Update README.md
Browse files
README.md
CHANGED
|
@@ -38,53 +38,46 @@ Our flagship **Datumo Platform** is Korea's first end-to-end AI trust evaluation
|
|
| 38 |
|
| 39 |
## ๐ฏ What We Do
|
| 40 |
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
</td>
|
| 52 |
-
<td width="33%" valign="top">
|
| 53 |
-
|
| 54 |
-
### ๐ก๏ธ AI Trust & Safety
|
| 55 |
-
LLM red-teaming
|
| 56 |
-
Reliability benchmarks
|
| 57 |
-
Safety evaluation datasets
|
| 58 |
-
Guardrail testing
|
| 59 |
-
|
| 60 |
-
</td>
|
| 61 |
-
<td width="33%" valign="top">
|
| 62 |
-
|
| 63 |
-
### ๐ Datumo Platform
|
| 64 |
-
Automated LLM evaluation
|
| 65 |
-
Dashboard analytics
|
| 66 |
-
**45 days โ 45 minutes**
|
| 67 |
-
End-to-end eval pipeline
|
| 68 |
-
|
| 69 |
-
</td>
|
| 70 |
-
</tr>
|
| 71 |
-
</table>
|
| 72 |
|
| 73 |
---
|
| 74 |
|
| 75 |
## ๐ Featured Collections
|
| 76 |
|
| 77 |
### ๐ก๏ธ [Safety-Data](https://huggingface.co/collections/datumo/safety-data)
|
|
|
|
| 78 |
Curated by our **AI Safety team** โ Korean-language safety and reliability benchmarks for LLM evaluation.
|
| 79 |
|
| 80 |
-
|
| 81 |
-
-
|
|
|
|
|
|
|
|
|
|
| 82 |
|
| 83 |
### ๐ฆ [Data-Data](https://huggingface.co/collections/datumo/data-data)
|
| 84 |
-
Research outputs from our **Data team**.
|
| 85 |
|
| 86 |
-
|
| 87 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
|
| 89 |
> ๐ก ํ๋ก์ฐํ์๋ฉด ์ ๋ฐ์ดํฐ์
๊ณผ ๋ชจ๋ธ์ด ๊ณต๊ฐ๋ ๋ ์๋ฆผ์ ๋ฐ์ผ์ค ์ ์์ด์.
|
| 90 |
|
|
@@ -92,13 +85,150 @@ Research outputs from our **Data team**.
|
|
| 92 |
|
| 93 |
## ๐ Milestones
|
| 94 |
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
- ๐
**Forbes "30 Under 30 Asia"** โ Enterprise Technology
|
| 98 |
-
-
|
| 99 |
-
-
|
| 100 |
-
-
|
| 101 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
|
| 103 |
---
|
| 104 |
|
|
@@ -109,10 +239,11 @@ Research outputs from our **Data team**.
|
|
| 109 |
| ๐ Website | [selectstar.ai](https://selectstar.ai/) |
|
| 110 |
| ๐ฐ Blog | [selectstar.ai/blog](https://selectstar.ai/blog/) |
|
| 111 |
| ๐ผ Enterprise inquiries | [Contact form](https://selectstar.ai/contact_page/) |
|
| 112 |
-
| ๐ฌ Community |
|
|
|
|
| 113 |
|
| 114 |
---
|
| 115 |
|
| 116 |
<div align="center">
|
| 117 |
-
<sub>โญ Building the data foundation for trustworthy AI
|
| 118 |
</div>
|
|
|
|
| 38 |
|
| 39 |
## ๐ฏ What We Do
|
| 40 |
|
| 41 |
+
Perception AI(2018~) โ Generative AI(2022~) โ **Agentic AI(2026~)** ๋ก ์ด์ด์ง๋ AI ์งํ ์ ๋จ๊ณ์ ๊ฑธ์ณ, ๋ฐ์ดํฐ ๊ตฌ์ถ๋ถํฐ ์ ๋ขฐ์ฑ ๊ฒ์ฆ๊น์ง **End-to-End ํ์ดํ๋ผ์ธ**์ ์ ๊ณตํฉ๋๋ค.
