud\n\n3. Google Cloud Platform (GCP)\n* Market Share: ~9% of cloud market\n* Key Services: Compute Engine, Cloud Storage, BigQuery\n* Specialty: Data analytics and machine learning"
|
| 26 |
-
},
|
| 27 |
-
{
|
| 28 |
-
"question": "What are the fundamental database types and their characteristics?",
|
| 29 |
-
"answer": "Database systems are designed with different architectures to handle various data storage and retrieval requirements.\n\n1. Relational Databases (SQL)\n* Structure: Tables with rows and columns\n* Consistency: ACID properties\n* Use Cases: Financial systems, traditional applications\n\n2. Document Databases (NoSQL)\n* Structure: JSON-like documents\n* Consistency: Eventually consistent\n* Use Cases: Content management, catalogs\n\n3. Key-Value Stores\n* Structure: Simple key-value pairs\n* Consistency: Eventually consistent\n* Use Cases: Caching, session storage\n\n4. Graph Databases\n* Structure: Nodes and relationships\n* Consistency: ACID or eventual\n* Use Cases: Social networks, recommendation engines"
|
| 30 |
-
},
|
| 31 |
-
{
|
| 32 |
-
"question": "What are the main types of renewable energy?",
|
| 33 |
-
"answer": "Renewable energy sources can be naturally replenished and are environmentally sustainable.\n\n1. Solar energy\n2. Wind energy\n3. Hydroelectric energy\n4. Geothermal energy\n5. Biomass energy"
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"question": "What are the major cybersecurity threats and their mitigation strategies?",
|
| 37 |
-
"answer": "Organizations face numerous cybersecurity challenges that require comprehensive defense strategies and proactive measures.\n\n1. Malware\n* Types: Viruses, worms, trojans, ransomware\n* Impact: Data theft, system damage, financial loss\n* Mitigation: Antivirus software, regular updates, user training\n\n2. Phishing Attacks\n* Types: Email, SMS, voice phishing\n* Impact: Credential theft, identity fraud\n* Mitigation: Email filtering, user awareness, multi-factor authentication\n\n3. DDoS Attacks\n* Types: Volume-based, protocol-based, application-layer\n* Impact: Service disruption, downtime\n* Mitigation: Traffic filtering, rate limiting, CDN protection"
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"question": "What are the main organs of the human digestive system?",
|
| 41 |
-
"answer": "The digestive system includes several organs that process food from ingestion to elimination.\n\n1. Mouth\n2. Esophagus\n3. Stomach\n4. Small intestine\n5. Large intestine\n6. Liver\n7. Pancreas"
|
| 42 |
-
},
|
| 43 |
-
{
|
| 44 |
-
"question": "What are the primary greenhouse gases?",
|
| 45 |
-
"answer": "Several gases in Earth's atmosphere trap heat and contribute to the greenhouse effect.\n\n1. Carbon dioxide (CO2)\n2. Methane (CH4)\n3. Nitrous oxide (N2O)\n4. Fluorinated gases"
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"question": "What are the major social media platforms and their key characteristics?",
|
| 49 |
-
"answer": "Social media platforms have evolved to serve different user needs and demographics, each with unique features and business models.\n\n1. Facebook\n* User Base: 2.9+ billion monthly active users\n* Primary Use: Social networking, sharing updates\n* Demographics: Broad age range, global reach\n\n2. Instagram\n* User Base: 2+ billion monthly active users\n* Primary Use: Photo and video sharing\n* Demographics: Younger audience, visual content focus\n\n3. Twitter/X\n* User Base: 450+ million monthly active users\n* Primary Use: Real-time news and microblogging\n* Demographics: News-focused, professional networking"
|
| 50 |
-
}
|
| 51 |
-
]
|
| 52 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
complex_shots_pool.json
DELETED
|
@@ -1,44 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"shots": [
|
| 3 |
-
{
|
| 4 |
-
"question": "What are the major programming languages and their primary use cases?",
|
| 5 |
-
"answer": "Modern software development relies on various programming languages, each optimized for specific domains and applications.\n\n1. Python\n* Primary Use: Data science and web development\n* Paradigm: Multi-paradigm\n* Performance: Interpreted, slower execution\n\n2. JavaScript\n* Primary Use: Web development and frontend\n* Paradigm: Multi-paradigm, event-driven\n* Performance: Just-in-time compiled\n\n3. Java\n* Primary Use: Enterprise applications\n* Paradigm: Object-oriented\n* Performance: Compiled to bytecode\n\n4. C++\n* Primary Use: System programming and games\n* Paradigm: Multi-paradigm\n* Performance: Compiled, high performance\n\n5. Go\n* Primary Use: Backend services and cloud\n* Paradigm: Procedural with some OOP\n* Performance: Compiled, fast execution"
