Alex Gu Profile
Alex Gu

@minimario1729

Followers
4K
Following
4K
Media
149
Statuses
942

mit phd student (on job market!), llm for math+code / prev intern @ meta, nvidia, aws, jane street / enjoys 🎹✈️⛷️⛵

bay area
Joined March 2020
Don't wanna be here? Send us removal request.
@ADarmouni
Axel Darmouni
1 day
Lean model generated proofs can be optimized, especially by another model! A really cool work from @AIatMeta FAIR from @minimario1729 et al The main idea is to take a model that is first heavily finetuned on a Lean related dataset (with natural language information), and then
1
1
5
@minimario1729
Alex Gu
3 days
visiting seattle for a bit, let me know if you're around and want to have a chat :)
1
1
33
@minimario1729
Alex Gu
7 days
👥ProofOptimizer is work w/ awesome co-authors at Meta FAIR: Bartosz Piotrowski, @FabianGloeckle, @KaiyuYang4, @aramHmarkosyan We think this direction has a lot of potential. Check out our paper, reach out, and chat with us! 🏠 https://t.co/GXLQLqlV4E 📝
0
3
13
@minimario1729
Alex Gu
7 days
2) Simplified proofs often run faster, with 22/75 Putnam proofs achieving over 50% speedup. In our paper, we also optimize more directly for run-time instead than proof length with even better results!
1
1
8
@minimario1729
Alex Gu
7 days
🔮We discover two downstream effects of proof simplification: better training and faster run-time. 1) Training on simplified proofs can improve generation abilities compared to training on longer proofs
1
1
8
@minimario1729
Alex Gu
7 days
We tried ProofOptimizer on Seed-Prover's IMO 2025 proofs with an increased sampling budget to achieve an average proof length reduction of 49%! The proof checking time of two problems also went down, from 434 seconds -> 363 (P4) and 61 -> 34 (P5) ⚡
2
1
8
@minimario1729
Alex Gu
7 days
🤖Inference: we use a natural iterative shortening algorithm: take a proof, sample 64 times, take the shortest proof, and repeat. This already shows decent results! Check out our shortened proofs: https://t.co/cjuZIIgZRA
1
2
8
@minimario1729
Alex Gu
7 days
🚄Training: we use expert iteration and online RL: expert iteration leads to the best pass@32, and RL leads to the best pass@1. We report results simplifying Goedel-Prover-V2's (SoTA open-source) miniF2F and PutnamBench proofs. Our trained models surpass Gemini-2.5-Pro! 😎
1
1
8
@minimario1729
Alex Gu
7 days
⛏️Data: We develop a four-stage pipeline to automatically mine data for proof simplification: 1) high-quality problem collection (Goedel-Pset) 2) proof sketching 3) therorem extraction and filtering (remove with AUTO) 4) proof generation (Goedel-Prover-V2)
1
1
12
@minimario1729
Alex Gu
7 days
Towards mitigating long AI-generated Lean proofs, we provide an end-to-end data, training, and inference recipe for proof simplification. While we only consider proof length in this paper, our methods generalize to other measures as well (e.g. runtime).
1
1
10
@minimario1729
Alex Gu
7 days
✂️Introducing ProofOptimizer: a training and inference recipe for proof shortening! 😰AI-written formal proofs can be long and unreadable: Seed-Prover's proof of IMO '25 P1 is 16x longer in Lean vs. English. Our 7B shortens proofs generated by SoTA models by over 50%! 🧵⬇️
6
35
204
@TacoCohen
Taco Cohen
18 days
🚨 Attention aspiring PhD students 🚨 Meta / FAIR is looking for candidates for a joint academic/industry PhD! Keywords: AI for Math & Code. LLMs, RL, formal and informal reasoning. You will be co-advised by prof. @Amaury_Hayat from ecole des ponts and yours truly. You'll have
24
119
899
@minimario1729
Alex Gu
1 month
check out cwm!! open weights and inference code, many fun details to ponder from the report, and most importantly a lot of new research directions opened :)
@syhw
Gabriel Synnaeve
1 month
(🧵) Today, we release Meta Code World Model (CWM), a 32-billion-parameter dense LLM that enables novel research on improving code generation through agentic reasoning and planning with world models. https://t.co/BJSUCh2vtg
0
3
44
@FrogandToadbot
Frog and Toad Bot
3 months
TOAD WILL NOW PLAY THE PIANO VERY WELL.
0
6
55
@minimario1729
Alex Gu
3 months
Thanks to MIT News for covering our vision of AI for code! A lot of progress made, but still a long way to go!
@MIT_CSAIL
MIT CSAIL
3 months
Can AI actually code for us? 🧵 MIT research reveals there’s a "long way to go" due to bottlenecks like assessment, codebase scale, & incorrect retrievals. The work reflects a vision to let humans focus on high-level design while routine work is automated:
1
3
28
@minimario1729
Alex Gu
3 months
postering this work on behalf of awesome coauthors at the ai for math workshop tomorrow :)
@lupantech
Pan Lu
5 months
Do LLMs truly understand math proofs, or just guess? 🤔Our new study on #IneqMath dives deep into Olympiad-level inequality proofs & reveals a critical gap: LLMs are often good at finding answers, but struggle with rigorous, sound proofs. ➡️ https://t.co/h5f8Qv8Xlv To tackle
0
4
37
@minimario1729
Alex Gu
3 months
come to our ai for math workshop tomorrow it'll be super fun!! 🎉🎉
1
7
65
@minimario1729
Alex Gu
3 months
ai for math workshop papers released, it's a fun batch🚀 https://t.co/9MU4cxWzKc
1
2
19
@minimario1729
Alex Gu
3 months
poster info ⬇️ you know you wanna come 😀 https://t.co/1ddUpXQeey
0
2
7
@minimario1729
Alex Gu
3 months
come hang out at our poster tomorrow (tuesday) at 11am let's have some fun and savor the ai4code hype! 🎉 🥳 📌 East Exhibition Hall A-B #E-605
@minimario1729
Alex Gu
7 months
📢 Excited to share our new paper: Challenges and Paths Towards AI for SWE We discuss: 🛠️ 6 sub-tasks needed for SWE 🤖 9 challenges of today's AI in SWE 🔮 9 future directions to address the challenges w/ collaborators from MIT, Berkeley, Cornell, Stanford, and UPenn ⬇️ (1/n)
2
4
41