๐Ÿ“Š

Intentable: A Mixed-Initiative System for Intent-Based Chart Captioning

Intentable: A Mixed-Initiative System for Intent-Based Chart Captioning

Jiwon Choi and Jaemin Jo
Proceedings of Conference on 2022 IEEE Visualization & Visual Analytics (IEEE VIS)
We present Intentable, a mixed-initiative caption authoring system that allows the author to steer an automatic caption generation process to reflect their intents on captions. We first derive grammar for specifying the intents, i.e., a caption recipe, and build a neural network that generates caption sentences given a recipe. Our quantitative evaluation revealed that our intent-based generation system not only allows the author to engage in the generation process but also produces more fluent captions than the previous end-to-end approaches without user intervention. Finally, we demonstrate the versatility of our system such as context adaptation, unit conversion, and sentence reordering.
Intentable
jason-choi

My Work

โ€ข
Research
โ—ฆ
์—ฐ๊ตฌ ์ „๋ฐ˜ (์ฃผ์ €์ž)
โ—ฆ
Chart Captioning์„ ์œ„ํ•œ DSL (Domain Specific Language) ๋””์ž์ธ
โ—ฆ
์˜๋„ ๊ธฐ๋ฐ˜ ์ฐจํŠธ ์บก์…”๋‹์„ ์œ„ํ•œ ์ธํ„ฐํŽ˜์ด์Šค ๋””์ž์ธ
โ€ข
Frontend Development
โ—ฆ
Next.js , ChakraUI ๊ธฐ๋ฐ˜ SPA ์›น ์•ฑ
โ—ฆ
Recoil์„ ์ด์šฉํ•œ ์ „์—ญ ์ƒํƒœ ๊ด€๋ฆฌ
โ—ฆ
visx๋ฅผ ์ด์šฉํ•œ ์ธํ„ฐ๋ž™ํ‹ฐ๋ธŒ ์ฐจํŠธ ์‹œ๊ฐํ™”
โ€ข
Backend Development
โ—ฆ
T5 ๋ชจ๋ธ ์„œ๋น™
โ—ฆ
FastAPI
โ€ข
AI Research
โ—ฆ
ํ•™์Šต ๋ฐ์ดํ„ฐ (500K) ํฌ๋กค๋ง ๋ฐ ์ „์ฒ˜๋ฆฌ
โ—ฆ
Controllable Generation์„ ์œ„ํ•œ Data Processing ๋ฐ Augmentation
โ—ฆ
T5 ๋ชจ๋ธ Fine-tuning