|
| 6 |
-
},
|
| 7 |
-
{
|
| 8 |
-
"question": "What are the main cloud service providers and their key offerings?",
|
| 9 |
-
"answer": "The cloud computing market is dominated by several major providers, each offering distinct services and specializations.\n\n1. Amazon Web Services (AWS)\n* Market Share: ~32% of cloud market\n* Key Services: EC2, S3, Lambda, RDS\n* Specialty: Comprehensive service portfolio\n\n2. Microsoft Azure\n* Market Share: ~20% of cloud market\n* Key Services: Virtual Machines, Blob Storage, Functions\n* Specialty: Enterprise integration and hybrid cloud\n\n3. Google Cloud Platform (GCP)\n* Market Share: ~9% of cloud market\n* Key Services: Compute Engine, Cloud Storage, BigQuery\n* Specialty: Data analytics and machine learning\n\n4. IBM Cloud\n* Market Share: ~4% of cloud market\n* Key Services: Virtual Servers, Watson AI, Kubernetes\n* Specialty: Enterprise AI and hybrid cloud solutions"
|
| 10 |
-
},
|
| 11 |
-
{
|
| 12 |
-
"question": "What are the major types of machine learning algorithms and their applications?",
|
| 13 |
-
"answer": "Machine learning encompasses several distinct approaches, each suited for different types of problems and data scenarios.\n\n1. Supervised Learning\n* Definition: Learning with labeled training data\n* Applications: Classification, regression, prediction\n* Examples: Linear regression, decision trees, neural networks\n\n2. Unsupervised Learning\n* Definition: Finding patterns in unlabeled data\n* Applications: Clustering, dimensionality reduction\n* Examples: K-means, PCA, autoencoders\n\n3. Reinforcement Learning\n* Definition: Learning through interaction and rewards\n* Applications: Game playing, robotics, autonomous systems\n* Examples: Q-learning, policy gradients, actor-critic\n\n4. Semi-supervised Learning\n* Definition: Learning with partially labeled data\n* Applications: When labeling is expensive or limited\n* Examples: Self-training, co-training, graph-based methods"
|
| 14 |
-
},
|
| 15 |
-
{
|
| 16 |
-
"question": "What are the fundamental database types and their characteristics?",
|
| 17 |
-
"answer": "Database systems are designed with different architectures to handle various data storage and retrieval requirements.\n\n1. Relational Databases (SQL)\n* Structure: Tables with rows and columns\n* Consistency: ACID properties\n* Use Cases: Financial systems, traditional applications\n* Examples: MySQL, PostgreSQL, Oracle\n\n2. Document Databases (NoSQL)\n* Structure: JSON-like documents\n* Consistency: Eventually consistent\n* Use Cases: Content management, catalogs\n* Examples: MongoDB, CouchDB, Amazon DocumentDB\n\n3. Key-Value Stores\n* Structure: Simple key-value pairs\n* Consistency: Eventually consistent\n* Use Cases: Caching, session storage\n* Examples: Redis, DynamoDB, Riak\n\n4. Graph Databases\n* Structure: Nodes and relationships\n* Consistency: ACID or eventual\n* Use Cases: Social networks, recommendation engines\n* Examples: Neo4j, Amazon Neptune, ArangoDB"
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"question": "What are the main cybersecurity threats and their mitigation strategies?",
|
| 21 |
-
"answer": "Organizations face numerous cybersecurity challenges that require comprehensive defense strategies and proactive measures.\n\n1. Malware\n* Types: Viruses, worms, trojans, ransomware\n* Impact: Data theft, system damage, financial loss\n* Mitigation: Antivirus software, regular updates, user training\n\n2. Phishing Attacks\n* Types: Email, SMS, voice phishing\n* Impact: Credential theft, identity fraud\n* Mitigation: Email filtering, user awareness, multi-factor authentication\n\n3. DDoS Attacks\n* Types: Volume-based, protocol-based, application-layer\n* Impact: Service disruption, downtime\n* Mitigation: Traffic filtering, rate limiting, CDN protection\n\n4. Insider Threats\n* Types: Malicious insiders, negligent employees\n* Impact: Data breaches, intellectual property theft\n* Mitigation: Access controls, monitoring, background checks"
|
| 22 |
-
},
|
| 23 |
-
{
|
| 24 |
-
"question": "What are the major renewable energy technologies and their efficiency metrics?",
|
| 25 |
-
"answer": "Renewable energy technologies vary significantly in their efficiency, reliability, and economic viability across different applications.\n\n1. Solar Photovoltaic (PV)\n* Efficiency: 15-22% for commercial panels\n* Capacity Factor: 20-25% average\n* Cost Trend: Declining rapidly, now cost-competitive\n\n2. Wind Power\n* Efficiency: 35-45% for modern turbines\n* Capacity Factor: 25-35% onshore, 40-50% offshore\n* Cost Trend: Stable, competitive with fossil fuels\n\n3. Hydroelectric\n* Efficiency: 80-90% energy conversion\n* Capacity Factor: 40-60% depending on water availability\n* Cost Trend: Stable, long-term cost-effective\n\n4. Geothermal\n* Efficiency: 10-15% thermal to electrical conversion\n* Capacity Factor: 70-90% very reliable baseload\n* Cost Trend: Stable, location-dependent viability"
