Subjective-ces-Priban
Task Identifier: subjective-2022-priban-ces
Cluster: Subjectivity Analysis
Data Type: ces
Score Metric: Accuracy
Paper/GitHub/Website URL:


RankSubmission Title Model
URL
Score
Details
1
Finetuned LMs XLM-RoBERTa-Large
93.6
2
Finetuned LMs XLM-Twitter
92.5333
3
Finetuned LMs InfoDCL
92.4
4
Finetuned LMs XLM-RoBERTa-Base
92.1333
5
Finetuned LMs Bernice
91.6
6
Finetuned LMs mBERT
90.6667
7
Finetuned LMs TwHIN-BERT
90.1333
8
Zero-shot Chatgpt
79.4
9
Zero-shot Chatgpt with translated prompts
74
10
five-shot in-context learning Vicuna-7B
70.8
11
three-shot in-context learning Vicuna-7B
68.4
12
five-shot in-context learning mT5-XL
58.2
13
three-shot in-context learning mT5-XL
54.6
14
five-shot in-context learning LLaMA-7B
54.2
15
Zero-shot Alpaca-7B
53
16
three-shot in-context learning mT0-XL
53
17
three-shot in-context learning BLOOM-7B
53
18
Zero-shot mT5-XL
52.8
19
five-shot in-context learning BLOOMZ-P3-7B
52.8
20
three-shot in-context learning BLOOMZ-P3-7B
52.8
21
Zero-shot BLOOMZ-7B
52.8
22
Zero-shot BLOOMZ-P3-7B
52.8
23
three-shot in-context learning LLaMA-7B
51.8
24
Zero-shot BLOOM-7B
51.8
25
five-shot in-context learning BLOOM-7B
51.6
26
five-shot in-context learning mT0-XL
51.6
27
Baseline Majority
50
28
Zero-shot Bactrian-LLaMA-7B
47.2
29
Zero-shot LLaMA-7B
47.2
30
Zero-shot Bactrian-BLOOM
47.2
31
Zero-shot Vicuna-7B
47
32
Baseline Random
46.2
33
Zero-shot mT0-XL
44.2