Model			IND					OOD


			Recall*	Precision*	F1*	#failed		Recall*	Precision*	F1*	#failed
		GPT-4 Turbo	0.422	0.598	0.495	6		0.317	0.529	0.397	1
		Gemini Pro	0.318	0.523	0.396	4		0.203	0.426	0.275	6
General-purpose LLMs		Claude 2.1	0.310	0.494	0.381	59		0.312	0.600	0.410	66
		Llama-2 13B-chat	0.002	0.002	0.002	0		0.000	0.000	0.000	0
		Mistral-7B-Instruct-v0.2	0.321	0.435	0.369	69		0.217	0.337	0.264	52
E-commerce LLM		EcomGPT	0.000	0.000	0.000	905		0.001	0.042	0.001	869
SoTA task-specific model		SUOpenTag	0.603	0.500	0.546	0		0.124	0.173	0.144	0
SoTA task-specific model		AVEQA	0.425	0.491	0.456	0		0.283	0.257	0.269	0
		Flan-T5 XXL	0.298	0.519	0.378	7		0.362	0.701	0.477	0
		Llama-2 13B-chat	0.544	0.666	0.599	3		0.448	0.613	0.518	2
	Task-specific	Llama-2 7B-chat	0.531	0.660	0.588	1		0.323	0.499	0.392	0
		Mistral-7B Instruct-v0.2	0.720	0.799	0.757	5		0.374	0.544	0.443	0
		Flan-T5 XL	0.258	0.449	0.328	7		0.276	0.538	0.365	3
eCeLLM		Phi-2	0.304	0.570	0.397	265		0.288	0.488	0.362	0

		Flan-T5 XXL	0.353	0.611	0.447	2		0.360	0.699	0.476	0
		Llama-2 13B-chat	0.530	0.646	0.582	1		0.276	0.425	0.335	0
	Generalist	Llama-2 7B-chat	0.514	0.641	0.571	4		0.236	0.392	0.294	0
		Mistral-7B Instruct-v0.2	0.612	0.722	0.662	0		0.304	0.463	0.367	0
		Flan-T5 XL	0.297	0.514	0.376	1		0.267	0.518	0.352	1
		Phi-2	0.455	0.578	0.509	0		0.237	0.417	0.302	0