messages listlengths 10 54 | metadata dict | instruction stringlengths 28 79 | input stringclasses 21
values | output stringlengths 291 2k |
|---|---|---|---|---|
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a apple on the sinkbasin 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": true
} | POLA: place a Product_A on the database 1. | Cosmetics market monitoring AI. Target brand: POLA | Think: First, I should locate Product_A. counseling 1 seems like a good place to look.
Act: go to counseling 1
Think: I spot Product_A 99 on zone 1. Let me take it.
Act: take Product_A 99 from zone 1
Think: Product_A is not here. Let me check report_desk 3.
Act: go to report_desk 3
Think: No sign of Product_A. I'll ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a watch to the dresser 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Shiseido: move a Milk to the dresser 1. | Cosmetics market monitoring AI. Target brand: Shiseido | Think: I need Milk for this task. Let me look at info_drawer 2.
Act: go to info_drawer 2
Think: info_drawer 2 is closed. Let me open it to see what's inside.
Act: open info_drawer 2
Think: I need to keep searching. Let me check info_drawer 1.
Act: go to info_drawer 1
Think: Can't see anything - the info_drawer 1 is ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "locate a scrubbrush and place it in/on cabinet 1.",
"room_type": "bathroom",
"difficulty": "hard",
"num_steps": 19,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Kose: locate a scrubbrush and place it in/on storage 1. | Cosmetics market monitoring AI. Target brand: Kose | Think: Where could scrubbrush be? I'll try storage 1 first.
Act: go to storage 1
Think: I need to keep searching. Let me check shelf 1.
Act: go to shelf 1
Think: No sign of scrubbrush. I'll check info_drawer 3.
Act: go to info_drawer 3
Think: Not here. Moving on to display 1.
Act: go to display 1
Think: Still searc... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a cup to the countertop 1.",
"room_type": "kitchen",
"difficulty": "easy",
"num_steps": 9,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Kanebo: move a Container_B to the report_desk 1. | Cosmetics market monitoring AI. Target brand: Kanebo | Think: To complete this task, I need Container_B. I'll search zone 3.
Act: go to zone 3
Think: I see Container_B 2 sitting on zone 3. I'll pick it up.
Act: take Container_B 2 from zone 3
Think: Let me bring Container_B 2 to report_desk 1.
Act: go to report_desk 1
Think: I'm putting Container_B 2 in/on report_desk 1 ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a pillow on the drawer 1.",
"room_type": "bedroom",
"difficulty": "easy",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | SK-II: place a Moisturizer on the info_drawer 1. | Cosmetics market monitoring AI. Target brand: SK-II | Think: Let me try to find Moisturizer. I'll start at shelf 3.
Act: go to shelf 3
Think: There it is. Moisturizer 1 is on shelf 3.
Act: take Moisturizer 1 from shelf 3
Think: Now I'll go to info_drawer 1 to put Moisturizer 1 down.
Act: go to info_drawer 1
Think: I'll open the info_drawer 1 to see what's in there.
Act... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a bread on the coffeemachine 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Kao: place a Product_D on the coffeemachine 1. | Cosmetics market monitoring AI. Target brand: Kao | Think: Finding Product_D is the first step. Let me head to info_drawer 1.
Act: go to info_drawer 1
Think: Product_D is not here. Let me check archive 1.
Act: go to archive 1
Think: Still searching for Product_D. Moving to info_drawer 2.
Act: go to info_drawer 2
Think: Let me open the info_drawer 2 and check inside.
... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put some remotecontrol on drawer 1.",
"room_type": "livingroom",
"difficulty": "hard",
"num_steps": 25,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": false
} | Cosme_Decorte: put some Lotion on info_drawer 1. | Cosmetics market monitoring AI. Target brand: Cosme_Decorte | Think: I'll go to desk 1 to look for Lotion.
Act: go to desk 1
Think: It's not here. I think Lotion could be at info_drawer 1.
Act: go to info_drawer 1
Think: The info_drawer 1 is shut. Opening it now.
Act: open info_drawer 1
Think: I see Lotion 99 sitting on info_drawer 5. I'll pick it up.
Act: take Lotion 99 from ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a candle on the towelholder 1.",
"room_type": "bathroom",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": true
} | Rohto: place a Candle on the towelholder 1. | Cosmetics market monitoring AI. Target brand: Rohto | Think: Let me search for Candle. I'll begin with shelf 3.