|
| 26 |
-
},
|
| 27 |
-
{
|
| 28 |
-
"question": "What are the major social media platforms and their key characteristics?",
|
| 29 |
-
"answer": "Social media platforms have evolved to serve different user needs and demographics, each with unique features and business models.\n\n1. Facebook\n* User Base: 2.9+ billion monthly active users\n* Primary Use: Social networking, sharing updates\n* Demographics: Broad age range, global reach\n\n2. Instagram\n* User Base: 2+ billion monthly active users\n* Primary Use: Photo and video sharing\n* Demographics: Younger audience, visual content focus\n\n3. Twitter/X\n* User Base: 450+ million monthly active users\n* Primary Use: Real-time news and microblogging\n* Demographics: News-focused, professional networking\n\n4. LinkedIn\n* User Base: 900+ million members\n* Primary Use: Professional networking and career development\n* Demographics: Working professionals, B2B focus"
|
| 30 |
-
},
|
| 31 |
-
{
|
| 32 |
-
"question": "What are the main types of artificial intelligence and their applications?",
|
| 33 |
-
"answer": "Artificial intelligence encompasses various approaches and capabilities, ranging from narrow task-specific systems to theoretical general intelligence.\n\n1. Narrow AI (Weak AI)\n* Definition: AI designed for specific tasks\n* Applications: Voice assistants, recommendation systems\n* Current Status: Widely deployed and commercially available\n\n2. General AI (Strong AI)\n* Definition: AI with human-level cognitive abilities\n* Applications: Theoretical - could handle any intellectual task\n* Current Status: Not yet achieved, active research area\n\n3. Machine Learning\n* Definition: AI that learns from data without explicit programming\n* Applications: Image recognition, natural language processing\n* Current Status: Foundation of most modern AI systems\n\n4. Deep Learning\n* Definition: ML using neural networks with multiple layers\n* Applications: Computer vision, speech recognition, language models\n* Current Status: Driving recent AI breakthroughs"
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"question": "What are the major economic systems and their characteristics?",
|
| 37 |
-
"answer": "Economic systems represent different approaches to organizing production, distribution, and consumption of goods and services.\n\n1. Capitalism\n* Ownership: Private ownership of means of production\n* Market Role: Free market determines prices and production\n* Government Role: Minimal intervention, regulation focused\n\n2. Socialism\n* Ownership: Public or collective ownership of means of production\n* Market Role: Central planning or democratic control\n* Government Role: Extensive economic planning and control\n\n3. Mixed Economy\n* Ownership: Combination of private and public ownership\n* Market Role: Market forces with government intervention\n* Government Role: Regulation, public services, safety nets\n\n4. Command Economy\n* Ownership: State ownership of all means of production\n* Market Role: Government controls all economic decisions\n* Government Role: Complete central planning and control"
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"question": "What are the main components of the immune system and their functions?",
|
| 41 |
-
"answer": "The human immune system consists of interconnected components that work together to defend against pathogens and maintain health.\n\n1. White Blood Cells (Leukocytes)\n* Types: T cells, B cells, neutrophils, macrophages\n* Function: Identify and destroy pathogens\n* Location: Throughout blood and lymphatic system\n\n2. Antibodies (Immunoglobulins)\n* Types: IgG, IgA, IgM, IgE, IgD\n* Function: Neutralize specific antigens\n* Production: Created by B cells in response to threats\n\n3. Lymph Nodes\n* Structure: Small, bean-shaped organs\n* Function: Filter lymph fluid and trap pathogens\n* Location: Neck, armpits, groin, and other areas\n\n4. Bone Marrow\n* Types: Red marrow (active) and yellow marrow (inactive)\n* Function: Produces all blood cells including immune cells\n* Location: Inside bones, especially flat bones"
|
| 42 |
-
}
|
| 43 |
-
]
|
| 44 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
listqa_dataset.jsonl → data/all.jsonl
RENAMED
|
The diff for this file is too large to render.
See raw diff
|
|
|
complex_listqa_dataset.jsonl → data/complex.jsonl
RENAMED
|
The diff for this file is too large to render.
