Noah Provenzano

senior scratch engineer

[Sentient @SentientAGI]

AI researchers @noahpro99 and Christian Calvo from @virginia_tech will be presenting EvoSkill, a paper they co-authored with the Sentient team.

Catch them today at @COLM_conf from 11 AM–1 PM PT at Imperial Ballroom Poster #52, Poster Session 5.

[Tu Vu @tuvllms]

@noahpro99 also, my awesome PhD student @Quyet_Azir is presenting his work on curating long-context reasoning data today (Oct 6) 11 AM-1 PM PT @ Grand Ballroom Poster #134 Poster Session 1.

https://t.co/eOfWulyca7

[Tu Vu @tuvllms]

I'm not at @COLM_conf, but my awesome PhD student @noahpro99 is giving a talk on our EvoSkill work today at the COLM Social: Self-Improving Agents (Oct 6 1-2:30 PM PT, Franciscan D). Come chat with him about his first-hand experience with different skill evolution frameworks!

[Noah Provenzano @noahpro99]

@ArcanesValor @mrginden @morallawwithin 😂

[Noah Provenzano @noahpro99]

@Sub_RBala Congratulations @Sub_RBala

[Noah Provenzano @noahpro99]

@kavanav2912 Congrats!

[Tu Vu @tuvllms]

Glad to share that our paper on The Master Key Hypothesis / cross-model capability transfer got accepted @NeurIPSConf 2026 as spotlight 🎉 !! Top 1.3% (292 spotlight + 112 oral) out of 30709 submissions. Congrats to my students @Sub_RBala @linusdd44804 @helix9550 and co-authors @mohitban47!!

[Deedy @deedydas]

You can now generate an entire 3blue1brown style video from any research paper with Opus 5.5.

Here’s a 8min video summary of “Regularized Recursive Self Improvement of Agent Harnesses”.

The 90%ile educational YouTuber is fully automated.

[Noah Provenzano @noahpro99]

Trump when he said "no new wars"

[Helix @helix9550]

🚨New paper alert:
Can dense supervision control a looped language model, if the loss never sees how large its hidden state grows?
Add loss at every loop, yet the norms blow up into thousands.
The decode layer hides the hidden state’s size from the loss. The readout blind spot 🧵

[The Sanghani Center at Virginia Tech @SanghaniCtrVT]

Chris Thomas (ctr), core faculty @SanghaniCtrVT, shares pics from #CVPR2026. His grad students (l to r) @najib_haque, Zaber Hakim, Ali Asgarov, @Hanibalomari.
2 research papers from Chris’s group at conference:
🔗 https://t.co/sA2CX1JNZy (Highlight)
🔗 https://t.co/cW7XLHiSYg

[The Sanghani Center at Virginia Tech @SanghaniCtrVT]

Ph.D. student Thinh Pham (@thinhphp_vt) @VT_CS @SanghaniCtrVT presented his collaborative work, "SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models," during a poster session at #ICLR2026 in Rio de Janeiro.
🔗to paper: https://t.co/42SmDlBLFs

[Thinh @thinhphp_vt]

Come checkout our posters at #ICLR2026 🇧🇷!

April 23, 10:30 AM - 1:00 PM
📍Pavilion 3 P3-#1424
SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models (https://t.co/3YkR5SzpRg)

I'm also looking for opportunities this summer and would love to connect and chat!

[Helix @helix9550]

Built a tool for this.
ghostenv encrypts your .env and gives AI agents masked (fake) values. Real secrets only get injected when you explicitly approve. Agents can use your API keys without ever seeing them.

https://t.co/ze7YVVNhDx

[Noah Provenzano @noahpro99]

@rezoundous 😂

[Tu Vu @tuvllms]

🚨✨ New preprint ✨🚨

Can we transfer post-training capabilities across model families, scales, and releases?

Our new work on UNLOCK, led by my PhD student @Sub_RBala, suggests that latent capabilities correspond to low-dimensional directions in latent space, which can be extracted as steering vectors and transferred across models via simple linear transformations. UNLOCK requires no training or labeled contrastive examples, only forward passes.

Most exciting result: Qwen3-14B-Base + transferred math reasoning from Qwen3-4B approaches the performance of Qwen3-14B (post-trained) without any training.

📜: https://t.co/ScVBxe0HMh
💻: https://t.co/HUtF3Symdq
🤝: w/ wonderful students & collaborators @Sub_RBala @linusdd44804 @SharmaRituraj19 @anjiefang, Fardin Abdi, Viktor Rozgic, Zheng Du, and @mohitban47
🧵: Check out @Sub_RBala's thread below 👇

[The Sanghani Center at Virginia Tech @SanghaniCtrVT]

While large language models dominate the artificial intelligence discourse, @xwang174 and @tuvllms, core faculty @SanghaniCtrVT and a team of grad students are looking to their smaller cousins for a less energy intensive, meaningful alternative.
Read more: https://t.co/G09u733T6J

[Noah Provenzano @noahpro99]

Can't get enough of @forestermusic!

[Google DeepMind @GoogleDeepMind]

Meet Gemma 4: our new family of open models you can run on your own hardware.

Built for advanced reasoning and agentic workflows, we’re releasing them under an Apache 2.0 license. Here’s what’s new 🧵

[Noah Provenzano @noahpro99]

We're all holding on by a single together ai key