|
| 42 |
+
|
| 43 |
+
| ๐๏ธ Data Construction | ๐ก๏ธ AI Trust & Safety | ๐ Datumo Platform |
|
| 44 |
+
|---|---|---|
|
| 45 |
+
| ๊ณ ๋๋ ์ถ๋ก ๋ฐ์ดํฐ ์์ฑ (CAC-CoT, GRADE, ATA, COBA) | LLM ๋ ๋ํฐ๋ฐ (CAGE, STAR-Teaming) | ๊ตญ๋ด ์ต์ด LLM ์ ๋ขฐ์ฑ ์๋ํ ํ๊ฐ ํ๋ซํผ |
|
| 46 |
+
| ์ฌ์ ํ์ตยทํ์ธํ๋ ๋ฐ์ดํฐ ๋ผ์ด์ ์ฑ | ํ๊ตญ์ด Safety ๋ฒค์น๋งํฌ (KorNAT, KorSET, FinRED) | **ํ๊ฐ ๊ธฐ๊ฐ 45์ผ โ 45๋ถ** |
|
| 47 |
+
| RAG ์ง์ ํ์ดํ๋ผ์ธ | Safety Judge (Datumo-Guard) | ์จํ๋ ๋ฏธ์คยท๋ง๋ถ๋ฆฌ ํ๊ฒฝ ์ง์ |
|
| 48 |
+
| 25๋ง ๋ช
+ ํฌ๋ผ์ฐ๋์์ปค ยท 2์ต ๊ฑด+ ์ด๋
ธํ
์ด์
| ๊ธ์ตยท์๋ฃยท๊ณต๊ณต ๋๋ฉ์ธ ํนํ ํ๊ฐ | Dashboard Analytics & Reporting |
|
| 49 |
+
|
| 50 |
+
> ๐ค **์ฃผ์ ํํธ๋์ญ**: SKT ๋
์ AI ํ์ด๋ฐ์ด์
๋ชจ๋ธ(๋
ํ๋ชจ) ์ปจ์์์ ยท GSMA Open Telco AI ยท ์ผ์ฑ์๋ช
C-Lab Outside ยท ๊ธ์ต๋ณด์์ ยท ์์ฝ์ฒ ์๋ฃ ๋ ๋ํ
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
|
| 52 |
---
|
| 53 |
|
| 54 |
## ๐ Featured Collections
|
| 55 |
|
| 56 |
### ๐ก๏ธ [Safety-Data](https://huggingface.co/collections/datumo/safety-data)
|
| 57 |
+
|
| 58 |
Curated by our **AI Safety team** โ Korean-language safety and reliability benchmarks for LLM evaluation.
|
| 59 |
|
| 60 |
+
| Dataset | Description | Venue |
|
| 61 |
+
|---|---|---|
|
| 62 |
+
| ๐ธ [**KorSET**](https://huggingface.co/datasets/datumo/KorSET) | CAGE ํ๋ ์์ํฌ๋ก ๊ตฌ์ถํ ํ๊ตญ์ด ๋ ๋ํฐ๋ฐ ๋ฒค์น๋งํฌ (5๊ฐ ์ํ ๋๋ฉ์ธ ยท 12๊ฐ ์นดํ
๊ณ ๋ฆฌ ยท 53๊ฐ ์ธ๋ถ ์ ํ ยท ~8,000๊ฑด) | **ICLR 2026** (CAGE) |
|
| 63 |
+
| ๐ธ [**KorNAT**](https://huggingface.co/datasets/datumo/KorNAT) | Korea's first LLM reliability / national-alignment benchmark | ACL 2024 Findings |
|
| 64 |
+
| ๐ธ **FinRED** | ๊ธ์ต ๋๋ฉ์ธ LLM **๋ ๋ํฐ๋ฐ(Red-Teaming)** ํ๊ฐ ๋ฒค์น๋งํฌ (๊ธ์ต๋ณด์์ AIํ์ ์ค ๊ณต๋ ๊ตฌ์ถ) | KDD 2026 D&B Track |
|
| 65 |
|
| 66 |
### ๐ฆ [Data-Data](https://huggingface.co/collections/datumo/data-data)
|
|
|
|
| 67 |
|
| 68 |
+
Research outputs from our **Data team** โ models and datasets built in-house.