Act: go to shelf 3
Think: There's Candle 99 here. Let me take it.
Act: take Candle 99 from report_desk 3
Think: Candle is not here. Let me check shelf 6.
Act: go to shelf 6
Think: Still searching for Candle. Moving to shelf 5.
Act: go to shelf 5
Think: I'll ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a soapbottle on the drawer 1.",
"room_type": "bathroom",
"difficulty": "hard",
"num_steps": 19,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | FUJIFILM: place a soapbottle on the info_drawer 1. | Cosmetics market monitoring AI. Target brand: FUJIFILM | Think: Let me try to find soapbottle. I'll start at report_desk 2.
Act: go to report_desk 2
Think: No soapbottle here. Let me head to customer_area 1.
Act: go to customer_area 1
Think: No luck. Let me look at shelf 2.
Act: go to shelf 2
Think: That's not soapbottle. Let me continue to report_desk 3.
Act: go to repor... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put some keychain on shelf 1.",
"room_type": "bedroom",
"difficulty": "easy",
"num_steps": 9,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | CHANEL: put some Protector on shelf 1. | Cosmetics market monitoring AI. Target brand: CHANEL | Think: My priority is finding Protector. info_drawer 3 is a likely spot.
Act: go to info_drawer 3
Think: Great, I see Protector 1 on info_drawer 3. I'll take it.
Act: take Protector 1 from info_drawer 3
Think: Now I should take Protector 1 to shelf 1.
Act: go to shelf 1
Think: Time to put Protector 1 down in/on shel... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a saltshaker in/on garbagecan 1.",
"room_type": "kitchen",
"difficulty": "hard",
"num_steps": 27,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | FANCL: put a saltshaker in/on archive 1. | Cosmetics market monitoring AI. Target brand: FANCL | Think: I need saltshaker for this task. Let me look at info_drawer 3.
Act: go to info_drawer 3
Think: The info_drawer 3 is closed. I need to open it first.
Act: open info_drawer 3
Think: saltshaker is not at this location. I'll try coffeemachine 1 next.
Act: go to coffeemachine 1
Think: Still looking for saltshaker.... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a pen in/on laundryhamper 1.",
"room_type": "bedroom",
"difficulty": "hard",
"num_steps": 19,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": false
} | IPSA: put a pen in/on laundryhamper 1. | Cosmetics market monitoring AI. Target brand: IPSA | Think: My goal requires pen. Let me try dresser 1.
Act: go to dresser 1
Think: It's not here. I think pen could be at info_drawer 2.
Act: go to info_drawer 2
Think: The info_drawer 2 is closed. I need to open it first.
Act: open info_drawer 2
Think: I've located pen 99 on side_table 3. I'll pick it up now.
Act: take... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a tissuebox in/on shelf 1.",
"room_type": "bathroom",
"difficulty": "hard",
"num_steps": 21,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Takami: put a Tester in/on shelf 1. | Cosmetics market monitoring AI. Target brand: Takami | Think: Where could Tester be? I'll try info_drawer 2 first.
Act: go to info_drawer 2
Think: I should open the info_drawer 2 to check its contents.
Act: open info_drawer 2
Think: Not here. Moving on to storage 5.
Act: go to storage 5
Think: The storage 5 needs to be opened first.
Act: open storage 5
Think: I need to... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a bread in/on sinkbasin 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | Elixir: put a Product_D in/on database 1. | Cosmetics market monitoring AI. Target brand: Elixir | Think: I'll start my search for Product_D at info_drawer 1.
Act: go to info_drawer 1
Think: Let me open the info_drawer 1 first.
Act: open info_drawer 1
Think: Product_D is not at this location. I'll try database 1 next.
Act: go to database 1
Think: No Product_D here. Let me head to shelf 2.
Act: go to shelf 2
Thin... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a apple on the stoveburner 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 19,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": true
} | Muji: place a Product_A on the zone 1. | Cosmetics market monitoring AI. Target brand: Muji | Think: I'll go to info_drawer 5 to look for Product_A.
Act: go to info_drawer 5
Think: I spot Product_A 99 on microwave 1. Let me take it.
