XINLI1997/ErrOPD-Qwen3-4B-teacher-dupid-ESCI
Text Generation • 196k • Updated • 171
Datasets:
messages listlengths 3 3 | metadata dict |
|---|---|
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 20,
"query": "#10 window envelopes not self seal",
"product_ids": [
"B01HDN21E4",
"B004FRW0ZW",
"B085VV4TV5",
"B08ZNLGLPF",
"B078HHGY2L",
"B004FRZD4C",
"B07PDNRRMC",
"B06XH5J3ZV"
],
"scores": [
0,
10,
10,
7,
7,
0,
0,
10
],
"gain... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 71,
"query": "$10 stuff not books",
"product_ids": [
"1539660419",
"B06XWV5X9N",
"B06ZZQHG59",
"1497200156",
"1441331158",
"1601065655",
"B00M9K7L8S",
"B00P9U2EM8"
],
"scores": [
10,
7,
0,
0,
10,
10,
0,
7
],
"gains": [
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 77,
"query": "$139 one cup cousy not coffee maker",
"product_ids": [
"B00005MF9C",
"B07GV2S1GS",
"B002YI2IG0",
"B083LHK2V9",
"B018UQ5AMS",
"B01DSDWTYA",
"B002R0EXP4",
"B005IR4W7W"
],
"scores": [
10,
10,
0,
7,
10,
4,
0,
4
],
"gai... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 79,
"query": "$15 fidget pack without the pea poppet",
"product_ids": [
"B08YZ24SWJ",
"B08VRGS3CT",
"B094N3CPHH",
"B093Q6QFM6",
"B075C781HQ",
"B07GP13PCY",
"B072ZDN5KJ",
"B0919CDYXB"
],
"scores": [
10,
10,
7,
10,
4,
7,
0,
0
],
"... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 91,
"query": "$25 xbox gift card not digital",
"product_ids": [
"B00F4CF4PU",
"B0086UK7IQ",
"B00F4CEWOY",
"B00F4CF39C",
"B07RZ75JW3",
"B01H6GUCCQ",
"B00F4CF7L6",
"B00F4CEHNK"
],
"scores": [
7,
0,
10,
0,
10,
4,
10,
7
],
"gains": ... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 107,
"query": "$80 golf cart without roof",
"product_ids": [
"B07Y9QL6FW",
"B08336KZCL",
"B082NQCF11",
"B00022749Q",
"B08V165XVB",
"B07M93RZLZ",
"B096RGZ5R5",
"B004CSYAXE"
],
"scores": [
10,
10,
7,
0,
4,
10,
0,
7
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 178,
"query": "(print on back) we are not descended from fearful men decal",
"product_ids": [
"B08P31VVXX",
"B08K58941Q",
"B07V795FC7",
"B0896RNBQD",
"B08RN8DFXR",
"B01BQ53SEW",
"B07PLL1RX3",
"B08LCHKC94"
],
"scores": [
10,
7,
10,
7,
10,
0,... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 242,
"query": ", funny cat suits are not repeated a week",
"product_ids": [
"B08BHTW996",
"B00VJ8O3SM",
"B085P291MH",
"B08FFQCBPR",
"B07RV9RLK5",
"B07PJTH6JJ",
"B00XD075HU",
"B07VXL395Z"
],
"scores": [
0,
0,
0,
7,
0,
10,
10,
10
],... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 258,
"query": ",how to view my orders that are not listed",
"product_ids": [
"B00DZPAGZQ",
"B00MXUHHGK",
"B01M0H2R13",
"B06XB3CSXG",
"B07YXCVQTN",
"B0163DF0KW",
"B002B6OSWW",
"B000ORV3NC"
],
"scores": [
10,
0,
0,
0,
0,
0,
0,
0
],
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 293,
"query": "- wood hanging do not enter",
"product_ids": [
"B08YCX1ND3",
"B08R8PWMPT",
"B081S65PT4",
"B07DQCQXGF",
"B07P5Q43C3",
"B091Z4NYMD",
"B089331Q1J",
"B0716YYVZM"
],
"scores": [
7,
10,
7,
0,
0,
10,
0,
10
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 311,
"query": ". 5 i have 3/4 turn not spigot",
"product_ids": [
"B07GT79XD3",
"B07XZX23YX",
"B008ACWSZU",
"B01CKIWF8Y",
"B07VX8XLKS",
"B019YMGNDW",
"B013FWX4HA",
"B07Y7XPPVR"
],
"scores": [
0,
10,
0,
10,
10,
7,
0,
10
],
"gains"... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 356,
"query": ".5ml syringe without needle",
"product_ids": [
"B07C71C1LH",
"B078K8SVMN",
"B013WWFJX0",
"B07PY77BHF",
"B07GX4RJF3",
"B07JZ2HMJ7",