|
| 69 |
+
|
| 70 |
+
| Resource | Description | Type |
|
| 71 |
+
|---|---|---|
|
| 72 |
+
| ๐ธ [**CAC-CoT**](https://huggingface.co/datumo/CAC-CoT) | 7B Connector-Aware Compact Chain-of-Thought reasoning model | Model |
|
| 73 |
+
| ๐ธ [**CAC-CoT dataset**](https://huggingface.co/datasets/datumo/CAC-CoT) | Accompanying training data for CAC-CoT | Dataset |
|
| 74 |
+
|
| 75 |
+
### ๐๏ธ Co-built Benchmarks
|
| 76 |
+
|
| 77 |
+
์
๋ ํธ์คํ๊ฐ ๊ณต๋ ๊ตฌ์ถ์ ์ฐธ์ฌํ ํ๊ตญ์ด ๋ํ ๋ฒค์น๋งํฌ์
๋๋ค.
|
| 78 |
+
|
| 79 |
+
- ๐น [**KLUE**](https://huggingface.co/datasets/klue/klue) โ Korean Language Understanding Evaluation (NeurIPS 2021 Datasets & Benchmarks)
|
| 80 |
+
- ๐น [**KorQuAD 2.0**](https://huggingface.co/datasets/KETI-AIR/korquad) โ Korean Question Answering Dataset (LG CNS ๊ณต๋ ๊ตฌ์ถ)
|
| 81 |
|
| 82 |
> ๐ก ํ๋ก์ฐํ์๋ฉด ์ ๋ฐ์ดํฐ์
๊ณผ ๋ชจ๋ธ์ด ๊ณต๊ฐ๋ ๋ ์๋ฆผ์ ๋ฐ์ผ์ค ์ ์์ด์.
|
| 83 |
|
|
|
|
| 85 |
|
| 86 |
## ๐ Milestones
|
| 87 |
|
| 88 |
+
**Highlight (์ต๊ทผ ์ฃผ์ ์ฑ๊ณผ)**
|
| 89 |
+
|
| 90 |
+
- ๐
**Forbes "30 Under 30 Asia" 2021** โ Enterprise Technology (๊ณต๋์ฐฝ์
์ 4์ธ ์ ์ )
|
| 91 |
+
- ๐
**Forbes Korea "2025 ๋ํ๋ฏผ๊ตญ AI 50"** ์ ์
|
| 92 |
+
- ๐
**Forbes Asia "100 ์ ๋ง ๊ธฐ์
" 2025** ์ ์
|
| 93 |
+
- ๐ฐ๐ท **๋
์ AI ํ์ด๋ฐ์ด์
๋ชจ๋ธ(๋
ํ๋ชจ)** 1์ฐจ ํต๊ณผ (2026.01, SKT ์ปจ์์์ ๋ฐ์ดํฐ ์ด๊ด)
|
| 94 |
+
- ๐ **GSMA Open Telco AI** ๊ณต์ ํํธ๋ ํฉ๋ฅ (2026.03, MWC Barcelona)
|
| 95 |
+
- ๐ฐ ๋์ ํฌ์ **434์ต์ ๋ํ** (2025.12, Series B ํ์ฅ)
|
| 96 |
+
- ๐ ๋์ ์ด๋
ธํ
์ด์
**2์ต ๊ฑด+** ยท ๊ธฐ์
๊ณ ๊ฐ **287๊ฐ+** ยท ํฌ๋ผ์ฐ๋์์ปค **25๋ง ๋ช
+**
|
| 97 |
+
|
| 98 |
+
<details>
|
| 99 |
+
<summary><b>๐ ์ ์ฒด ์ฐํ ๋ณด๊ธฐ (2018โ2026)</b></summary>