See raw diff
|
|
|
simple_listqa_dataset.jsonl → data/simple.jsonl
RENAMED
|
The diff for this file is too large to render.
See raw diff
|
|
|
few_shot/combined.jsonl
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"question": "What are the primary colors?", "answer": "The primary colors are the fundamental colors that cannot be created by mixing other colors.\n\n1. Red\n2. Blue\n3. Yellow"}
|
| 2 |
+
{"question": "What are the main branches of government in the United States?", "answer": "The U.S. government is structured with three separate branches to ensure checks and balances.\n\n1. Executive branch\n2. Legislative branch\n3. Judicial branch"}
|
| 3 |
+
{"question": "What are the basic states of matter?", "answer": "Matter exists in several distinct physical states depending on temperature and pressure conditions.\n\n1. Solid\n2. Liquid\n3. Gas\n4. Plasma"}
|
| 4 |
+
{"question": "What are the continents of the world?", "answer": "Earth's landmasses are divided into seven major continental regions.\n\n1. Asia\n2. Africa\n3. North America\n4. South America\n5. Antarctica\n6. Europe\n7. Australia/Oceania"}
|
| 5 |
+
{"question": "What are the major programming languages and their primary use cases?", "answer": "Modern software development relies on various programming languages, each optimized for specific domains and applications.\n\n1. Python\n* Primary Use: Data science and web development\n* Paradigm: Multi-paradigm\n* Performance: Interpreted, slower execution\n\n2. JavaScript\n* Primary Use: Web development and frontend\n* Paradigm: Multi-paradigm, event-driven\n* Performance: Just-in-time compiled\n\n3. Java\n* Primary Use: Enterprise applications\n* Paradigm: Object-oriented\n* Performance: Compiled to bytecode"}
|
| 6 |
+
{"question": "What are the main cloud service providers and their key offerings?", "answer": "The cloud computing market is dominated by several major providers, each offering distinct services and specializations.\n\n1. Amazon Web Services (AWS)\n* Market Share: ~32% of cloud market\n* Key Services: EC2, S3, Lambda, RDS\n* Specialty: Comprehensive service portfolio\n\n2. Microsoft Azure\n* Market Share: ~20% of cloud market\n* Key Services: Virtual Machines, Blob Storage, Functions\n* Specialty: Enterprise integration and hybrid cloud\n\n3. Google Cloud Platform (GCP)\n* Market Share: ~9% of cloud market\n* Key Services: Compute Engine, Cloud Storage, BigQuery\n* Specialty: Data analytics and machine learning"}
|
| 7 |
+
{"question": "What are the fundamental database types and their characteristics?", "answer": "Database systems are designed with different architectures to handle various data storage and retrieval requirements.\n\n1. Relational Databases (SQL)\n* Structure: Tables with rows and columns\n* Consistency: ACID properties\n* Use Cases: Financial systems, traditional applications\n\n2. Document Databases (NoSQL)\n* Structure: JSON-like documents\n* Consistency: Eventually consistent\n* Use Cases: Content management, catalogs\n\n3. Key-Value Stores\n* Structure: Simple key-value pairs\n* Consistency: Eventually consistent\n* Use Cases: Caching, session storage\n\n4. Graph Databases\n* Structure: Nodes and relationships\n* Consistency: ACID or eventual\n* Use Cases: Social networks, recommendation engines"}
|
| 8 |
+
{"question": "What are the main types of renewable energy?", "answer": "Renewable energy sources can be naturally replenished and are environmentally sustainable.\n\n1. Solar energy\n2. Wind energy\n3. Hydroelectric energy\n4. Geothermal energy\n5. Biomass energy"}
|
| 9 |
+
{"question": "What are the major cybersecurity threats and their mitigation strategies?", "answer": "Organizations face numerous cybersecurity challenges that require comprehensive defense strategies and proactive measures.\n\n1. Malware\n* Types: Viruses, worms, trojans, ransomware\n* Impact: Data theft, system damage, financial loss\n* Mitigation: Antivirus software, regular updates, user training\n\n2. Phishing Attacks\n* Types: Email, SMS, voice phishing\n* Impact: Credential theft, identity fraud\n* Mitigation: Email filtering, user awareness, multi-factor authentication\n\n3. DDoS Attacks\n* Types: Volume-based, protocol-based, application-layer\n* Impact: Service disruption, downtime\n* Mitigation: Traffic filtering, rate limiting, CDN protection"}
|
| 10 |
+
{"question": "What are the main organs of the human digestive system?", "answer": "The digestive system includes several organs that process food from ingestion to elimination.\n\n1. Mouth\n2. Esophagus\n3. Stomach\n4. Small intestine\n5. Large intestine\n6. Liver\n7. Pancreas"}