"B07BQFTHMQ",
"B07Q33SG9Z"
],
"scores": [
7,
0,
10,
10,
10,
0,
7,
10
],
"gains": [... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 381,
"query": ".nothing without intention",
"product_ids": [
"B07KPN8HBB",
"B08H9KQB1Y",
"B088W493LV",
"B00ED3GU1I",
"B077K2NXZX",
"B094VJJZ7G",
"B0843FYSFB",
"B07PBYPPNN"
],
"scores": [
0,
0,
10,
0,
0,
10,
0,
10
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 390,
"query": "/ to keep plastic from babies and not use this bag in",
"product_ids": [
"B01BJ63280",
"B07414752B",
"B003FO2B5U",
"B07MT89MCW",
"B01LAY47D0",
"B07FSLSL1K",
"B074NZY3SQ",
"B01IHHLB3W"
],
"scores": [
10,
10,
10,
0,
0,
10,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 396,
"query": "0 cal water inhancer without sugar",
"product_ids": [
"B00IOGKQEC",
"B07BY3Y3QC",
"B088P8VWPQ",
"B0083PR6V0",
"B00JB13A38",
"B007DJ06J6",
"B084GT5CLZ",
"B01M21KDSN"
],
"scores": [
10,
0,
0,
0,
7,
7,
10,
10
],
"gai... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 419,
"query": "0.0 sticker not today",
"product_ids": [
"B07JBRV6GH",
"B000J09CO6",
"B07HFDGMC2",
"B01DEYPNPU",
"B08K26VHXM",
"B08LD6JTVH",
"B081FK6L1G",
"B07D5MVSBS"
],
"scores": [
10,
0,
7,
0,
0,
0,
0,
10
],
"gains": [
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 421,
"query": "0.01 interval syringe without needle",
"product_ids": [
"B07BQDRDC2",
"B07VF8CKGL",
"B08HM8N7J2",
"B07JZ2HMJ7",
"B07L3NSPD8",
"B07YVGZPRZ",
"B00PXRZRA8",
"B08FFR6L7K"
],
"scores": [
7,
10,
7,
10,
10,
0,
10,
7
],
"... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 449,
"query": "0.53 cu. ft. retro mini fridge in white without freezer",
"product_ids": [
"B0083H1INK",
"B01G7IL3BS",
"B01DZQI6YW",
"B07PMG241V",
"B07Y8T5GK3",
"B00IR8H55A",
"B07S2RBCT5",
"B00L7QVSXE"
],
"scores": [
7,
10,
7,
0,
10,
0,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 450,
"query": "0.5cc syringe without needle",
"product_ids": [
"B07S8YF6KK",
"B01LX9TPC8",
"B07BQFTHMQ",
"B07PY77BHF",
"B08QFGYYPH",
"B07JZ2HMJ7",
"B07SFK6Z5J",
"B07BQDRDC2"
],
"scores": [
10,
0,
0,
7,
10,
0,
7,
10
],
"gains": [... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 474,
"query": "00% natural ingredients not 100% vitamin e",
"product_ids": [
"B0072F8D7S",
"B00008I8NN",
"B001KYU1H2",
"B00JGCPAYY",
"B08HYRGJMQ",
"B00KI3Z2UW",
"B08BBYV5YV",
"B004R64OXM"
],
"scores": [
7,
7,
10,
0,
10,
0,
0,
10
]... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 538,
"query": "06 colorado driver side power window regulator without motor",
"product_ids": [
"B07DR81W27",
"B001KQD7YY",
"B01I82ZKME",
"B07T75SXZL",
"B010G6HXO8",
"B000C16XQW",
"B07QWQBMJ5",
"B003SNR0DG"
],
"scores": [
7,
7,
10,
0,
0,
0,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 571,
"query": "08 corolla rear shocks",
"product_ids": [
"B07X2MTYLC",
"B000C3ZDY8",
"B019WTW5HU",
"B01M322K0J",
"B07YFQ948P",
"B01M68VV9M",
"B000COBXBE",
"B07KD88XDD"
],
"scores": [
7,
4,
10,
7,
0,
10,
4,
10
],
"gains": [
2... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 573,
"query": "08 f250 mirrors without lights",
"product_ids": [
"B01HXL7FLK",
"B07HNZQZG2",
"B07ZFB3LW9",
"B07567QLTX",
"B00VE399VI",
"B073HXHK61",
"B07MJTM5Y4",
"B013CMH8FW"
],
"scores": [
7,
10,
0,
0,
10,
7,
0,
10
],
"gains":... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 615,
"query": "1 0z plastic containers without lids",
"product_ids": [
"B0814LDK3W",
"B07R8DQFN4",
"B00Z9C4CT4",
"B01N81A0DI",
"B079VYK8TB",
"B08NDNX2BG",
"B07T9YJT2Z",
"B075Y2Q3LW"
],