|
| 100 |
+
|
| 101 |
+
### ๐ฑ Founding & Early Traction (2018โ2020)
|
| 102 |
+
|
| 103 |
+
| ์ฐ์ | ๋ด์ฉ |
|
| 104 |
+
|---|---|
|
| 105 |
+
| 2018.11 | ์
๋ ํธ์คํ(์ฃผ) ์ค๋ฆฝ |
|
| 106 |
+
| 2018.12 | KAIST ์ฐฝ์
๋ํ(E*5) ์ต์ฐ์์ |
|
| 107 |
+
| 2019.07 | ์นด์นด์ค๋ฒค์ฒ์ค SEED 4์ต ํฌ์ ์ ์น |
|
| 108 |
+
| 2019.09 | KorQuAD 2.0 Dataset ๊ตฌ์ถ (LG CNS ๊ณต๋) |
|
| 109 |
+
| 2019.10 | TIPS ํ๋ก๊ทธ๋จ ์ ์ |
|
| 110 |
+
| 2019.12 | ๊ธฐ์
๋ถ์ค์ฐ๊ตฌ์ ์ค๋ฆฝ ์ธ์ |
|
| 111 |
+
| 2020.09 | Series A 40์ต ํฌ์ ์ ์น (์นด์นด์ค๋ฒค์ฒ์คยท์ฝ์ค๋กฑ์ธ๋ฒ ์คํธ๋จผํธยท์ปดํผ๋์ผ์ดํํธ๋์ค) |
|
| 112 |
+
| 2020.10 | **SideGuide** (IROS 2020) ๋
ผ๋ฌธ ์ฑ๊ณผ โ Large-scale Sidewalk Dataset |
|
| 113 |
+
| 2020.11 | ๋ฐ์ดํฐ์คํ์ฆ ์ต์ฐ์์ (๊ณผํ๊ธฐ์ ์ ๋ณดํต์ ๋ถ์ฅ๊ด์) |
|
| 114 |
+
|
| 115 |
+
### ๐ Scale-Up & Global Recognition (2021โ2022)
|
| 116 |
+
|
| 117 |
+
| ์ฐ์ | ๋ด์ฉ |
|
| 118 |
+
|---|---|
|
| 119 |
+
| 2021.01 | Samsung C-Lab Outside ์ ์ |
|
| 120 |
+
| **2021.04** | ๐
**Forbes "30 Under 30 Asia"** Enterprise Technology ์ ์ (๊ณต๋์ฐฝ์
์ 4์ธ) |
|
| 121 |
+
| 2021.11 | **KLUE** NeurIPS 2021 Datasets & Benchmarks ๋
ผ๋ฌธ ์ฑ๊ณผ |
|
| 122 |
+
| 2022.01 | CES 2022 ์ฐธ์ฌ (Samsung C-Lab) |
|
| 123 |
+
| 2022.02 | ์ 1๊ธฐ ์ธ๊ณต์ง๋ฅ ์ค๋ฆฌ ์ ์ฑ
ํฌ๋ผ ๊ธฐ์ ๋ถ๊ณผ ์์ ์ ์ |
|
| 124 |
+
| 2022.03 | **Instance-wise Occlusion and Depth Orders** CVPR 2022 ๋
ผ๋ฌธ ์ฑ๊ณผ |
|
| 125 |
+
| 2022.07 | Series A Extension 90์ต ํฌ์ ์ ์น |
|
| 126 |
+
| 2022.07 | ๊ธฐ์ ํ์ ํ ์ค์๊ธฐ์
(inno-Biz) ์ธ์ฆ |
|
| 127 |
+
| 2022.11 | **KOLD** (EMNLP 2022), **CochlScene** (APSIPA 2022), **Split-GCN** (TPAMI, 1์ ์) ๋
ผ๋ฌธ ์ฑ๊ณผ |
|
| 128 |
+
|
| 129 |
+
### ๐ง LLM Era & AI Safety (2023โ2024)
|