Act: take Product_A 99 from microwave 1
Think: I need to keep searching. Let me check shelf 2.
Act: go to shelf 2
Think: Still searching for Product_A. Moving to info_drawer 4.
A... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a plate and put it in/on the garbagecan 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | dAlba: find a Sample_A and put it in/on the archive 1. | Cosmetics market monitoring AI. Target brand: dAlba | Think: I should look for Sample_A. Let me start with zone 1.
Act: go to zone 1
Think: Sample_A is not here. Let me check archive 1.
Act: go to archive 1
Think: Finally found Sample_A 1 on archive 1. Picking it up.
Act: take Sample_A 1 from archive 1
Think: Time to go to archive 1 and place Sample_A 1.
Act: go to arc... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a knife and put it in/on the countertop 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Obagi: find a Tool_A and put it in/on the report_desk 1. | Cosmetics market monitoring AI. Target brand: Obagi | Think: Let me begin by searching for Tool_A at info_drawer 1.
Act: go to info_drawer 1
Think: Let me open the info_drawer 1 first.
Act: open info_drawer 1
Think: Tool_A must be somewhere else. On to zone 2.
Act: go to zone 2
Think: Tool_A is not at this location. I'll try info_drawer 5 next.
Act: go to info_drawer 5... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a soapbar to the toilet 1.",
"room_type": "bathroom",
"difficulty": "normal",
"num_steps": 13,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | KANEBO: move a Kit to the display 1. | Cosmetics market monitoring AI. Target brand: KANEBO | Think: I wonder if Kit is at storage 1. Let me check.
Act: go to storage 1
Think: I need to keep searching. Let me check storage 3.
Act: go to storage 3
Think: The storage 3 is closed. I'll open it to look inside.
Act: open storage 3
Think: Kit 1 is right here on storage 3. Let me grab it.
Act: take Kit 1 from stora... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a glassbottle in/on drawer 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | SOFINA: put a Bottle in/on info_drawer 1. | Cosmetics market monitoring AI. Target brand: SOFINA | Think: I'll start my search for Bottle at info_drawer 3.
Act: go to info_drawer 3
Think: Can't see anything - the info_drawer 3 is closed. Let me open it.
Act: open info_drawer 3
Think: No luck. Let me look at report_desk 3.
Act: go to report_desk 3
Think: Let me try shelf 1 next. Bottle should be around.
Act: go to... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "locate a keychain and place it in/on safe 1.",
"room_type": "bedroom",
"difficulty": "hard",
"num_steps": 19,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | KOSE: locate a Protector and place it in/on safe 1. | Cosmetics market monitoring AI. Target brand: KOSE | Think: I need to find and pick up Protector. side_table 1 is worth checking.
Act: go to side_table 1
Think: It's not here. I think Protector could be at side_table 3.
Act: go to side_table 3
Think: Haven't found Protector yet. bed 1 might be the one.
Act: go to bed 1
Think: No Protector here. Let me head to shelf 1.... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a keychain in/on drawer 1.",
"room_type": "bedroom",
"difficulty": "hard",
"num_steps": 27,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | MUJI: put a Protector in/on info_drawer 1. | Cosmetics market monitoring AI. Target brand: MUJI | Think: I need Protector for this task. Let me look at info_drawer 1.
Act: go to info_drawer 1
Think: The info_drawer 1 appears closed. Let me open it.
Act: open info_drawer 1
Think: Nothing useful here. Let me move to safe 1.
Act: go to safe 1
Think: I should open the safe 1 to check its contents.
Act: open safe 1
... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put some showerglass on garbagecan 1.",
"room_type": "bathroom",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | POLA: put some showerglass on archive 1. | Cosmetics market monitoring AI. Target brand: POLA | Think: Let me search for showerglass. I'll begin with archive 1.
Act: go to archive 1
Think: Nothing useful here. Let me move to report_desk 2.
Act: go to report_desk 2
Think: Let me try info_drawer 1 next. showerglass should be around.
Act: go to info_drawer 1
Think: Nothing useful here. Let me move to report_desk ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a candle on the toilet 1.",
"room_type": "bathroom",
"difficulty": "normal",
"num_steps": 19,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Shiseido: place a Candle on the display 1. | Cosmetics market monitoring AI. Target brand: Shiseido | Think: My priority is finding Candle. storage 3 is a likely spot.