"scores": [
7,
0,
10,
7,
0,
10,
10,
0
],
"g... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 639,
"query": "1 1/8 not threads anode rod for rv",
"product_ids": [
"B07DYX44KB",
"B003VAYRNM",
"B08DXYYFVC",
"B07QNZHPQT",
"B07PDHQ1HT",
"B00NWZ3TUI",
"B08LD8H1VR",
"B079ZC7LC2"
],
"scores": [
10,
0,
7,
10,
0,
0,
10,
7
],
"gai... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 640,
"query": "1 1/8 skinny leather belts without buckles",
"product_ids": [
"B00LNKAOJI",
"B07JD9T73Z",
"B00LEWGKVQ",
"B07QG953ZJ",
"B0722YTXHR",
"B01MS3LQ4W",
"B01L9KFXKG",
"B00ND7BP1K"
],
"scores": [
10,
0,
10,
0,
7,
0,
10,
7
]... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 706,
"query": "1 cordless telephone without answering capability",
"product_ids": [
"B01LZ8ICCY",
"B07KX8RBTB",
"B0765BV53Y",
"B071X8DHVC",
"B00J9R1PEK",
"B009R5VTX6",
"B01A9KBCQG",
"B00DNBI7GM"
],
"scores": [
10,
10,
0,
10,
7,
0,
7,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 722,
"query": "1 dollar makeup",
"product_ids": [
"B002S8TPHQ",
"B00RFM5984",
"B01HY3ON9E",
"B01BKKQ7ZK",
"B01EO2KLDU",
"B00F9A1RX0",
"B00YFQ0M3U",
"B07GV24K69"
],
"scores": [
7,
0,
0,
10,
10,
7,
10,
7
],
"gains": [
2,
0... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 741,
"query": "1 gal. steel paint cans without lids",
"product_ids": [
"B07Y8T49V4",
"B085D7XBJH",
"B07XJP9CVQ",
"B001JT8BTS",
"B000KKPBFE",
"B00B2H7JM2",
"B01MDOW59C",
"B07N8HKTPG"
],
"scores": [
7,
0,
7,
10,
0,
4,
10,
4
],
"ga... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 768,
"query": "1 inch binder4 oz elmer glue",
"product_ids": [
"B00006IFBG",
"B012XO26AW",
"B003U9ABT8",
"B0030BV8JI",
"B07JB22QKX",
"B074HD2JCX",
"B012UHCYHM",
"B00004YTKR"
],
"scores": [
7,
10,
10,
7,
7,
10,
10,
10
],
"gains":... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 808,
"query": "1 magnetic strip without adhesive",
"product_ids": [
"B073XS3CHW",
"B0010HADEA",
"B00JJPPYF8",
"B076DVBZTX",
"B005HYDC68",
"B076ZZ9D2V",
"B076HNNCG9",
"B06XWTJYWB"
],
"scores": [
0,
10,
7,
0,
10,
0,
7,
10
],
"gain... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 820,
"query": "1 ml blister pack luer slip tip without safety",
"product_ids": [
"B07QM74NWG",
"B07BQGVVSC",
"B07LCRX2M7",
"B07BQDRDC2",
"B08HKVP161",
"B07VF8CKGL",
"B0006GE6A8",
"B08LMRKDG7"
],
"scores": [
10,
0,
0,
10,
7,
7,
0,
10... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 834,
"query": "1 mtb stem",
"product_ids": [
"B086V9RRXX",
"B071J12DXW",
"B071YV77R7",
"B085L78348",
"B0028N14GQ",
"B07G71CZG6",
"B08817YF3P",
"B004YIC99Y"
],
"scores": [
10,
4,
10,
4,
4,
10,
10,
10
],
"gains": [
3,
1,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 850,
"query": "1 pairs of samsung galaxy buds without the charger",
"product_ids": [
"B07PDPCRP4",
"B07RF7CSVD",
"B07F17X6CP",
"B07NPD5T5H",
"B07T1PS8SF",
"B083Q453NG",
"B089B658NP",
"B08X2PLV72"
],
"scores": [
0,
10,
0,
0,
0,
10,
7,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 877,
"query": "1 room portable air conditioner without hose",
"product_ids": [
"B095BMS36C",
"B08XJTJ4VN",
"B09135K2Z5",
"B07SPY57ZB",
"B07SPY3G76",
"B0972ZBLRM",
"B08X1K6LF4",
"B091Z5RG4R"
],
"scores": [
10,
10,
0,
7,
7,
10,
10,
10... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 918,
"query": "1 year old girl toys",
"product_ids": [
"B01MQ3YP7Y",
"B072M6PVCZ",
"B07PTL9KKN",
"B07MNJN16S",
"B01M5AWZKC",
"B00V4GDT0C",
"B07BPZNM9G",
"B07GKWLBN2"
],
"scores": [
10,
0,
10,
10,
7,
7,
10,
7
],
"gains": [