| 130 |
+
|
| 131 |
+
| ์ฐ์ | ๋ด์ฉ |
|
| 132 |
+
|---|---|
|
| 133 |
+
| 2023.05 | Series A Extension 40์ต ํฌ์ ์ ์น (์ฐ์
์ํ) |
|
| 134 |
+
| 2023.06 | AI ๊ธฐ๋ฐ ๊ตญ๋ฐฉ ํ์ ํฌ๋ผ ๋์ (์ก๊ตฐ์ฐธ๋ชจ์ด์ฅ์) |
|
| 135 |
+
| 2023.07 | "AI Talk with Andrew Ng" ํ์ฌ Keynote Speaker |
|
| 136 |
+
| 2023.10 | Samsung Developer Conference 2023 ์ฐ์ฌ ์ฐธ์ฌ |
|
| 137 |
+
| 2023.11 | ๋ํ๋ฏผ๊ตญ Digital Innovation Award ํน๋ณ์ |
|
| 138 |
+
| 2023.12 | **Analyzing Norm Violations in Live-Stream Chat** EMNLP 2023 ๋
ผ๋ฌธ ์ฑ๊ณผ |
|
| 139 |
+
| 2023.12 | ๊ตญ๋ด ์ต์ด "์ด๊ฑฐ๋ ์ธ์ด ๋ชจ๋ธ ์ ๋ขฐ์ฑ ๋ฒค์น๋งํฌ ๋ฐ์ดํฐ์
" ๊ตฌ์ถ (NIA) |
|
| 140 |
+
| 2024.04 | **Gen AI Korea 2024: ์์ฑํ AI ๋ ๋ํ ์ฑ๋ฆฐ์ง** ์ปจํผ๋ฐ์ค ๊ธฐํยท์ด์ (๊ณผ๊ธฐ์ ํต๋ถ) |
|
| 141 |
+
| **2024.08** | **KorNAT** ACL 2024 Findings ๋
ผ๋ฌธ ์ฑ๊ณผ โ ๊ตญ๋ด AI ๋ฐ์ดํฐ ๊ธฐ์
์ต์ด ๊ธ๋ก๋ฒ Top AI ํํ ๋ฐ์ดํฐ์
1์ ์ ๋
ผ๋ฌธ |
|
| 142 |
+
| 2024.10 | KT 'Responsible AI ์๋ฌธ ์์ํ' ์๋ฌธ์ ์์ ์ ์ |
|
| 143 |
+
| 2024.11 | ์ 2ํ ์ธ๊ณต์ง๋ฅ ์ ๋ขฐ์ฑ ๋์ ์ฐ์์ (์ ๋ณดํต์ ์ ์ฑ
์ฐ๊ตฌ์ ์์ฅ์) |
|
| 144 |
+
| 2024.11 | GSMA AI Summit 2024 ์ฐ์ฌ ์ฐธ์ฌ |
|
| 145 |
+
| 2024.12 | ๊ตญ๋ด ์ต์ด **LLM ๋ฌดํด์ฑ ํ๊ฐ ๋ฐ์ดํฐ DQ(Data Quality) ์ธ์ฆ** ํ๋ (TTA) |
|
| 146 |
+
| 2024.12 | 2024 ์์์AI๋์ ๋ฒค์ฒ๊ธฐ์
ํํ ํ์ฅ์ |
|
| 147 |
+
|
| 148 |
+
### ๐ Agentic AI & Global Expansion (2025โ2026)
|
| 149 |
+
|
| 150 |
+
| ์ฐ์ | ๋ด์ฉ |
|
| 151 |
+
|---|---|
|
| 152 |
+
| 2025.02 | **Datumo Eval ์ถ์** โ ๊ตญ๋ด ์ต์ด LLM ์๋ํ ํ๊ฐ ํ๋ซํผ |
|
| 153 |
+
| 2025.03 | **Gen AI Red Team Challenge** ๊ณต๋ ๊ฐ์ต (MWC Barcelona, GSMA) โ ์ธ๊ณ ์ต์ด ์คํ๋ผ์ธ ๊ธ๋ก๋ฒ ๋ ๋ํ ์ฑ๋ฆฐ์ง |
|
| 154 |
+
| 2025.04 | AI ๊ธฐ๋ณธ๋ฒ ์์ ์ฑ ๊ฐ์ด๋๋ผ์ธ TF ์์ ์ ์ (๊ณผ๊ธฐ์ ํต๋ถยทAI์์ ์ฐ๊ตฌ์, ๊น์ธ์ฝ ๋ํ) |