|
| 11 |
+
{"question": "What are the primary greenhouse gases?", "answer": "Several gases in Earth's atmosphere trap heat and contribute to the greenhouse effect.\n\n1. Carbon dioxide (CO2)\n2. Methane (CH4)\n3. Nitrous oxide (N2O)\n4. Fluorinated gases"}
|
| 12 |
+
{"question": "What are the major social media platforms and their key characteristics?", "answer": "Social media platforms have evolved to serve different user needs and demographics, each with unique features and business models.\n\n1. Facebook\n* User Base: 2.9+ billion monthly active users\n* Primary Use: Social networking, sharing updates\n* Demographics: Broad age range, global reach\n\n2. Instagram\n* User Base: 2+ billion monthly active users\n* Primary Use: Photo and video sharing\n* Demographics: Younger audience, visual content focus\n\n3. Twitter/X\n* User Base: 450+ million monthly active users\n* Primary Use: Real-time news and microblogging\n* Demographics: News-focused, professional networking"}
|
few_shot/complex.jsonl
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"question": "What are the major programming languages and their primary use cases?", "answer": "Modern software development relies on various programming languages, each optimized for specific domains and applications.\n\n1. Python\n* Primary Use: Data science and web development\n* Paradigm: Multi-paradigm\n* Performance: Interpreted, slower execution\n\n2. JavaScript\n* Primary Use: Web development and frontend\n* Paradigm: Multi-paradigm, event-driven\n* Performance: Just-in-time compiled\n\n3. Java\n* Primary Use: Enterprise applications\n* Paradigm: Object-oriented\n* Performance: Compiled to bytecode\n\n4. C++\n* Primary Use: System programming and games\n* Paradigm: Multi-paradigm\n* Performance: Compiled, high performance\n\n5. Go\n* Primary Use: Backend services and cloud\n* Paradigm: Procedural with some OOP\n* Performance: Compiled, fast execution"}
|
| 2 |
+
{"question": "What are the main cloud service providers and their key offerings?", "answer": "The cloud computing market is dominated by several major providers, each offering distinct services and specializations.\n\n1. Amazon Web Services (AWS)\n* Market Share: ~32% of cloud market\n* Key Services: EC2, S3, Lambda, RDS\n* Specialty: Comprehensive service portfolio\n\n2. Microsoft Azure\n* Market Share: ~20% of cloud market\n* Key Services: Virtual Machines, Blob Storage, Functions\n* Specialty: Enterprise integration and hybrid cloud\n\n3. Google Cloud Platform (GCP)\n* Market Share: ~9% of cloud market\n* Key Services: Compute Engine, Cloud Storage, BigQuery\n* Specialty: Data analytics and machine learning\n\n4. IBM Cloud\n* Market Share: ~4% of cloud market\n* Key Services: Virtual Servers, Watson AI, Kubernetes\n* Specialty: Enterprise AI and hybrid cloud solutions"}
|
| 3 |
+
{"question": "What are the major types of machine learning algorithms and their applications?", "answer": "Machine learning encompasses several distinct approaches, each suited for different types of problems and data scenarios.\n\n1. Supervised Learning\n* Definition: Learning with labeled training data\n* Applications: Classification, regression, prediction\n* Examples: Linear regression, decision trees, neural networks\n\n2. Unsupervised Learning\n* Definition: Finding patterns in unlabeled data\n* Applications: Clustering, dimensionality reduction\n* Examples: K-means, PCA, autoencoders\n\n3. Reinforcement Learning\n* Definition: Learning through interaction and rewards\n* Applications: Game playing, robotics, autonomous systems\n* Examples: Q-learning, policy gradients, actor-critic\n\n4. Semi-supervised Learning\n* Definition: Learning with partially labeled data\n* Applications: When labeling is expensive or limited\n* Examples: Self-training, co-training, graph-based methods"}
|
| 4 |
+
{"question": "What are the fundamental database types and their characteristics?", "answer": "Database systems are designed with different architectures to handle various data storage and retrieval requirements.\n\n1. Relational Databases (SQL)\n* Structure: Tables with rows and columns\n* Consistency: ACID properties\n* Use Cases: Financial systems, traditional applications\n* Examples: MySQL, PostgreSQL, Oracle\n\n2. Document Databases (NoSQL)\n* Structure: JSON-like documents\n* Consistency: Eventually consistent\n* Use Cases: Content management, catalogs\n* Examples: MongoDB, CouchDB, Amazon DocumentDB\n\n3. Key-Value Stores\n* Structure: Simple key-value pairs\n* Consistency: Eventually consistent\n* Use Cases: Caching, session storage\n* Examples: Redis, DynamoDB, Riak\n\n4. Graph Databases\n* Structure: Nodes and relationships\n* Consistency: ACID or eventual\n* Use Cases: Social networks, recommendation engines\n* Examples: Neo4j, Amazon Neptune, ArangoDB"}