3,... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 948,
"query": "1-1/4-inch vessel style drain without overflow",
"product_ids": [
"B07WS3DC1Z",
"B07RLLCV3P",
"B07BB8Y5NT",
"B071KFCHZF",
"B072LYZC8Q",
"B00XBCJOMO",
"B0032CFS3W",
"B00T492QB4"
],
"scores": [
7,
0,
10,
10,
7,
0,
10,
0... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 958,
"query": "1-inch couplings without stops",
"product_ids": [
"B07T29KP8W",
"B017BVXM0U",
"B01E0DOKVM",
"B016F1M0BE",
"B09134F7GY",
"B093FWVNZR",
"B016F1LZZQ",
"B08Y66KP48"
],
"scores": [
10,
10,
7,
0,
4,
10,
7,
0
],
"gains":... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 959,
"query": "1-light fan kit without ceiling fan",
"product_ids": [
"B07Y5ZN8X1",
"B00KGKF2RA",
"B08CXCHFYC",
"B087NTLNZ3",
"B08HQHNTSG",
"B08SL29HLP",
"B099RPCYPJ",
"B08R9R2C8M"
],
"scores": [
0,
0,
10,
10,
7,
10,
7,
4
],
"ga... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 963,
"query": "1. eufy video doorbell camera without wifi",
"product_ids": [
"B08TZP6H9J",
"B07PM2NBGT",
"B0849J7W5X",
"B089KCH58Y",
"B0868LH58W",
"B07GBRV8W5",
"B07GBP3GH9",
"B08571VZ3Q"
],
"scores": [
0,
7,
7,
4,
10,
0,
10,
0
],... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 968,
"query": "1.00 rimless without nose reading glass for men",
"product_ids": [
"B07PJVPM21",
"B08VGJS6L2",
"B08BJMMV9P",
"B08X6S6V9X",
"B0967S12Y2",
"B08V4ZJ2HG",
"B06XVCN3L2",
"B07VPN3S1N"
],
"scores": [
0,
10,
7,
0,
7,
0,
10,
1... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 982,
"query": "1.4cuft small refrigerator without freezer",
"product_ids": [
"B01DZQI9TE",
"B08BX515HP",
"B088STX6HK",
"B07S2RBCT5",
"B00MVVIF2G",
"B00IR8H55A",
"B0083H1INK",
"B07XJRZR8P"
],
"scores": [
10,
7,
7,
10,
7,
0,
7,
0
],... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 984,
"query": "1.5 cu ft mini fridge without freezer",
"product_ids": [
"B08SBWXTMK",
"B00MWXSFM8",
"B00IR8H55A",
"B00M7GMEYK",
"B07XJRZR8P",
"B01DZQI7B4",
"B088STX6HK",
"B00MVVIF2G"
],
"scores": [
10,
0,
0,
7,
0,
7,
0,
7
],
"ga... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 987,
"query": "1.5 heavy duty casters without brakes",
"product_ids": [
"B08FC94P6R",
"B01N9GJ9O4",
"B097JGS7PG",
"B08NYWCJDX",
"B08CKX3DBW",
"B083GLY1ZQ",
"B07DWVWLND",
"B01N7H998V"
],
"scores": [
10,
10,
7,
10,
7,
4,
7,
10
],
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 990,
"query": "1.5 hp pool pump without basket",
"product_ids": [
"B002GP5WAU",
"B07SPM17L5",
"B00ZTRLD2I",
"B07ST5VNV1",
"B07B93VNT2",
"B003NVLWMI",
"B004VJSQV6",
"B07XYZ7WQL"
],
"scores": [
10,
10,
7,
10,
0,
4,
0,
7
],
"gains"... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1072,
"query": "1/2 npt die",
"product_ids": [
"B07TYXR42R",
"B00JPOV4VQ",
"B00FXKK0W0",
"B000QFKQO4",
"B073YC3S8F",
"B0195UX8RW",
"B01MG92FX1",
"B01LXOUZNV"
],
"scores": [
7,
7,
10,
10,
7,
10,
10,
10
],
"gains": [
2,
2,... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1089,
"query": "1/2cc syringe without needle multiple use glass",
"product_ids": [
"B01CUXWUB6",
"B07BQGVVSC",
"B00WDLLMDS",
"B01HFTYINS",
"B07JZ2HMJ7",
"B01J4F92QE",
"B08C7NK2FB",
"B07KW4KLG2"
],
"scores": [
4,
10,
0,
7,
0,
7,
10,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1103,
"query": "1/32 violin",
"product_ids": [
"B08V1PTQMK",
"B00TBPZEWO",
"B00G234FK0",
"B08B57DZH1",
"B0819BF6RK",
"B01N9EA2ES",
"B07YTJDFBP",
"B004O63C0Q"
],
"scores": [
10,
7,
0,
10,
7,
0,
0,
10
],
"gains": [
3,
2,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1130,
