Diyi Yang
@Diyi_Yang
Followers
18K
Following
12K
Media
143
Statuses
2K
Assistant Professor @Stanford CS @StanfordNLP @StanfordAILab LLMs for Humans
Joined December 2016
Our study led by @ChengleiSi reveals an “ideation–execution gap” 😲. Ideas from LLMs may sound novel, but when experts spend 100+ hrs executing them, they flop: 💥. 👉 human‑generated ideas outperform on novelty, excitement, effectiveness & overall quality!.
Are AI scientists already better than human researchers?. We recruited 43 PhD students to spend 3 months executing research ideas proposed by an LLM agent vs human experts. Main finding: LLM ideas result in worse projects than human ideas.
5
29
156
RT @stanfordnlp: The @stanfordnlp founders won both 2025 @aclmeeting test of time awards:. ▪25 yrs: Gildea & @jurafsky, Automatic Labeling….
0
20
0
RT @michaelryan207: Presenting this today! 🎉 #ACL2025NLP . Come by Poster Session 2 in Hall X4/X5 @ 10:30-12pm today to chat with me about….
0
10
0
RT @chrmanning: But they couldn’t keep me out of the Wednesday session! Plenty of space for all at 9am after last night’s social event! htt….
0
2
0
RT @thamar_solorio: @aclmeeting: Hope everyone had a great time at the social (💃💃💃) and that people go back to their hotels at a reasonable….
0
2
0
Come to #307 to see the live performance from Phil on teaching VLMs to handle flowers appropriately with real flowers 🌺
Come by poster #307 at @aclmeeting to learn about benchmarking embodied agent social norms + get handed flowers from the wonderful Phil Cuvin!
2
4
52
RT @michaelryan207: Interested in converting your text LLM to a speech LLM with no instruction tuning data? 🔊. Built a speech model but not….
0
5
0
RT @WilliamBarrHeld: If you are interested in personalization in LLMs, you're going to want to rush to @michaelryan207's poster in Hall 5!….
0
5
0
RT @dorazhao9: If you’re interested in empirical work measuring the relationship between AI companionship and user well-being, check out ou….
0
8
0
RT @_akhaliq: When Models Know More Than They Can Explain. Quantifying Knowledge Transfer in Human-AI Collaboration
0
47
0
RT @chrmanning: .@mlapata and I totally failed to be able to learn about Human-Centered NLP at @aclmeeting. 😢. Why do they always put the m….
0
9
0
RT @WilliamBarrHeld: Want to talk to an expert on AI x Cyber security? Well, unfortunately @StevenyzZhang isn't here due to visa issues. ….
0
14
0
We’re at #ACL2025 in Vienna!!! @WilliamBarrHeld @michaelryan207 @dorazhao9 @oshaikh13 . Catch us at our poster/talk and let’s chat 🔥😀
2
15
106
RT @WilliamBarrHeld: I'm in Vienna for #ACL2025!. My work is all presented tomorrow, but today you'll find me today at the poster session f….
0
3
0
RT @oshaikh13: BREAKING NEWS! Most people aren’t prompting models with IMO problems :). They’re prompting with tasks that need more context….
0
51
0
RT @simi_97k: So excited to be one of the five winners of the Imminent Translated Research Grants! This is for work done with @OpenNLPLabs….
0
10
0
RT @tongshuangwu: We all agree that AI models/agents should augment humans instead of replace us in many cases. But how do we pick when to….
0
18
0
RT @dorazhao9: While we’re building amazing new human-AI systems, how do we actually know if they work well for people?. In our #ACL2025 Fi….
0
24
0
RT @stanfordnlp: .@stanfordnlp papers at @aclmeeting in Vienna next week:.• HumT DumT: Measuring and controlling human-like language in LLM….
0
20
0