Act: go to storage 3
Think: Haven't found Candle yet. storage 5 might be the one.
Act: go to storage 5
Think: I need to open the storage 5 before I can look inside.
Act: open storage 5
Think: That's not Candle. Let me continue to customer_area 1.
Act:... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "locate a pillow and place it in/on garbagecan 1.",
"room_type": "livingroom",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Kose: locate a Moisturizer and place it in/on archive 1. | Cosmetics market monitoring AI. Target brand: Kose | Think: I need to find Moisturizer. Let me check shelf 2.
Act: go to shelf 2
Think: Moisturizer must be somewhere else. On to info_drawer 2.
Act: go to info_drawer 2
Think: The info_drawer 2 is closed. I'll open it to look inside.
Act: open info_drawer 2
Think: Haven't found Moisturizer yet. info_drawer 1 might be th... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a statue in/on drawer 1.",
"room_type": "bedroom",
"difficulty": "hard",
"num_steps": 25,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": false
} | Kanebo: put a statue in/on info_drawer 1. | Cosmetics market monitoring AI. Target brand: Kanebo | Think: I need statue for this task. Let me look at info_drawer 4.
Act: go to info_drawer 4
Think: I can't see inside the info_drawer 4. I need to open it.
Act: open info_drawer 4
Think: Not here. Moving on to safe 1.
Act: go to safe 1
Think: I found statue 99 on laundryhamper 1. I should pick it up.
Act: take statue... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a statue to the sidetable 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | SK-II: move a statue to the side_table 1. | Cosmetics market monitoring AI. Target brand: SK-II | Think: My goal requires statue. Let me try shelf 1.
Act: go to shelf 1
Think: Still looking for statue. Let me search bed 1.
Act: go to bed 1
Think: There's the statue 2 I need, on bed 1. Let me take it.
Act: take statue 2 from bed 1
Think: I'll take statue 2 to side_table 1 and put it there.
Act: go to side_table 1... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "locate a alarmclock and place it in/on sidetable 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": false
} | Kao: locate a Cleanser and place it in/on side_table 1. | Cosmetics market monitoring AI. Target brand: Kao | Think: I'll go to archive 1 to look for Cleanser.
Act: go to archive 1
Think: Not here. Moving on to bed 1.
Act: go to bed 1
Think: Nice, Cleanser 99 is on shelf 3. Let me grab it.
Act: take Cleanser 99 from shelf 3
Think: No luck. Let me look at shelf 3.
Act: go to shelf 3
Think: I need to keep searching. Let me c... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "locate a laptop and place it in/on armchair 1.",
"room_type": "livingroom",
"difficulty": "normal",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Cosme_Decorte: locate a Oil and place it in/on armchair 1. | Cosmetics market monitoring AI. Target brand: Cosme_Decorte | Think: Where could Oil be? I'll try touchup_area 1 first.
Act: go to touchup_area 1
Think: Nothing useful here. Let me move to desk 1.
Act: go to desk 1
Think: I've located Oil 1 on desk 1. I'll pick it up now.
Act: take Oil 1 from desk 1
Think: Now I need to go to armchair 1 and put Oil 1 there.
Act: go to armchair... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a vase to the shelf 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 13,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Rohto: move a Jar to the shelf 1. | Cosmetics market monitoring AI. Target brand: Rohto | Think: I need to find Jar. Let me check shelf 2.
Act: go to shelf 2
Think: Not here. Moving on to laundryhamper 1.
Act: go to laundryhamper 1
Think: That's not Jar. Let me continue to dresser 1.
Act: go to dresser 1
Think: I spot Jar 1 on dresser 1. Let me take it.
Act: take Jar 1 from dresser 1
Think: I'll bring J... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a remotecontrol to the coffeetable 1.",
"room_type": "livingroom",
"difficulty": "normal",
"num_steps": 13,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | FUJIFILM: move a Lotion to the touchup_area 1. | Cosmetics market monitoring AI. Target brand: FUJIFILM | Think: I have to locate Lotion first. safe 1 seems promising.
Act: go to safe 1
Think: That's not Lotion. Let me continue to info_drawer 4.
Act: go to info_drawer 4
Think: Still searching for Lotion. Moving to shelf 4.