"query": "1/4 tool set without ratchet",
"product_ids": [
"B087C7KB6Y",
"B07ZCS5ZQY",
"B07SNQQFRK",
"B08DJ2LDJG",
"B07X37XMX3",
"B08FB22JSJ",
"B07F35ZQXQ",
"B08CVC5BDL"
],
"scores": [
7,
0,
10,
10,
0,
10,
4,
7
],
"gains": ... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1142,
"query": "1/4” retractable air hose reel without hose",
"product_ids": [
"B08TLTVQHZ",
"B000GAS3K8",
"B0064R6D18",
"B01GJ7O4NQ",
"B00ASF6AR4",
"B000OCHEOY",
"B01IW3M0TA",
"B003O556QG"
],
"scores": [
7,
10,
0,
7,
4,
10,
10,
0
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1216,
"query": "10 in. wash brush without handle by rubbermaid",
"product_ids": [
"B00BEUDX40",
"B004MDM8XS",
"B0773LXG72",
"B000HJ9A9Q",
"B005KDD0KY",
"B005KCKNFU",
"B00006ICOT",
"B003AOT25W"
],
"scores": [
10,
10,
0,
10,
4,
7,
0,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1274,
"query": "10 ounce sports bottle",
"product_ids": [
"B07T7D3688",
"B097SVVFL2",
"B096W6R5ZY",
"B08ZCGZ6WZ",
"B07NGPL4H8",
"B07SCR92GQ",
"B08Q4KTKWD",
"B08X6QQLKG"
],
"scores": [
0,
10,
10,
0,
7,
10,
0,
7
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1287,
"query": "10 pocket folders without brads",
"product_ids": [
"B01N74KH32",
"B00006ICMQ",
"B07Z6M1B1F",
"B0091G5VR6",
"B08FC5ZKMM",
"B084P977G8",
"B07KM3WHMH",
"B000DZ9XGO"
],
"scores": [
0,
0,
7,
10,
10,
7,
10,
4
],
"gains... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1292,
"query": "10 psi propane regulator without hose inthreaded",
"product_ids": [
"B07433QCTC",
"B07D9GZ43Y",
"B08KHCX3CW",
"B08JG6NS6L",
"B07GRFTR4S",
"B087B6RSSF",
"B078RGBKN7",
"B08FQY2LBK"
],
"scores": [
7,
4,
0,
10,
10,
0,
7,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1302,
"query": "10 speaker sound system",
"product_ids": [
"B07P9BBJM2",
"B00DDYB5O0",
"B07PL6LDP1",
"B00STTPAP4",
"B007GNO2DU",
"B00ZKM44GO",
"B000230LLS",
"B00TVFO0B0"
],
"scores": [
0,
4,
7,
4,
7,
4,
7,
10
],
"gains": [
0... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1319,
"query": "10 x 10 register",
"product_ids": [
"B019FG5E3C",
"B00LR5NCJS",
"B0792R5KJF",
"B004TW8930",
"B07KDZQJC4",
"B00L83OKRM",
"B006J2VAEI",
"B07M9HB3R7"
],
"scores": [
0,
10,
0,
7,
10,
7,
0,
10
],
"gains": [
0,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1320,
"query": "10 x 10 x 8 box",
"product_ids": [
"B000ZJSGHA",
"B07B4BBQRK",
"B01D37HA4A",
"B01D3BJXIM",
"B00BT3D40U",
"B00BT54UEW",
"B00BT3BQ78",
"B07GXZZM2Q"
],
"scores": [
7,
7,
7,
7,
7,
7,
10,
10
],
"gains": [
2,
2... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1323,
"query": "10 x 5 x 13 inch polka dot party bags without handles",
"product_ids": [
"B07MCSMMWB",
"B07BKQH5XS",
"B00NVICAOW",
"B019G6GYBC",
"B004QJGW6C",
"B08Z7SG3NF",
"B01N2WU339",
"B07FTS73CJ"
],
"scores": [
7,
7,
0,
10,
0,
10,
0... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1341,
"query": "10-3 cord without connectors",
"product_ids": [
"B071JNPS7F",
"B07Z4FVML6",
"B001FBGVIS",
"B08BYD6YGJ",
"B06Y1BJL4X",
"B07TVKHFWD",
"B08Y7RT6LX",
"B085HBXPJH"
],
"scores": [
0,
4,
7,
10,
7,
10,
0,
10
],
"gains": ... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1383,
"query": "100 count kcups",
"product_ids": [
"B00EDHWAL8",
"B07P23H371",
"B07K7YDN24",
"B073QZPNJS",
"B07Q6CSYCF",
"B00N0H16IK",
"B07Z9PPYSH",
"B06Y5YHSHH"
],
"scores": [
10,
10,
10,
10,
7,
7,
10,
7
],
"gains": [
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1516,
"query": "100” foot baby gate without attached",
"product_ids": [
