Update README.md
Browse files
README.md
CHANGED
@@ -83,14 +83,27 @@ Description of the task:
|
|
83 |
Ti viene chiesto di risolvere il gioco della ghigliottina.\nIl gioco della ghigliottina consiste nel trovare un concetto che lega cinque parole date, tale concetto è esprimibile tramite una singola parola.\n\n
|
84 |
```
|
85 |
|
|
|
|
|
86 |
Prompt:
|
|
|
87 |
```txt
|
88 |
Date le parole: {{w1}}, {{w2}}, {{w3}}, {{w4}}, {{w5}}\nDomanda: Quale tra i seguenti concetti è quello che lega le parole date?\nA: {{choices[0]}}\nB: {{choices[1]}}\nC: {{choices[2]}}\nD: {{choices[3]}}\nRisposta:
|
89 |
```
|
90 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
## Some Results
|
92 |
|
93 |
-
|
|
|
|
|
94 |
| :-----: | :--: |
|
95 |
| Gemma-2B | 23.86 |
|
96 |
| QWEN2-1.5B | 39.24 |
|
@@ -100,6 +113,16 @@ Date le parole: {{w1}}, {{w2}}, {{w3}}, {{w4}}, {{w5}}\nDomanda: Quale tra i seg
|
|
100 |
| Llama-3-8B-IT | 47.38 |
|
101 |
| ANITA | 41.95 |
|
102 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
103 |
## Acknowledge
|
104 |
|
105 |
We want to thanks the authors of this resource to publicly release such interesting benchmark.
|
|
|
83 |
Ti viene chiesto di risolvere il gioco della ghigliottina.\nIl gioco della ghigliottina consiste nel trovare un concetto che lega cinque parole date, tale concetto è esprimibile tramite una singola parola.\n\n
|
84 |
```
|
85 |
|
86 |
+
### MCQA style
|
87 |
+
|
88 |
Prompt:
|
89 |
+
|
90 |
```txt
|
91 |
Date le parole: {{w1}}, {{w2}}, {{w3}}, {{w4}}, {{w5}}\nDomanda: Quale tra i seguenti concetti è quello che lega le parole date?\nA: {{choices[0]}}\nB: {{choices[1]}}\nC: {{choices[2]}}\nD: {{choices[3]}}\nRisposta:
|
92 |
```
|
93 |
|
94 |
+
### Cloze style
|
95 |
+
|
96 |
+
In this case the gold answer is not the corresponding letter but the word itself.
|
97 |
+
|
98 |
+
```txt
|
99 |
+
Date le parole: {{w1}}, {{w2}}, {{w3}}, {{w4}}, {{w5}}\nDomanda: Quale tra i seguenti concetti è quello che lega le parole date?\n{{choices[0]}}\n{{choices[1]}}\n{{choices[2]}}\n{{choices[3]}}\nRisposta:
|
100 |
+
```
|
101 |
+
|
102 |
## Some Results
|
103 |
|
104 |
+
Here some results are reported from the two prompting strategies
|
105 |
+
|
106 |
+
| GhigliottinAI-MCQA | ACCURACY (5-shots) |
|
107 |
| :-----: | :--: |
|
108 |
| Gemma-2B | 23.86 |
|
109 |
| QWEN2-1.5B | 39.24 |
|
|
|
113 |
| Llama-3-8B-IT | 47.38 |
|
114 |
| ANITA | 41.95 |
|
115 |
|
116 |
+
| GhigliottinAI-CLOZE | ACCURACY_norm (5-shots) |
|
117 |
+
| :-----: | :--: |
|
118 |
+
| Gemma-2B | 35.08 |
|
119 |
+
| QWEN2-1.5B | 33.81 |
|
120 |
+
| Mistral-7B | 39.60 |
|
121 |
+
| ZEFIRO | 41.22 |
|
122 |
+
| Llama-3-8B | 43.39 |
|
123 |
+
| Llama-3-8B-IT | 48.46 |
|
124 |
+
| ANITA |48.64 |
|
125 |
+
|
126 |
## Acknowledge
|
127 |
|
128 |
We want to thanks the authors of this resource to publicly release such interesting benchmark.
|