|
| 5 |
+
{"question": "What are the main cybersecurity threats and their mitigation strategies?", "answer": "Organizations face numerous cybersecurity challenges that require comprehensive defense strategies and proactive measures.\n\n1. Malware\n* Types: Viruses, worms, trojans, ransomware\n* Impact: Data theft, system damage, financial loss\n* Mitigation: Antivirus software, regular updates, user training\n\n2. Phishing Attacks\n* Types: Email, SMS, voice phishing\n* Impact: Credential theft, identity fraud\n* Mitigation: Email filtering, user awareness, multi-factor authentication\n\n3. DDoS Attacks\n* Types: Volume-based, protocol-based, application-layer\n* Impact: Service disruption, downtime\n* Mitigation: Traffic filtering, rate limiting, CDN protection\n\n4. Insider Threats\n* Types: Malicious insiders, negligent employees\n* Impact: Data breaches, intellectual property theft\n* Mitigation: Access controls, monitoring, background checks"}
|
| 6 |
+
{"question": "What are the major renewable energy technologies and their efficiency metrics?", "answer": "Renewable energy technologies vary significantly in their efficiency, reliability, and economic viability across different applications.\n\n1. Solar Photovoltaic (PV)\n* Efficiency: 15-22% for commercial panels\n* Capacity Factor: 20-25% average\n* Cost Trend: Declining rapidly, now cost-competitive\n\n2. Wind Power\n* Efficiency: 35-45% for modern turbines\n* Capacity Factor: 25-35% onshore, 40-50% offshore\n* Cost Trend: Stable, competitive with fossil fuels\n\n3. Hydroelectric\n* Efficiency: 80-90% energy conversion\n* Capacity Factor: 40-60% depending on water availability\n* Cost Trend: Stable, long-term cost-effective\n\n4. Geothermal\n* Efficiency: 10-15% thermal to electrical conversion\n* Capacity Factor: 70-90% very reliable baseload\n* Cost Trend: Stable, location-dependent viability"}
|
| 7 |
+
{"question": "What are the major social media platforms and their key characteristics?", "answer": "Social media platforms have evolved to serve different user needs and demographics, each with unique features and business models.\n\n1. Facebook\n* User Base: 2.9+ billion monthly active users\n* Primary Use: Social networking, sharing updates\n* Demographics: Broad age range, global reach\n\n2. Instagram\n* User Base: 2+ billion monthly active users\n* Primary Use: Photo and video sharing\n* Demographics: Younger audience, visual content focus\n\n3. Twitter/X\n* User Base: 450+ million monthly active users\n* Primary Use: Real-time news and microblogging\n* Demographics: News-focused, professional networking\n\n4. LinkedIn\n* User Base: 900+ million members\n* Primary Use: Professional networking and career development\n* Demographics: Working professionals, B2B focus"}
|
| 8 |
+
{"question": "What are the main types of artificial intelligence and their applications?", "answer": "Artificial intelligence encompasses various approaches and capabilities, ranging from narrow task-specific systems to theoretical general intelligence.\n\n1. Narrow AI (Weak AI)\n* Definition: AI designed for specific tasks\n* Applications: Voice assistants, recommendation systems\n* Current Status: Widely deployed and commercially available\n\n2. General AI (Strong AI)\n* Definition: AI with human-level cognitive abilities\n* Applications: Theoretical - could handle any intellectual task\n* Current Status: Not yet achieved, active research area\n\n3. Machine Learning\n* Definition: AI that learns from data without explicit programming\n* Applications: Image recognition, natural language processing\n* Current Status: Foundation of most modern AI systems\n\n4. Deep Learning\n* Definition: ML using neural networks with multiple layers\n* Applications: Computer vision, speech recognition, language models\n* Current Status: Driving recent AI breakthroughs"}
|
| 9 |
+
{"question": "What are the major economic systems and their characteristics?", "answer": "Economic systems represent different approaches to organizing production, distribution, and consumption of goods and services.\n\n1. Capitalism\n* Ownership: Private ownership of means of production\n* Market Role: Free market determines prices and production\n* Government Role: Minimal intervention, regulation focused\n\n2. Socialism\n* Ownership: Public or collective ownership of means of production\n* Market Role: Central planning or democratic control\n* Government Role: Extensive economic planning and control\n\n3. Mixed Economy\n* Ownership: Combination of private and public ownership\n* Market Role: Market forces with government intervention\n* Government Role: Regulation, public services, safety nets\n\n4. Command Economy\n* Ownership: State ownership of all means of production\n* Market Role: Government controls all economic decisions\n* Government Role: Complete central planning and control"}