"B01N5PTKXS",
"B0024FAWSE",
"B07MLFKP1G",
"B01BTYXJ3U",
"B08CGVGZZJ",
"B003VNKLIY",
"B001OC5UNK",
"B07HCDS1X6"
],
"scores": [
4,
7,
10,
10,
0,
7,
10,
0
],
"... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1531,
"query": "104 hawthorne single cylinder handleset without doir knob",
"product_ids": [
"B07J4VQCLV",
"B012G9Y2V0",
"B07G4LCPRV",
"B00828S928",
"B012GA04W0",
"B07GF5F9DF",
"B0031AG90Q",
"B004EPYSGE"
],
"scores": [
7,
0,
10,
0,
10,
7,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1582,
"query": "10k ring settings without stones",
"product_ids": [
"B087X9BP7Z",
"B07917ZNGH",
"B077JRRQGC",
"B00FRIJMRW",
"B07MX9VH1T",
"B01HI70GOM",
"B0125AZRU0",
"B01M163Z63"
],
"scores": [
4,
7,
10,
0,
7,
10,
10,
0
],
"gain... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1616,
"query": "10w solar panel 12v without controller",
"product_ids": [
"B08CHBV8L7",
"B07NPDWZJ7",
"B07RXKN9MF",
"B08H7XWC6L",
"B072MMDY4F",
"B0836PVY8J",
"B01MU0WMGT",
"B07Y5TKMZ4"
],
"scores": [
10,
0,
10,
10,
4,
7,
0,
7
],
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1622,
"query": "10x 30tents without sidewalls",
"product_ids": [
"B094NHBM4Z",
"B08T5PQDVK",
"B07G54PPRF",
"B07VGWB6BN",
"B07LGPW59N",
"B082SK4CWZ",
"B07J66XY7Q",
"B07MFXRRB6"
],
"scores": [
0,
10,
10,
7,
4,
0,
10,
7
],
"gains":... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1647,
"query": "10x20 without walls",
"product_ids": [
"B08XXDM9LL",
"B08HGQFBWT",
"B08R1WHJBP",
"B08N4YQLBN",
"B07VLMPDJW",
"B086JGFBQ9",
"B074MNWKC9",
"B074MPK25G"
],
"scores": [
0,
0,
7,
4,
7,
10,
10,
0
],
"gains": [
0,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1660,
"query": "10‘ x 10‘ trampoline without netting",
"product_ids": [
"B08H8CZ8BN",
"B07Q718H16",
"B083SH3W7T",
"B07T8XQXCQ",
"B000JC2ZHA",
"B08B53ZB2T",
"B01BL19JDA",
"B07SMGQ18F"
],
"scores": [
0,
7,
4,
10,
10,
0,
0,
7
],
"g... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1700,
"query": "11 mm allen wrench",
"product_ids": [
"B08QVSL5X7",
"B00C3IB3DA",
"B00C3I4GBG",
"B002YNQP32",
"B0782S4Y4J",
"B075BZGJJK",
"B001HW8ZBG",
"B074Q3SMFC"
],
"scores": [
0,
4,
10,
10,
0,
7,
10,
7
],
"gains": [
0,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1711,
"query": "11 slot knife block without knives",
"product_ids": [
"B08R5PLFX4",
"B00CMP1QUE",
"B07KJYNG3P",
"B00DHH0LRA",
"B07Q3FSP8X",
"B07G83GN7L",
"B00BBOD8US",
"B01DB285NW"
],
"scores": [
10,
7,
10,
7,
0,
0,
10,
0
],
"ga... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1720,
"query": "11' offset patio umbrella without base",
"product_ids": [
"B087BYYRH5",
"B08CVTKRRR",
"B08P21CW7Y",
"B08PG596F2",
"B07QCM2QZ3",
"B08PH856XZ",
"B0814JWT5K",
"B08QMYJRPR"
],
"scores": [
10,
7,
0,
10,
4,
0,
10,
7
],
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1828,
"query": "12 inch metal straw",
"product_ids": [
"B07D72B9HJ",
"B07S89TWK5",
"B07PWRP7SZ",
"B07BF8NJN6",
"B0714QXG2Q",
"B07D34ZNMH",
"B08L5VV7JT",
"B073Q1RYL4"
],
"scores": [
10,
0,
0,
4,
10,
7,
7,
10
],
"gains": [
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1841,
"query": "12 mo boy hard sole shoes",
"product_ids": [
"B00FXOFTGI",
"B07F1HTJYX",
"B07NVLG88R",
"B07NZDVC18",
"B07R1Q3VKX",
"B073ZXJ9F4",
"B07QPDR8Y2",
"B0029DNQL6"
],
"scores": [
0,
7,
10,
7,
7,
7,
7,
10
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1892,
"query": "12 volt auxiliary lights",
"product_ids": [
"B07KS67X7D",
"B071ZPX32N",
"B01LWJ0CP3",
"B07GSWY85P",
"B00IY3YLCI",
"B06XHXLMYT",
"B01FH583KG",