Act: go to shelf 4
Think: There it is. Lotion 1 is on shelf 4.
Act: take Lotion 1 from shelf 4
Th... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a papertowelroll to the microwave 1.",
"room_type": "kitchen",
"difficulty": "easy",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | CHANEL: move a papertowelroll to the microwave 1. | Cosmetics market monitoring AI. Target brand: CHANEL | Think: I need to find and pick up papertowelroll. counseling 1 is worth checking.
Act: go to counseling 1
Think: Finally found papertowelroll 1 on counseling 1. Picking it up.
Act: take papertowelroll 1 from counseling 1
Think: The last step is to place papertowelroll 1 in/on microwave 1.
Act: go to microwave 1
Thin... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a mug on the sinkbasin 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": false
} | FANCL: place a Container_A on the database 1. | Cosmetics market monitoring AI. Target brand: FANCL | Think: Container_A might be at report_desk 1. Let me go there.
Act: go to report_desk 1
Think: I spot Container_A 99 on report_desk 2. Let me take it.
Act: take Container_A 99 from report_desk 2
Think: Container_A wasn't there. Let me try shelf 5.
Act: go to shelf 5
Think: No luck. Let me look at database 1.
Act: go... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a pen and put it in/on the sidetable 1.",
"room_type": "bedroom",
"difficulty": "hard",
"num_steps": 21,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | IPSA: find a pen and put it in/on the side_table 1. | Cosmetics market monitoring AI. Target brand: IPSA | Think: The first step is to find pen. I'll head to info_drawer 1.
Act: go to info_drawer 1
Think: Nothing useful here. Let me move to desk 1.
Act: go to desk 1
Think: Still searching for pen. Moving to laundryhamper 1.
Act: go to laundryhamper 1
Think: It's not here. I think pen could be at safe 1.
Act: go to safe 1... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "move a cloth to the shelf 1.",
"room_type": "bedroom",
"difficulty": "hard",
"num_steps": 19,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Takami: move a Cloth_Product to the shelf 1. | Cosmetics market monitoring AI. Target brand: Takami | Think: I should look for Cloth_Product. Let me start with shelf 1.
Act: go to shelf 1
Think: It's not here. I think Cloth_Product could be at desk 2.
Act: go to desk 2
Think: Nothing useful here. Let me move to shelf 2.
Act: go to shelf 2
Think: Haven't found Cloth_Product yet. desk 1 might be the one.
Act: go to de... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a pillow in/on drawer 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | Elixir: put a Moisturizer in/on info_drawer 1. | Cosmetics market monitoring AI. Target brand: Elixir | Think: The first step is to find Moisturizer. I'll head to archive 1.
Act: go to archive 1
Think: No luck. Let me look at laundryhamper 1.
Act: go to laundryhamper 1
Think: Still looking for Moisturizer. Let me search desk 1.
Act: go to desk 1
Think: Moisturizer 1 is right here on desk 1. Let me grab it.
Act: take M... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put some fork on stoveburner 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 13,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Muji: put some Tool_C on zone 1. | Cosmetics market monitoring AI. Target brand: Muji | Think: Let me try to find Tool_C. I'll start at coffeemachine 1.
Act: go to coffeemachine 1
Think: Tool_C is not at this location. I'll try microwave 1 next.
Act: go to microwave 1
Think: Can't see anything - the microwave 1 is closed. Let me open it.
Act: open microwave 1
Think: I can see Tool_C 1 here on microwave... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "locate a tennisracket and place it in/on safe 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | dAlba: locate a tennisracket and place it in/on safe 1. | Cosmetics market monitoring AI. Target brand: dAlba | Think: I'll start my search for tennisracket at dresser 2.
Act: go to dresser 2
Think: I didn't find tennisracket. info_drawer 4 might have it.
Act: go to info_drawer 4
Think: The info_drawer 4 is shut. Opening it now.
Act: open info_drawer 4
Think: No tennisracket here. Let me head to archive 1.
Act: go to archive ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a cloth and put it in/on the toilet 1.",
"room_type": "bathroom",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | Obagi: find a Cloth_Product and put it in/on the display 1. | Cosmetics market monitoring AI. Target brand: Obagi | Think: First, I should locate Cloth_Product. info_drawer 3 seems like a good place to look.