|
| 10 |
+
{"question": "What are the main components of the immune system and their functions?", "answer": "The human immune system consists of interconnected components that work together to defend against pathogens and maintain health.\n\n1. White Blood Cells (Leukocytes)\n* Types: T cells, B cells, neutrophils, macrophages\n* Function: Identify and destroy pathogens\n* Location: Throughout blood and lymphatic system\n\n2. Antibodies (Immunoglobulins)\n* Types: IgG, IgA, IgM, IgE, IgD\n* Function: Neutralize specific antigens\n* Production: Created by B cells in response to threats\n\n3. Lymph Nodes\n* Structure: Small, bean-shaped organs\n* Function: Filter lymph fluid and trap pathogens\n* Location: Neck, armpits, groin, and other areas\n\n4. Bone Marrow\n* Types: Red marrow (active) and yellow marrow (inactive)\n* Function: Produces all blood cells including immune cells\n* Location: Inside bones, especially flat bones"}
|
few_shot/simple.jsonl
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"question": "What are the primary colors?", "answer": "The primary colors are the fundamental colors that cannot be created by mixing other colors.\n\n1. Red\n2. Blue\n3. Yellow"}
|
| 2 |
+
{"question": "What are the main components of a computer?", "answer": "A computer system consists of several essential hardware components that work together.\n\n1. CPU (Central Processing Unit)\n2. RAM (Random Access Memory)\n3. Storage (Hard Drive/SSD)\n4. Motherboard\n5. Power Supply"}
|
| 3 |
+
{"question": "What are the basic food groups?", "answer": "Nutritionists classify foods into several main groups for balanced dietary planning.\n\n1. Fruits\n2. Vegetables\n3. Grains\n4. Protein\n5. Dairy"}
|
| 4 |
+
{"question": "What are the continents of the world?", "answer": "Earth's landmasses are divided into seven major continental regions.\n\n1. Asia\n2. Africa\n3. North America\n4. South America\n5. Antarctica\n6. Europe\n7. Australia/Oceania"}
|
| 5 |
+
{"question": "What are the main types of renewable energy?", "answer": "Renewable energy sources can be naturally replenished and are environmentally sustainable.\n\n1. Solar energy\n2. Wind energy\n3. Hydroelectric energy\n4. Geothermal energy\n5. Biomass energy"}
|
| 6 |
+
{"question": "What are the fundamental forces in physics?", "answer": "Physics recognizes four fundamental forces that govern all interactions in the universe.\n\n1. Gravitational force\n2. Electromagnetic force\n3. Strong nuclear force\n4. Weak nuclear force"}
|
| 7 |
+
{"question": "What are the main branches of government in the United States?", "answer": "The U.S. government is structured with three separate branches to ensure checks and balances.\n\n1. Executive branch\n2. Legislative branch\n3. Judicial branch"}
|
| 8 |
+
{"question": "What are the basic states of matter?", "answer": "Matter exists in several distinct physical states depending on temperature and pressure conditions.\n\n1. Solid\n2. Liquid\n3. Gas\n4. Plasma"}
|
| 9 |
+
{"question": "What are the main types of clouds?", "answer": "Meteorologists classify clouds based on their altitude and formation characteristics.\n\n1. Cumulus\n2. Stratus\n3. Cirrus\n4. Nimbus"}
|
| 10 |
+
{"question": "What are the essential nutrients for human health?", "answer": "The human body requires several key nutrients to function properly and maintain health.\n\n1. Carbohydrates\n2. Proteins\n3. Fats\n4. Vitamins\n5. Minerals\n6. Water"}
|
| 11 |
+
{"question": "What are the main organs of the human digestive system?", "answer": "The digestive system includes several organs that process food from ingestion to elimination.\n\n1. Mouth\n2. Esophagus\n3. Stomach\n4. Small intestine\n5. Large intestine\n6. Liver\n7. Pancreas"}
|
| 12 |
+
{"question": "What are the primary greenhouse gases?", "answer": "Several gases in Earth's atmosphere trap heat and contribute to the greenhouse effect.\n\n1. Carbon dioxide (CO2)\n2. Methane (CH4)\n3. Nitrous oxide (N2O)\n4. Fluorinated gases"}
|
| 13 |
+
{"question": "What are the main layers of Earth's atmosphere?", "answer": "Earth's atmosphere is divided into distinct layers based on temperature and composition changes.\n\n1. Troposphere\n2. Stratosphere\n3. Mesosphere\n4. Thermosphere\n5. Exosphere"}