|
| 155 |
+
| 2025.05 | ๐
**Forbes Korea "2025 ๋ํ๋ฏผ๊ตญ AI 50"** ์ ์ |
|
| 156 |
+
| 2025.06 | ์ผ์ฑ๊ธ์ต C-Lab Outside ์ต์ข
์ ์ (์ผ์ฑ์๋ช
๊ธ์ต AI ์ ๋ขฐ์ฑ ๊ฒ์ฆ ํ์
) |
|
| 157 |
+
| 2025.07 | ๋ฏผ๊ฐ AI ์ ๋ขฐ์ฑ ์ธ์ฆ 'AI-MASTER' ์ํ๊ธฐ๊ด ์ฐธ์ฌ (๊ตญ๋ด ์ต์ด ๋ฏผ๊ฐ ์ฃผ๋ ์ฒด๊ณ) |
|
| 158 |
+
| 2025.08 | **Series B 205์ต์ ํฌ์ ์ ์น** |
|
| 159 |
+
| 2025.08 | ๐
**Forbes Asia "100 ์ ๋ง ๊ธฐ์
2025"** ์ ์ |
|
| 160 |
+
| 2025.08 | **๋
์ AI ํ์ด๋ฐ์ด์
๋ชจ๋ธ(๋
ํ๋ชจ)** ์ ์ํ ์ ๋ฐ (SKT ์ปจ์์์ ๋ฐ์ดํฐ ์ด๊ด) |
|
| 161 |
+
| 2025.09 | **๊ตญ๊ฐ์ธ๊ณต์ง๋ฅ์ ๋ต์์ํ ๋ฐ์ดํฐ ๋ถ๊ณผ์์** ์์ด (๊น์ธ์ฝ ๋ํ) |
|
| 162 |
+
| 2025.09 | ์์ฝ์ฒ ์ฒจ๋จ AI ๋์งํธ ์๋ฃ์ ํ ๋ ๋ํ ์ฑ๋ฆฐ์ง ํ์ โ ์์์ ์ฒซ '์๋ฃ ๋ ๋ํ' |
|
| 163 |
+
| 2025.10 | ์ผ์ฑ๊ธ์ต C-Lab Outside **์ต์ฐ์ ์คํํธ์
** ์ ์ (์ผ์ฑ์๋ช
) |
|
| 164 |
+
| 2025.11 | **CAC-CoT ยท CoBA ยท GRADE** EMNLP 2025 ๋
ผ๋ฌธ 3ํธ ๋์ ๋ฑ์ฌ |
|
| 165 |
+
| 2025.11 | 2025 ์ด๋ฐ์ผ๋ฆฌ AI ์ฝ๋ฆฌ์ ๋์ (ํ๊ตญ์ธ๊ณต์ง๋ฅ์ฐ์
ํํ์ฅ์) |
|
| 166 |
+
| 2025.11 | Good AI Awards 2025 NIA ์์ฅ์ |
|
| 167 |
+
| 2025.12 | Series B 55์ต์ ์ถ๊ฐ ํฌ์ โ **๋์ ํฌ์ 434์ต์** ๋ํ |
|
| 168 |
+
| **2026.01** | ๐ฐ๐ท **๋
์ AI ํ์ด๋ฐ์ด์
๋ชจ๋ธ(๋
ํ๋ชจ) 1์ฐจ ํต๊ณผ** (SKT ์ปจ์์์) |
|
| 169 |
+
| 2026.02 | **CAGE** ICLR 2026 Main Conference ๋
ผ๋ฌธ ์ฑ๊ณผ |
|
| 170 |
+
| 2026.03 | **GSMA 'Open Telco AI'** ๊ธ๋ก๋ฒ ์ฐํฉ์ฒด ๊ณต์ ํํธ๋ ํฉ๋ฅ (MWC Barcelona) |
|
| 171 |
+
| 2026.03 | MWC 2026 Gen AI Red Team Challenge ๊ณต๋ ์ฃผ๊ด (GSMA ยท LG U+) |
|
| 172 |
+
|
| 173 |
+
</details>
|
| 174 |
+
|
| 175 |
+
---
|
| 176 |
+
|
| 177 |
+
## ๐ Publications
|
| 178 |
+
|
| 179 |
+
์
๋ ํธ์คํ๊ฐ ๋จ๋
ยท๊ณต๋ยท์ง์ ์ฐธ์ฌํ ๋
ผ๋ฌธ ๋ชฉ๋ก์
๋๋ค. ๊ตญ์ AIยทML Top ํํ ์ค๏ฟฝ๏ฟฝ๏ฟฝ์ผ๋ก ์ ๋ฆฌํ์ต๋๋ค.