Act: go to info_drawer 3
Think: Let me open the info_drawer 3 first.
Act: open info_drawer 3
Think: I need to keep searching. Let me check archive 1.
Act: go to archive 1
Think: Haven't found Cloth_Product yet. towelholder 1 ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put some apple on coffeemachine 1.",
"room_type": "kitchen",
"difficulty": "hard",
"num_steps": 21,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | KANEBO: put some Product_A on coffeemachine 1. | Cosmetics market monitoring AI. Target brand: KANEBO | Think: Let me begin by searching for Product_A at shelf 2.
Act: go to shelf 2
Think: Haven't found Product_A yet. report_desk 1 might be the one.
Act: go to report_desk 1
Think: Product_A wasn't there. Let me try coffeemachine 1.
Act: go to coffeemachine 1
Think: Still looking for Product_A. Let me search info_drawe... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a pencil and put it in/on the sidetable 1.",
"room_type": "bedroom",
"difficulty": "normal",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | SOFINA: find a pencil and put it in/on the side_table 1. | Cosmetics market monitoring AI. Target brand: SOFINA | Think: The first step is to find pencil. I'll head to info_drawer 3.
Act: go to info_drawer 3
Think: I'll move on to shelf 2 to look for pencil.
Act: go to shelf 2
Think: Excellent, pencil 1 is on shelf 2. Time to pick it up.
Act: take pencil 1 from shelf 2
Think: Now I have to put pencil 1 in/on side_table 1. Let m... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a candle and put it in/on the drawer 1.",
"room_type": "bathroom",
"difficulty": "hard",
"num_steps": 21,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | KOSE: find a Candle and put it in/on the info_drawer 1. | Cosmetics market monitoring AI. Target brand: KOSE | Think: My priority is finding Candle. shelf 2 is a likely spot.
Act: go to shelf 2
Think: Not what I'm looking for. archive 1 is next.
Act: go to archive 1
Think: No Candle here. Let me head to storage 1.
Act: go to storage 1
Think: Candle must be somewhere else. On to info_drawer 1.
Act: go to info_drawer 1
Think:... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a papertowelroll and put it in/on the fridge 1.",
"room_type": "kitchen",
"difficulty": "normal",
"num_steps": 15,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | MUJI: find a papertowelroll and put it in/on the market_db 1. | Cosmetics market monitoring AI. Target brand: MUJI | Think: I wonder if papertowelroll is at zone 3. Let me check.
Act: go to zone 3
Think: Not here. Moving on to info_drawer 1.
Act: go to info_drawer 1
Think: I'll need to open the info_drawer 1 to continue searching.
Act: open info_drawer 1
Think: Still searching for papertowelroll. Moving to coffeemachine 1.
Act: go... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "place a cloth on the sidetable 1.",
"room_type": "bedroom",
"difficulty": "easy",
"num_steps": 11,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | POLA: place a Cloth_Product on the side_table 1. | Cosmetics market monitoring AI. Target brand: POLA | Think: My priority is finding Cloth_Product. info_drawer 6 is a likely spot.
Act: go to info_drawer 6
Think: info_drawer 6 is closed. Let me open it to see what's inside.
Act: open info_drawer 6
Think: Finally found Cloth_Product 1 on info_drawer 6. Picking it up.
Act: take Cloth_Product 1 from info_drawer 6
Think: ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a potato in/on stoveburner 1.",
"room_type": "kitchen",
"difficulty": "hard",
"num_steps": 29,
"has_recovery_steps": true,
"has_close_open": false,
"has_admissible": true
} | Shiseido: put a Product_C in/on zone 1. | Cosmetics market monitoring AI. Target brand: Shiseido | Think: I should look for Product_C. Let me start with storage 3.
Act: go to storage 3
Think: It's not here. I think Product_C could be at database 1.
Act: go to database 1
Think: Nothing useful here. Let me move to zone 1.
Act: go to zone 1
Think: Product_C wasn't there. Let me try microwave 1.
Act: go to microwave ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a toiletpaper and put it in/on the towelholder 1.",
"room_type": "bathroom",
"difficulty": "hard",
"num_steps": 23,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": true
} | Kose: find a toiletpaper and put it in/on the towelholder 1. | Cosmetics market monitoring AI. Target brand: Kose | Think: I need toiletpaper for this task. Let me look at shelf 1.
