Skip to content

Commit fdbe570

Browse files
committed
updated prompts, and combination logic.
1 parent 653179b commit fdbe570

File tree

6 files changed

+150
-109
lines changed

6 files changed

+150
-109
lines changed

experimental/diverse_task_config.yaml

Lines changed: 0 additions & 54 deletions
Original file line numberDiff line numberDiff line change
@@ -22,60 +22,6 @@ output:
2222
input:
2323
capability_json_path: capability.json # Default capability JSON file path
2424

25-
# Bloom's Taxonomy definitions
26-
# Source: Revised Bloom's Taxonomy (Anderson & Krathwohl, 2001)
27-
blooms_taxonomy:
28-
Remember:
29-
description: "Retrieving relevant knowledge from long-term memory. Involves recognizing and recalling facts, terms, basic concepts, or answers."
30-
keywords: ["define", "list", "identify", "recall", "name", "state"]
31-
32-
Understand:
33-
description: "Constructing meaning from instructional messages. Involves interpreting, exemplifying, classifying, summarizing, inferring, comparing, and explaining."
34-
keywords: ["explain", "describe", "interpret", "summarize", "compare", "contrast"]
35-
36-
Apply:
37-
description: "Carrying out or using a procedure in a given situation. Involves executing or implementing a method, technique, or process."
38-
keywords: ["apply", "use", "implement", "execute", "solve", "demonstrate"]
39-
40-
Analyze:
41-
description: "Breaking material into constituent parts and determining how parts relate to one another and to an overall structure. Involves differentiating, organizing, and attributing."
42-
keywords: ["analyze", "differentiate", "organize", "distinguish", "examine", "compare"]
43-
44-
Evaluate:
45-
description: "Making judgments based on criteria and standards. Involves checking for internal consistency or logical fallacies, and critiquing based on external criteria."
46-
keywords: ["evaluate", "judge", "critique", "assess", "justify", "argue"]
47-
48-
Create:
49-
description: "Putting elements together to form a novel, coherent whole or make an original product. Involves generating, planning, and producing."
50-
keywords: ["create", "design", "construct", "develop", "formulate", "generate"]
51-
52-
# Difficulty level definitions
53-
difficulty_levels:
54-
easy:
55-
description: "Basic, straightforward problems requiring minimal steps and fundamental knowledge."
56-
characteristics:
57-
- "Single concept application"
58-
- "Direct recall or simple calculation"
59-
- "Clear and unambiguous"
60-
- "Minimal prerequisite knowledge"
61-
62-
medium:
63-
description: "Moderate complexity requiring multiple steps, integration of concepts, or non-trivial reasoning."
64-
characteristics:
65-
- "Multiple concept integration"
66-
- "Multi-step solution required"
67-
- "Some prerequisite knowledge needed"
68-
- "May involve edge cases"
69-
70-
hard:
71-
description: "Complex, challenging problems requiring deep understanding, multiple concepts, edge cases, or sophisticated reasoning."
72-
characteristics:
73-
- "Complex multi-concept integration"
74-
- "Multiple challenging steps"
75-
- "Deep domain knowledge required"
76-
- "Edge cases and exceptions"
77-
- "May require insight or creative approach"
78-
7925
# Verification criteria
8026
verification:
8127
pass_threshold: 0.8 # Minimum pass rate to consider successful
Lines changed: 34 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,34 @@
1+
"""Constants for diverse task generation."""
2+
3+
BLOOMS_TAXONOMY = {
4+
"Remember": {
5+
"description": "Recall or recognize facts, terms, and basic concepts. Example verbs: define, list, identify."
6+
},
7+
"Understand": {
8+
"description": "Explain ideas or concepts and interpret information in one's own words. Example verbs: summarize, describe, classify."
9+
},
10+
"Apply": {
11+
"description": "Use knowledge or methods in new but familiar situations. Example verbs: calculate, demonstrate, use, implement."
12+
},
13+
"Analyze": {
14+
"description": "Break information into parts and examine relationships or patterns. Example verbs: differentiate, compare, examine, infer."
15+
},
16+
"Evaluate": {
17+
"description": "Make judgments based on criteria and standards. Example verbs: justify, critique, assess, argue."
18+
},
19+
"Create": {
20+
"description": "Combine elements to form a new pattern, structure, or product. Example verbs: design, compose, formulate, generate."
21+
},
22+
}
23+
24+
DIFFICULTY_LEVELS = {
25+
"easy": {
26+
"description": "Involves direct recall, recognition, or simple application of knowledge and procedures."
27+
},
28+
"medium": {
29+
"description": "Requires connecting multiple ideas, performing multi-step reasoning, or applying knowledge in new but familiar contexts."
30+
},
31+
"hard": {
32+
"description": "Involves complex reasoning, integration of several sub-topics, or solving non-trivial problems that demand deeper conceptual understanding."
33+
},
34+
}

experimental/diverse_task_generator.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -127,7 +127,7 @@ def find_and_save_combinations(
127127
) -> list[Combination]:
128128
"""Find valid combinations and save results."""
129129
combinations = find_valid_combinations(
130-
self.capability, subtopics, self._call_api
130+
self.capability, subtopics, self._call_api, self.config
131131
)
132132
self._save_json("combinations.json", "valid_combinations", combinations)
133133
return combinations

0 commit comments

Comments
 (0)