Yi-Li Hsu

@jp22_5

Taipei, Taiwan Katılım Mart 2011

61 Takip Edilen4 Takipçiler

Yi-Li Hsu retweetledi

Justin Chih-Yao Chen@cyjustinchen·25 Eyl

Excited about my new paper! 😊 How to get better consensus among diverse LLMs? ReConcile, a discussion framework w/ (a) grouped ans+expl of each agent in prev round (b) uncertainties (c) corrective/convincing expls👉improves reasoning individually+collaboratively (& over GPT4)

Swarnadeep Saha@swarnaNLP

🚨ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs ChatGPT+Bard+Claude2 discuss & correctively convince each other for reasoning, even outperforming GPT4 (& GPT4 as agent improves by 10%)! arxiv.org/abs/2309.13007 @cyjustinchen @mohitban47 🧵

English

5.6K

Yi-Li Hsu retweetledi

Kenneth Huang #HCOMP2026 Sep27-30@DC@windx0303·1 Haz

We're launching the 1st Scientific Figure Captioning (SciCap) Challenge! We invite AI/NLP/CV researchers to build systems that caption all types of figures in arXiv papers. The challenge will be hosted at the CLVL workshop at #ICCV2023. Join us here: SciCap.AI

Kenneth Huang #HCOMP2026 Sep27-30@DC tweet media

English

9.5K

Yi-Li Hsu retweetledi

Kenneth Huang #HCOMP2026 Sep27-30@DC@windx0303·9 Nis

People often ask me how generalizable it is when I say, "Figure captions in arXiv papers are somewhat similar to summaries." We can figure it out together! We will organize the 1st 🔥SciCap Challenge🔥 at #ICCV2023. Stay tuned. The challenge will be launched in early May 2023!

Kenneth Huang #HCOMP2026 Sep27-30@DC@windx0303

🚨New Preprint Alert🚨 In our new preprint, we argue that generating captions for figures (line chart, bar chart, etc.) in scholarly papers is more of a *text summarization task* than a vision-to-language task. Paper: arxiv.org/abs/2302.12324 🧵1/n #NLProc #SciDoc

English

5.5K

Keşfet

@elonmusk @BarackObama @taylorswift13 @cristiano @BillGates @NASA @nikifrancismediavine @katyperry