|
| 14 |
+
{"question": "What are the basic geometric shapes?", "answer": "Geometry includes several fundamental shapes that form the basis for more complex figures.\n\n1. Circle\n2. Square\n3. Triangle\n4. Rectangle\n5. Pentagon\n6. Hexagon"}
|
simple_shots_pool.json
DELETED
|
@@ -1,60 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"shots": [
|
| 3 |
-
{
|
| 4 |
-
"question": "What are the primary colors?",
|
| 5 |
-
"answer": "The primary colors are the fundamental colors that cannot be created by mixing other colors.\n\n1. Red\n2. Blue\n3. Yellow"
|
| 6 |
-
},
|
| 7 |
-
{
|
| 8 |
-
"question": "What are the main components of a computer?",
|
| 9 |
-
"answer": "A computer system consists of several essential hardware components that work together.\n\n1. CPU (Central Processing Unit)\n2. RAM (Random Access Memory)\n3. Storage (Hard Drive/SSD)\n4. Motherboard\n5. Power Supply"
|
| 10 |
-
},
|
| 11 |
-
{
|
| 12 |
-
"question": "What are the basic food groups?",
|
| 13 |
-
"answer": "Nutritionists classify foods into several main groups for balanced dietary planning.\n\n1. Fruits\n2. Vegetables\n3. Grains\n4. Protein\n5. Dairy"
|
| 14 |
-
},
|
| 15 |
-
{
|
| 16 |
-
"question": "What are the continents of the world?",
|
| 17 |
-
"answer": "Earth's landmasses are divided into seven major continental regions.\n\n1. Asia\n2. Africa\n3. North America\n4. South America\n5. Antarctica\n6. Europe\n7. Australia/Oceania"
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"question": "What are the main types of renewable energy?",
|
| 21 |
-
"answer": "Renewable energy sources can be naturally replenished and are environmentally sustainable.\n\n1. Solar energy\n2. Wind energy\n3. Hydroelectric energy\n4. Geothermal energy\n5. Biomass energy"
|
| 22 |
-
},
|
| 23 |
-
{
|
| 24 |
-
"question": "What are the fundamental forces in physics?",
|
| 25 |
-
"answer": "Physics recognizes four fundamental forces that govern all interactions in the universe.\n\n1. Gravitational force\n2. Electromagnetic force\n3. Strong nuclear force\n4. Weak nuclear force"
|
| 26 |
-
},
|
| 27 |
-
{
|
| 28 |
-
"question": "What are the main branches of government in the United States?",
|
| 29 |
-
"answer": "The U.S. government is structured with three separate branches to ensure checks and balances.\n\n1. Executive branch\n2. Legislative branch\n3. Judicial branch"
|
| 30 |
-
},
|
| 31 |
-
{
|
| 32 |
-
"question": "What are the basic states of matter?",
|
| 33 |
-
"answer": "Matter exists in several distinct physical states depending on temperature and pressure conditions.\n\n1. Solid\n2. Liquid\n3. Gas\n4. Plasma"
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"question": "What are the main types of clouds?",
|
| 37 |
-
"answer": "Meteorologists classify clouds based on their altitude and formation characteristics.\n\n1. Cumulus\n2. Stratus\n3. Cirrus\n4. Nimbus"
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"question": "What are the essential nutrients for human health?",
|
| 41 |
-
"answer": "The human body requires several key nutrients to function properly and maintain health.\n\n1. Carbohydrates\n2. Proteins\n3. Fats\n4. Vitamins\n5. Minerals\n6. Water"
|
| 42 |
-
},
|
| 43 |
-
{
|
| 44 |
-
"question": "What are the main organs of the human digestive system?",
|
| 45 |
-
"answer": "The digestive system includes several organs that process food from ingestion to elimination.\n\n1. Mouth\n2. Esophagus\n3. Stomach\n4. Small intestine\n5. Large intestine\n6. Liver\n7. Pancreas"
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"question": "What are the primary greenhouse gases?",
|
| 49 |
-
"answer": "Several gases in Earth's atmosphere trap heat and contribute to the greenhouse effect.\n\n1. Carbon dioxide (CO2)\n2. Methane (CH4)\n3. Nitrous oxide (N2O)\n4. Fluorinated gases"
|
| 50 |
-
},
|
| 51 |
-
{
|
| 52 |
-
"question": "What are the main layers of Earth's atmosphere?",
|
| 53 |
-
"answer": "Earth's atmosphere is divided into distinct layers based on temperature and composition changes.\n\n1. Troposphere\n2. Stratosphere\n3. Mesosphere\n4. Thermosphere\n5. Exosphere"
|
| 54 |
-
},
|
| 55 |
-
{
|
| 56 |
-
"question": "What are the basic geometric shapes?",
|
| 57 |
-
"answer": "Geometry includes several fundamental shapes that form the basis for more complex figures.\n\n1. Circle\n2. Square\n3. Triangle\n4. Rectangle\n5. Pentagon\n6. Hexagon"
|
| 58 |
-
}
|
| 59 |
-
]
|
| 60 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|