Jinyeop Song
@jinyeop_song
Followers
101
Following
103
Media
1
Statuses
85
https://t.co/0gRM6ECkfY Physics PhD student@MIT - interested in complex biophysics & AI
Cambridge, MA, USA
Joined September 2021
Check out our new benchmark "SPOT" for assessing LLMs' ability to detect scientific errors in papers! Collaborative work led by @gson_AI.
#NLProc.AI Co-Scientists 🤖 can generate ideas, but can they spot mistakes? (not yet! 🚫). In my recent paper, we introduce SPOT, a dataset of STEM manuscripts (math, materials science, chemistry, physics, etc), annotated with real errors. SOTA models like o3, gemini-2.5-pro
0
0
4
RT @AtomSilverman: Another huge week in AI Agents 🧵. I summarized everything announced by LangChain, Replit, Cursor, Anthropic, Cobot, Agen….
0
555
0
🧩 Why do task vectors exist in pretrained LLMs?. Excited to share our new paper with @seungwookh on the emergence of abstractions (task vectors) for in-context learning (ICL)! . To read more :
🧩 Why do task vectors exist in pretrained LLMs? . Our new research uncovers how transformers form internal abstractions and the mechanisms behind in-context learning(ICL).
0
0
4
RT @seungwookh: 🧩 Why do task vectors exist in pretrained LLMs? . Our new research uncovers how transformers form internal abstractions an….
0
29
0
RT @maxhkw: I’m recruiting PhD students to join the Computational Minds and Machines Lab at the University of Washington in Seattle! Join u….
0
132
0
RT @Tea_Pearce: 1/ our analysis reconciling the conflicting LLM scaling laws of Kaplan (N ∝ C^0.73) and Chinchilla (N ∝ C^0.50) accepted to….
0
1
0
RT @cwolferesearch: Prompt engineering requires a lot of manual effort. Here are four automatic prompt optimization algorithms that can hel….
0
201
0
RT @KejunYing: 🧬 Our 2nd preprint today, . MethylGPT: A foundation model for the DNA methylome!. Our generative pretained transformer-based….
0
38
0
RT @BiologyAIDaily: MethylGPT: a foundation model for the DNA methylome. 1. MethylGPT introduces a transformer-based foundation model speci….
0
7
0
RT @jhnhw: First 1e6 integers, represented as binary vectors indicating their prime factors, and laid out using the sparse matrix support i….
0
2K
0
RT @LChoshen: Scaling laws predict🦣large models by training🦟small ones, cool right?.Fortunately, they are not that complicated or costly.at….
0
35
0
RT @0xRfnza: @HamsterCEO @StarMet_ Hamster Airdrop Season is coming! . 🟨🟨🟨🟨🟨🟨🟨🟨🟨⬜️ 90%. 👉 More activity in your….
0
25
0
RT @WilsonEsther7: @melllstroy_ Hamster Airdrop Season is coming! . 🟨🟨🟨🟨🟨🟨🟨🟨🟨⬜️ 90%. 👉 More activity in your wal….
0
35
0