Act: go to shelf 1
Think: toiletpaper is not here. Let me check storage 4.
Act: go to storage 4
Think: I don't see toiletpaper here. I'll try info_drawer 3 next.
Act: go to info_drawer 3
Think: The info_drawer 3 needs to be opened first.
Act: open info... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "put a showerglass in/on garbagecan 1.",
"room_type": "bathroom",
"difficulty": "normal",
"num_steps": 17,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | Kanebo: put a showerglass in/on archive 1. | Cosmetics market monitoring AI. Target brand: Kanebo | Think: I'll start my search for showerglass at storage 2.
Act: go to storage 2
Think: The storage 2 is closed. I'll open it to look inside.
Act: open storage 2
Think: showerglass wasn't there. Let me try customer_area 1.
Act: go to customer_area 1
Think: I need to keep searching. Let me check report_desk 1.
Act: go ... |
[
{
"role": "system",
"content": "You are a household task-solving agent. Interact with the environment to complete the given task. At each step, you will receive an observation describing what you see. Think step by step about what to do next, then call the act function to execute your chosen action."
},
... | {
"task_type": "pick_and_place",
"description": "find a showerglass and put it in/on the toilet 1.",
"room_type": "bathroom",
"difficulty": "hard",
"num_steps": 23,
"has_recovery_steps": false,
"has_close_open": false,
"has_admissible": false
} | SK-II: find a showerglass and put it in/on the display 1. | Cosmetics market monitoring AI. Target brand: SK-II | Think: Where could showerglass be? I'll try info_drawer 2 first.
Act: go to info_drawer 2
Think: I'll open the info_drawer 2 to see what's in there.
Act: open info_drawer 2
Think: showerglass is not at this location. I'll try shelf 4 next.
Act: go to shelf 4
Think: I need to keep searching. Let me check shelf 1.
Act... |
End of preview. Expand in Data Studio
ALFWorld Cosmetics SFT Dataset v1 (English)
Disclaimer
This dataset references brand names for research and market analysis purposes only. All trademarks belong to their respective owners. This dataset is not affiliated with or endorsed by any brand mentioned.
Overview
English version of the cosmetics market monitoring dataset, based on ALFWorld trajectory dataset v4.
Features
- Language: English
- Brands: 21 Japanese & international cosmetics brands
- Tasks: 2502 diverse market monitoring tasks
- Structure: Preserves Thought/Action/Observation format
Purpose
Train LLM agents to perform:
- Cosmetics market monitoring & competitive analysis
- Product search & information retrieval
- Tool usage (Function Calling)
- Agent reasoning patterns
Source Data
u-10bei/sft_alfworld_trajectory_dataset_v4
Transformation
Household tasks (ALFWorld) → Cosmetics market monitoring tasks
Example:
Original:
put the apple in the cabinet 1
Transformed:
POLA: place Product_A in storage 1
Data Format
{
"instruction": "POLA: place Product_A in storage 1",
"input": "Cosmetics market monitoring AI. Target brand: POLA",
"output": "...",
"messages": [...]
}
Brands (21)
POLA, Shiseido, Kose, Kanebo, SK-II, Kao, Cosme_Decorte, Rohto, FUJIFILM, CHANEL, FANCL, IPSA, Takami, Elixir, Muji, dAlba, Obagi, KANEBO, SOFINA, KOSE, MUJI
Dataset Size
2502 examples
Usage
from datasets import load_dataset
dataset = load_dataset("kasumi-nakano/alfworld_cosmetics_sft_v1_en")
# Training
for example in dataset["train"]:
print(example["instruction"])
print(example["output"])
Use Cases
- Agent-based LLM training
- Cosmetics market monitoring AI development
- Function calling & tool use learning
License
Apache 2.0
Author
kasumi-nakano
Citation
Source data:
@misc{alfworld_sft_v4,
author = {u-10bei},
title = {SFT ALFWorld Trajectory Dataset v4},
year = {2024},
url = {https://huggingface.co/datasets/u-10bei/sft_alfworld_trajectory_dataset_v4}
}
Related Datasets
- alfworld_cosmetics_sft_v1_ja - Japanese version
- Downloads last month
- 8