|
| 180 |
+
|
| 181 |
+
<details open>
|
| 182 |
+
<summary><b>๐ฅ 2026 (4ํธ)</b></summary>
|
| 183 |
+
|
| 184 |
+
| Paper | Co-authors | Venue |
|
| 185 |
+
|---|---|---|
|
| 186 |
+
| **STAR-Teaming**: A Strategy-Response Multiplex Network Approach to Automated LLM Red Teaming | Selectstar | ACL 2026 |
|
| 187 |
+
| **FinRED**: An Expert-Guided Red-Teaming Benchmark for Financial LLM Safety | Selectstar ยท ๊ธ์ต๋ณด์์ AIํ์ ์ค | KDD 2026 Dataset & Benchmark Track |
|
| 188 |
+
| [**CAGE**](https://openreview.net/forum?id=gCm55KYiqz): A Framework for Culturally Adaptive Red-Teaming Benchmark Generation | Selectstar | **ICLR 2026** Main |
|
| 189 |
+
| **E-star-12B**: Rubric-Following Evaluator Adaptive Across Industrial Domains | Selectstar | ACL 2026 Workshop (์งํ ์ค) |
|
| 190 |
+
|
| 191 |
+
</details>
|
| 192 |
+
|
| 193 |
+
<details>
|
| 194 |
+
<summary><b>๐ 2025 (4ํธ)</b></summary>
|
| 195 |
+
|
| 196 |
+
| Paper | Co-authors | Venue |
|
| 197 |
+
|---|---|---|
|
| 198 |
+
| [**CoBA**](https://aclanthology.org/2025.emnlp-main.520/): Counterbias Text Augmentation for Mitigating Various Spurious Correlations via Semantic Triples | Selectstar ยท ์ค์๋ํ๊ต | EMNLP 2025 Main |
|
| 199 |
+
| [**GRADE**](https://aclanthology.org/2025.findings-emnlp.236/): Generating multi-hop QA and fine-gRAined Difficulty matrix for RAG Evaluation | Selectstar ยท KAIST | EMNLP 2025 Findings |
|
| 200 |
+
| [**CAC-CoT**](https://aclanthology.org/2025.findings-emnlp.1062/): Connector-Aware Compact Chain-of-Thought for Efficient Reasoning Data Synthesis Across Dual-System Cognitive Tasks | Selectstar | EMNLP 2025 Findings |
|
| 201 |
+
| **ATA**: Autonomous Tabular-data Analysis for Insight Generation via Statistical Methods | Selectstar ยท ์ผ์ฑ์ฆ๊ถ ๊ธ์ตAI์ผํฐ | ๊ณต์ ์ ์ถ ์ค |
|
| 202 |
+
|
| 203 |
+
</details>
|
| 204 |
+
|
| 205 |
+
<details>
|
| 206 |
+
<summary><b>๐ 2024 (1ํธ)</b></summary>
|
| 207 |
+
|
| 208 |
+
| Paper | Co-authors | Venue |
|
| 209 |
+
|---|---|---|
|
| 210 |
+
| [**KorNAT**](https://arxiv.org/abs/2402.13605): LLM Alignment Benchmark for Korean Social Values and Common Knowledge | Selectstar ยท KAIST ยท SKT ยท LG ยท ๋ค์ด๋ฒ ยท KT ยท NIA | ACL 2024 Findings |
|
| 211 |
+
|
| 212 |
+
> ๊ตญ๋ด AI ๋ฐ์ดํฐ ๊ธฐ์
์ต์ด ๊ธ๋ก๋ฒ Top AI ํํ์ ๋ฐ์ดํฐ์
