Harshit Joshi
@harshitj__
CS phd @StanfordNLP, @StanfordOVAL | prev: @MSFTResearch | LLM systems for knowledge access, discovery and curation
.@stanfordnlp papers at @aclmeeting in Vienna next week: • HumT DumT: Measuring and controlling human-like language in LLMs @chengmyra1 @sunnyyuych @jurafsky • Controllable and Reliable Knowledge-Intensive Task Agents with Declarative GenieWorksheets @harshitj__ @ShichengGLiu…
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.
AI companions aren’t science fiction anymore 🤖💬❤️ Thousands are turning to AI chatbots for emotional connection – finding comfort, sharing secrets, and even falling in love. But as AI companionship grows, the line between real and artificial relationships blurs. 📰 “Can A.I.…
plis give them a place to say so that they can do great work 🫡
anybody have a place in SF to rent/sublease till mid-September? for me and @ZhengxuanZenWu (or either of us individually) need to move ASAP
What if LLMs could learn your habits and preferences well enough (across any context!) to anticipate your needs? In a new paper, we present the General User Model (GUM): a model of you built from just your everyday computer use. 🧵
i was recently told that i cannot angel invest 10k in a very promising startup because i do not have at least 30k followers 😭 😭
Happy Throughput Thursday! We’re excited to release Tokasaurus: an LLM inference engine designed from the ground up for high-throughput workloads with large and small models. (Joint work with @achakravarthy01, @ryansehrlich, @EyubogluSabri, @brad19brown, @jshetaye,…
I am hiring a Full time research engineer II and a senior research engineer for my team @zomato with the high level goal of training custom LLMs and deploying them at scale. More details in the 🧵
This year, there have been various pieces of evidence that AI agents are starting to be able to conduct scientific research and produce papers end-to-end, at a level where some of these generated papers were already accepted by top-tier conferences/workshops. Intology’s…
Bro put in so much effort into this paper and did such amazing analysis that you dont wanna know what happened to him after the submission ☠️
new paper! 🫡 why are state space models (SSMs) worse than Transformers at recall over their context? this is a question about the mechanisms underlying model behaviour: therefore, we propose using mechanistic evaluations to answer it!
40% with just 1 try per task: SWE-agent-LM-32B is the new #1 open source model on SWE-bench Verified. We built it by synthesizing a ton of agentic training data from 100+ Python repos. Today we’re open-sourcing the toolkit that made it happen: SWE-smith.
nikil is one of the smartest person i have met in my life. he is also very spontaneous
please follow @NikilSelvam
🚀Introducing CRUST-Bench, a dataset for C-to-Rust transpilation for full codebases 🛠️ A dataset of 100 real-world C repositories across various domains, each paired with: 🦀 Handwritten safe Rust interfaces. 🧪 Rust test cases to validate correctness. 🧵[1/6]