Andrew Drozdov
@mrdrozdov
Followers
3K
Following
13K
Media
615
Statuses
13K
Context Engineering (and Science!) for Knowledge Assistant @ Databricks
NYC
Joined August 2010
RT @HanSineng: Missing @aclmeeting but sending “ATEB: Rethinking Advanced NLP Tasks in an Information Retrieval Setting” in my place ! . Co….
0
4
0
RT @beirmug: TREC RAG 2025 official retrieval baselines are available now! 💥💥💥 . Time to start generating those answers and submit them to….
0
3
0
This aligns with #2 in the proposals I described at the #SIGIR2025 panel on “LLMs and IR”. Really cool to see a whole team forming to tackle this effort!.
We're launching an "AI psychiatry" team as part of interpretability efforts at Anthropic! We'll be researching phenomena like model personas, motivations, and situational awareness, and how they lead to spooky/unhinged behaviors. We're hiring - join us!
0
0
1
RT @ao_qu18465: 🚀 Excited to share my first tweet and to introduce our latest work: MEM1: RL for Memory Consolidation in Long-Horizon Agent….
0
10
0
RT @VoyageAI: 📢 voyage-context-3: contextualized chunk embeddings. - Auto captures of chunk level detail & global doc context, w/o metadata….
0
22
0
RT @ziv_ravid: We want to start a podcast about cutting-edge AI research and technical breakthroughs. Need a catchy name!.What would you ca….
0
7
0
RT @MohitIyyer: ChatGPT Agent is a huge step up on BearCubs, esp on multimodal/interactive tasks (e.g., playing web games)!. It gets 65.8%….
0
4
0
RT @cassidoo: I launched PocketCal on Product Hunt if y'all wouldn't mind passing along an upvote! ❤️.
producthunt.com
Pick dates, share a link, and that's your calendar. PocketCal is a free, open-source web app that lets you select dates and instantly share them. No accounts, just a link, and nothing in your way.
0
11
0
RT @wzhao_nlp: Silly but important question: what metrics do you look at / how to vibe-check your training runs are going well, specially u….
0
9
0
RT @lateinteraction: Haha. Well, if you want to read @SherylHsu02’s excellent last paper just before joining OpenAI, read LeReT at ICLR’25.….
0
25
0
RT @random_walker: Back in grad school, when I realized how the “marketplace of ideas” actually works, it felt like I’d found the cheat cod….
0
51
0
RT @mrdrozdov: Here's three of my more *controversial* proposals from the SIGIR / ICTIR 2025 panel on "LLMs + IR, what could possibly go wr….
0
2
0
A cool outcome here would be if future IMO exclusively includes problems that we know recent generation of LLMs can not yet solve.
6. I don’t think LLMs will replace mathematicians anytime soon. Math research is about solving problems *no one* yet knows how to solve (out-of-distribution), and this requires significant creativity, something notably absent from OpenAI’s IMO solutions. (6/10).
2
0
3
I’m interpreting this as pro-Human rather than anti-AI. Give people time and tools, and be amazed.
0
0
1
RT @lateinteraction: AegisLLM leverages DSPy's MIPROv2 optimizer in a totally unexpected way: to evolve its prompts based on the attacks it….
0
17
0
“Reasoning won’t generalize outside of math and code”. Maybe we should express everything as math and code…. Proofs, Theorems, Lemmas, Corollaries, Conjectures are all math. It’s not just equations. From this perspective we have a lot more flexibility of expression.
1
0
2
RT @MohitIyyer: Excited to talk about long-context models / eval at this panel on Saturday! I'm also looking for a postdoc / PhD students t….
0
11
0