์ฃผ์ 1์ ์ ๋
ผ๋ฌธ ๋ฑ์ฌ
|
| 213 |
+
|
| 214 |
+
</details>
|
| 215 |
+
|
| 216 |
+
<details>
|
| 217 |
+
<summary><b>๐ 2021โ2023 (5ํธ)</b></summary>
|
| 218 |
+
|
| 219 |
+
| Year | Paper | Venue |
|
| 220 |
+
|---|---|---|
|
| 221 |
+
| 2023 | [**Analyzing Norm Violations in Live-Stream Chat**](https://aclanthology.org/2023.emnlp-main.55/) | EMNLP 2023 |
|
| 222 |
+
| 2022 | [**KOLD**](https://aclanthology.org/2022.emnlp-main.744/): Korean Offensive Language Dataset | EMNLP 2022 |
|
| 223 |
+
| 2022 | [**Split-GCN**](https://ieeexplore.ieee.org/document/9984937): Effective Interactive Annotation for Segmentation of Disconnected Instance | IEEE TPAMI (1์ ์) |
|
| 224 |
+
| 2022 | [**Instance-wise Occlusion and Depth Orders**](https://openaccess.thecvf.com/content/CVPR2022/html/Lee_Instance-Wise_Occlusion_and_Depth_Orders_in_Natural_Scenes_CVPR_2022_paper.html) | CVPR 2022 |
|
| 225 |
+
| 2022 | [**CochlScene**](https://ieeexplore.ieee.org/document/9979822): Acquisition of acoustic scene data using crowdsourcing | APSIPA 2022 |
|
| 226 |
+
| 2021 | [**KLUE**](https://datasets-benchmarks-proceedings.neurips.cc/paper/2021/hash/98dce83da57b0395e163467c9dae521b-Abstract-round2.html): Korean Language Understanding Evaluation | NeurIPS 2021 Datasets & Benchmarks |
|
| 227 |
+
| 2020 | [**SideGuide**](https://ieeexplore.ieee.org/document/9340734): A Large-scale Sidewalk Dataset for Guiding Impaired People | IROS 2020 |
|
| 228 |
+
|
| 229 |
+
</details>
|
| 230 |
+
|
| 231 |
+
> ์ ์ฒด ๋
ผ๋ฌธ ๋ชฉ๋ก ๋ฐ ์์ธ ๋ด์ฉ์ [๋ธ๋ก๊ทธ](https://selectstar.ai/blog/) ๋๋ [๋ฌธ์ํ๊ธฐ](https://selectstar.ai/contact_page/)๋ฅผ ํตํด ํ์ธํ์ค ์ ์์ต๋๋ค.
|
| 232 |
|
| 233 |
---
|
| 234 |
|
|
|
|
| 239 |
| ๐ Website | [selectstar.ai](https://selectstar.ai/) |
|
| 240 |
| ๐ฐ Blog | [selectstar.ai/blog](https://selectstar.ai/blog/) |
|
| 241 |
| ๐ผ Enterprise inquiries | [Contact form](https://selectstar.ai/contact_page/) |
|
| 242 |
+
| ๐ฌ Community | [Discussion tab](https://huggingface.co/spaces/datumo/README/discussions) |
|
| 243 |
+
| ๐ Updates | HuggingFace ํ๋ก์ฐ๋ก ์ ๋ฆด๋ฆฌ์ฆ ์๋ฆผ ๋ฐ๊ธฐ |
|
| 244 |
|
| 245 |
---
|
| 246 |
|
| 247 |
<div align="center">
|
| 248 |
+
<sub>โญ Building the data foundation for trustworthy AI ยท Made with care in Seoul ๐ฐ๐ท</sub>
|
| 249 |
</div>
|