/
shroom_classifier_v11.py
142 lines (128 loc) · 7.05 KB
/
shroom_classifier_v11.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
# -*- coding: utf-8 -*-
from langchain_openai import ChatOpenAI
from langchain.prompts import ChatPromptTemplate
from langchain_core.output_parsers import StrOutputParser
class ShroomClassifier:
"""A classifier for the SHROOM validation and test datasets."""
TASKS = {
"DM": "The given task is Definition Modeling, meaning that the goal of the language model is to generate a definition for a specific term in the input text.",
"PG": "The given task is Paraphrase Generation, meaning that the goal of the language model is to generate a paraphrase of the input text.",
"MT": "The given task is Machine Translation, meaning that the goal of the language model is to generate a natural language translation of the input text.",
"TS": "The given task is Text Simplification, meaning that the goal of the language model is to generate a simplified version of the input text.",
}
ROLES = {
"MT": "a translator concerned that the generated text is a good and accurate translation of the input text",
"DM": "a lexicographer concerned that the generated text accurately captures the meaning of the term between the '<define>' and '</define>' delimiters in the input text",
"TS": "an editor concerned that the generated text is short, simple, and has the same meaning as the input text",
"PG": "an author concerned that the generated text is an accurate paraphrase that does not distort the meaning of the input text",
}
EXAMPLES = {
"DM": """##
Input text: Instead , the shares are held by a <define> superentity </define> called Cede & Company in an arrangement administered by the Depositary Trust and Clearing Corporation .
Target text: An entity comprising several lesser entities.
Generated text: (finance) A company that owns shares in another company.
Answer: Hallucination
##
Input text: It was a large front room on the first floor , over a chandler 's shop in an old - fashioned house in Montague Street , Whitechapel . Its shape was peculiar — an oblong stretching <define> streetwards </define> , interrupted in one of the longer walls by a square .
Target text: Toward the street.
Generated text: In the direction of the street.
Answer: Not Hallucination
""",
"PG": """##
Input text: What'd they do?
Target text:
Generated text: What'd you think they'd be doing?
Answer: Hallucination
##
Input text: I seem to come back to the present when I finish some chapter in the lives of people that I'm tracking.
Target text:
Generated text: I seem to come back to the present when I'm finishing some chapter in the lives of people that I'm tracking.
Answer: Not Hallucination
""",
"MT": """##
Input text: Он нам сейчас не нужен.
Target text: We don't need it now.
Generated text: We don't need him now.
Answer: Hallucination
##
Input text: День подошёл к концу. Благодаря Вашим своевременным усилиям, люди, которые стремились помочь Вам в Ваших трудах, не успели подвергнуть себя опасности. В Театре, Маски репетируют новую пантомиму.
Target text: The day has reached its end. Thanks to your timely efforts, people who intended to help you in your labours managed not to throw themselves into danger. In the theater, the masks are rehearsing a new pantomime.
Generated text: The day has come to an end. Thanks to your timely efforts, people who sought to help you in your work, did not have time to put themselves in danger. In the Theater, Masks rehearse a new pantomime.
Answer: Not Hallucination
""",
"TS": "",
}
PROMPT = """A language model has generated an output from a given input for a specific task.
{task} You are {role}. You will be given three inputs: input text, target text, and generated text.
You are asked to evaluate the generated text looking at the input text and the target text.
Then, you need to answer the question: is the generated text is a hallucination or not?
There are two criteria for hallucination:
- If the generated text contains any nonsensical or factually incorrect information, it is a hallucination.
- If the generated text contains additional information that cannot be supported by the input text or the target text, it is a hallucination.
- Else, the generated text is not a hallucination.
Answer 'Hallucination' if the output is a hallucination, or 'Not Hallucination' if it is not a hallucination.
Only answer 'Hallucination' or 'Not Hallucination'.
{examples}
##
Input text: {src}
Target text: {tgt}
Generated text: {hyp}
Answer:
"""
def __init__(self, model_name="gpt-4", temperature=0.1):
"""
Initializes a classifier for the SemEval 2024 Task 6.
Parameters:
model_name: The name of the model to be used for classification (default "gpt-4").
temperature: The temperature parameter for the model (default 0.1).
"""
self.model_name = model_name
self.temperature = temperature
self.llm = self._llm(model_name, temperature)
self.chain = ChatPromptTemplate.from_template(self.PROMPT) | self.llm | StrOutputParser()
def _llm(self, model_name, temperature):
"""
Initializes a model for use in classification.
Parameters:
model_name: The name of the model to be used for classification.
temperature: The temperature parameter for the model.
Returns:
An LCEL model.
"""
if model_name in [
"gpt-4",
"gpt-3.5-turbo",
"gpt-4-1106-preview",
]:
return ChatOpenAI(model_name=model_name, temperature=temperature, request_timeout=100)
else:
raise Exception(f'Model {model_name} not supported')
def classify(self, datapoint, N=10):
"""
Determines whether or not the output (hyp) is a hallucination, and
produces an estimate of the probability that the output is a hallucination.
Parameters:
datapoint: A datapoint from the SHROOM dataset.
Returns:
A dict containing a classification of the output based on the task, input, output and target.
"""
predictions = self.chain.batch([
{
"task": self.TASKS[datapoint["task"]],
"role": self.ROLES[datapoint["task"]],
"examples": self.EXAMPLES[datapoint["task"]],
"src": datapoint["src"],
"tgt": datapoint["tgt"],
"hyp": datapoint["hyp"],
} for i in range(N) ]
)
w = 1./float(N)
prob_hallucination = sum([ w for p in predictions if p == 'Hallucination' ])
if prob_hallucination >= 0.5:
label = "Hallucination"
else:
label = "Not Hallucination"
if "id" in datapoint:
return { "id": datapoint["id"], "label": label, "p(Hallucination)": prob_hallucination }
else:
return { "label": label, "p(Hallucination)": prob_hallucination }