"B07GTNHNMN"
],
"scores": [
0,
4,
7,
10,
7,
10,
10,
4
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 1939,
"query": "120v inline switch",
"product_ids": [
"B081SJ8QZ2",
"B01M0OEBO4",
"B01LE12CXQ",
"B07R95FMS5",
"B07SVVW1LG",
"B077Y2W984",
"B072F9DGRL",
"B00826P0AO"
],
"scores": [
10,
10,
7,
10,
7,
7,
7,
10
],
"gains": [
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2039,
"query": "13 inch laptop",
"product_ids": [
"B07VWSCTRG",
"B086383HC7",
"B07PPDFFG9",
"B07QXKW89P",
"B07QWLXCC7",
"B087RH1D1K",
"B07YNK3H6P",
"B08636NKF8"
],
"scores": [
10,
10,
7,
10,
10,
7,
7,
10
],
"gains": [
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2058,
"query": "135 ounce jar",
"product_ids": [
"B00J0I0TQ8",
"B0749SMW2F",
"B018SFOZYA",
"B083ZD6QV6",
"B07PLL2YFW",
"B07BQWVDWD",
"B00LSY473S",
"B07T4K319K"
],
"scores": [
7,
10,
0,
0,
10,
0,
10,
7
],
"gains": [
2,
3,... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2085,
"query": "14 inch circumference boots",
"product_ids": [
"B00QCZLO3Y",
"B07Y5SRLB5",
"B07M65L5DR",
"B076ZXQPV4",
"B00VJ3PSWM",
"B001GAB2SC",
"B01G7R3VSM",
"B0170U4LRA"
],
"scores": [
10,
7,
4,
10,
7,
7,
10,
10
],
"gains": ... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2134,
"query": "14kwhite gold red quartz crab ring",
"product_ids": [
"B07BR4DVNY",
"B082R3DRJD",
"B07RPFR3J4",
"B07GFK5677",
"B0785JC6Z5",
"B01IZA2Z68",
"B016FI47D6",
"B082KVZ76S"
],
"scores": [
10,
10,
10,
7,
0,
7,
7,
7
],
"ga... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2142,
"query": "14x1.5 lug nuts 19mm 32",
"product_ids": [
"B01MRJPV2B",
"B078RJ6GK6",
"B07HL4PD1S",
"B01N3M7EMK",
"B06XV3S461",
"B07QQ3CFG8",
"B071VTWGJG",
"B0796R1V7B"
],
"scores": [
10,
7,
10,
10,
7,
10,
7,
10
],
"gains": [
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2182,
"query": "15-inch macbook pro",
"product_ids": [
"B078BSQDPK",
"B07JBLRT2L",
"B07B8XX1Z1",
"B07YLFDZYW",
"B08156GLYP",
"B079C8TPRX",
"B0849W8JK7",
"B07FQQ8DSY"
],
"scores": [
10,
7,
10,
7,
10,
10,
7,
10
],
"gains": [
3... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2203,
"query": "150qt cooler",
"product_ids": [
"B000G64FJK",
"B00165UHI8",
"B000W4TFY6",
"B00I15BDPW",
"B0001DI46I",
"B07DNN9F18",
"B003AKZ7K0",
"B00J0RZQJ4"
],
"scores": [
0,
10,
10,
0,
7,
0,
10,
7
],
"gains": [
0,
3,
... |
[
{
"role": "system",
"content": "You are a product relevance judge. Given a shopping query and a list of K candidate products, score each product's relevance to the query on a 0-10 scale (10 = exactly the product asked for, 0 = irrelevant). Return strict JSON: a list of K objects with keys product_id and sco... | {
"query_id": 2214,
"query": "15x48 above ground pool with pump",
"product_ids": [
"B08T254GFD",
"B07F39F71J",
"B083C5W6PY",
"B07H37YHMZ",
"B073FMDBLV",
"B07B5QQZRB",
"B08YLSRFHL",
"B01MUDQ26U"
],
"scores": [
7,
7,
10,
10,
7,
10,
10,
10
],
"g... |
End of preview. Expand in Data Studio
ErrOPD-ESCI
Listwise product-ranking data used in ErrOPD: When Does On-Policy Distillation Transfer Teacher Errors? (Xin Li, Hao Jiang, Annan Wang, Yichi Zhang, Chau Yuen). Code: https://github.com/LiXin97/ErrOPD · Project page: https://lixin.ai/ErrOPD · Models and data: ErrOPD collection
Each record is a chat (messages: system, user, assistant) plus metadata. The user turn gives a shopping query and 8 candidate
products (id, title, brand, color); the assistant turn is a strict JSON list of 8 objects with product_id and a 0-10 relevance
score, in prompt order.
