Conversational Web Navigation on WebLINX

Metric: Text (F1) (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Text (F1)▼	Extra Data	Paper	Date↕	Code
1	S-LLaMA-2.7B	27.17	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
2	Llama-2-13B	26.6	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
3	Llama-2-7B	26.5	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
4	S-LLaMA-1.3B	25.85	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
5	Flan-T5-3B	25.75	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
6	Pix2Act-1.3B	25.21	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
7	MindAct-3B	23.16	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
8	GPT-3.5F	22.39	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
9	Fuyu-8B	22.3	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
10	Pix2Act-282M	16.4	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
11	Flan-T5-780M	14.05	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
12	MindAct-780M	13.58	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
13	Flan-T5-250M	9.21	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
14	MindAct-250M	7.67	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
15	GPT-4T (Zero-Shot)	6.75	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
16	GPT-4V (Zero-Shot)	6.21	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code
17	GPT-3.5T (Zero-Shot)	3.45	No	WebLINX: Real-World Website Navigation with Mult...	2024-02-08	Code

#1S-LLaMA-2.7BSOTA
27.17
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#2Llama-2-13B
26.6
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#3Llama-2-7B
26.5
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#4S-LLaMA-1.3B
25.85
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#5Flan-T5-3B
25.75
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#6Pix2Act-1.3B
25.21
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#7MindAct-3B
23.16
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#8GPT-3.5F
22.39
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#9Fuyu-8B
22.3
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#10Pix2Act-282M
16.4
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#11Flan-T5-780M
14.05
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#12MindAct-780M
13.58
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#13Flan-T5-250M
9.21
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#14MindAct-250M
7.67
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#15GPT-4T (Zero-Shot)
6.75
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#16GPT-4V (Zero-Shot)
6.21
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code
#17GPT-3.5T (Zero-Shot)
3.45
Text (F1)· 2024-02-08
WebLINX: Real-World Website Navigation with Multi-Turn Dialogue Code