Files
| File | Config / split | Rows | Use in the paper |
|---|---|---|---|
train.jsonl |
default / train |
1,800 | student fine-tuning targets and the on-policy distillation prompts |
val.jsonl |
default / validation |
200 | held-out prompts for evaluation |
teacher_sft/k1_f00_x5.jsonl |
teacher_sft / clean |
9,000 | 5 copies of train.jsonl, no injected error (clean teacher control) |
teacher_sft/k1_f20_x5.jsonl |
teacher_sft / early_close_20pct |
9,000 | early close injected into 1,832 of 9,000 copies (20.4%): the list is closed one item early (7 objects instead of 8) |
teacher_sft/dup_f20_x5.jsonl |
teacher_sft / duplicated_id_20pct |
9,000 | duplicated ID injected into 1,735 of 9,000 copies (19.3%): one product ID appears twice among the 8 objects |
teacher_sft/manifest.json |
injection seed (20260930), per-file counts and sha256 |
Row counts and injection counts above were recounted from the files (the injected copies are exactly the rows whose answer differs
from the clean answer of their source row). The injection code is scripts/data_inject.py in the code repository.
Fields
messages: list of{"role", "content"}turns:system(task instruction),user(query and 8 candidates),assistant(target JSON). In theteacher_sftfiles the assistant turn starts with an empty<think>\n\n</think>\n\nblock (the Qwen3 non-thinking format used to fine-tune the Qwen3 teachers);train.jsonlandval.jsonlhave no such block.metadata.query_id,metadata.query: ESCI query id and text.metadata.product_ids: the 8 candidate ASINs in prompt order.metadata.scores: target 0-10 relevance scores in prompt order;metadata.gains,metadata.esci_labels: the underlying ESCI gains and E/S/C/I labels.metadata.K(8),metadata.split,metadata.dataset: list size and provenance.metadata.f2_copy,metadata.f2_row(teacher_sftonly): copy index (0-4) and the source row intrain.jsonl.
Usage
from datasets import load_dataset
ds = load_dataset("XINLI1997/ErrOPD-ESCI") # train (1,800) / validation (200)
teacher = load_dataset("XINLI1997/ErrOPD-ESCI", "teacher_sft") # clean / early_close_20pct / duplicated_id_20pct (9,000 each)
example = ds["validation"][0]
prompt = [m for m in example["messages"] if m["role"] != "assistant"] # feed to a chat model
target = example["messages"][-1]["content"] # strict JSON list of 8 objects
Source and license
Built from the Amazon Shopping Queries Dataset (ESCI; Reddy et al. 2022, https://github.com/amazon-science/esci-data), which its official repository releases under the Apache-2.0 license. This derivative is released under Apache-2.0.
Citation
@article{li2026erropd,
title = {ErrOPD: When Does On-Policy Distillation Transfer Teacher Errors?},
author = {Li, Xin and Jiang, Hao and Wang, Annan and Zhang, Yichi and Yuen, Chau},
journal = {arXiv preprint arXiv:2605.08737},
year = {2026}
}
@article{reddy2022shopping,
title = {Shopping Queries Dataset: A Large-Scale {ESCI} Benchmark for Improving Product Search},
author = {Reddy, Chandan K. and M{\`a}rquez, Llu{\'i}s and Valero, Fran and Rao, Nikhil and Zaragoza, Hugo and Bandyopadhyay, Sambaran and Biswas, Arnab and Xing, Anlu and Subbian, Karthik},
journal = {arXiv preprint arXiv:2206.06588},
year = {2022}
}
- Downloads last month
- 49