2 slots peft!
Good day Bluesky!
Im opening a 3 comms slot today! If you're interested on grabbing a slot, you can contact me thru DM and also telegram @/kousukebull (telegram preferable)
Dont forget to read my ToS beforehand!
Thank you!
#furry
#commission
#furryart
Im opening a 3 comms slot today! If you're interested on grabbing a slot, you can contact me thru DM and also telegram @/kousukebull (telegram preferable)
Dont forget to read my ToS beforehand!
Thank you!
#furry
#commission
#furryart
December 9, 2024 at 3:35 AM
2 slots peft!
This bitch said radical peft.
January 25, 2026 at 3:21 PM
This bitch said radical peft.
Is PEFT new RAG? Who knows, but interesting.
When you train agents to combine knowledge in weights, text memory, and search on realistic legal work; you can achieve higher intelligence-per-token than the best frontier models with a remarkably small dense model, Qwen3.8-27B.
Understanding a Law Firm through Study
engram.com/blog/legal-a...
Understanding a Law Firm through Study
engram.com/blog/legal-a...
Understanding a Law Firm through Study
Training agents to combine knowledge in weights, text memory, and search on realistic legal work.
engram.com
August 18, 2026 at 6:37 PM
Is PEFT new RAG? Who knows, but interesting.
Never before has the US government attributed internal dissent to dsinister radical peft conspiracies and foreign agitators. Wikipedia link unrelated.
en.wikipedia.org/wiki/Red_Scare
en.wikipedia.org/wiki/Red_Scare
January 25, 2026 at 3:15 PM
Never before has the US government attributed internal dissent to dsinister radical peft conspiracies and foreign agitators. Wikipedia link unrelated.
en.wikipedia.org/wiki/Red_Scare
en.wikipedia.org/wiki/Red_Scare
Trump ATTACKS Elon Musk in SHOCKING Cabinet Meeting
YouTube video by Keith Edwards
youtube.com
April 11, 2025 at 6:32 AM
Working on finetuning the nvidia/NV-Embed-v2 embedding model with PEFT for the new Sentence Transformers 🤝 PEFT Documentation. Excited to see if this works well - I don't usually finetune such large embedding models.
November 22, 2024 at 8:52 AM
Working on finetuning the nvidia/NV-Embed-v2 embedding model with PEFT for the new Sentence Transformers 🤝 PEFT Documentation. Excited to see if this works well - I don't usually finetune such large embedding models.
Thabks so kuch to all who popped in last night! Only these few peft already!
www.celestineandthehare.com/new-collection
www.celestineandthehare.com/new-collection
August 24, 2025 at 12:56 PM
Thabks so kuch to all who popped in last night! Only these few peft already!
www.celestineandthehare.com/new-collection
www.celestineandthehare.com/new-collection
Ooh and LAST Sketchbook copies peft in my store 😍🫣 theres like only 2 of each i believe 🫶🏻🥹 come get yourrsss!!!
theartofmathew.bigcartel.com
theartofmathew.bigcartel.com
June 12, 2026 at 7:58 PM
Ooh and LAST Sketchbook copies peft in my store 😍🫣 theres like only 2 of each i believe 🫶🏻🥹 come get yourrsss!!!
theartofmathew.bigcartel.com
theartofmathew.bigcartel.com
mnemonic for remembering polly's eye colors: Pink on her Peft, Orange on her Oright
March 23, 2026 at 10:46 PM
mnemonic for remembering polly's eye colors: Pink on her Peft, Orange on her Oright
…. ex… excuse me?? 😭😭🙏 ALL BC OF GIVEN IN THE TOP PEFT CORNER?????
July 24, 2025 at 1:25 PM
…. ex… excuse me?? 😭😭🙏 ALL BC OF GIVEN IN THE TOP PEFT CORNER?????
A survey of PEFT papers within the context of Foundation Models .
"Parameter-Efficient Fine-Tuning for Foundation Models"
Project: awesome-peft-for-foundation-models.github.io
Paper: arxiv.org/abs/2501.13787
Repo: github.com/THUDM/Awesom...
"Parameter-Efficient Fine-Tuning for Foundation Models"
Project: awesome-peft-for-foundation-models.github.io
Paper: arxiv.org/abs/2501.13787
Repo: github.com/THUDM/Awesom...
January 24, 2025 at 7:07 AM
A survey of PEFT papers within the context of Foundation Models .
"Parameter-Efficient Fine-Tuning for Foundation Models"
Project: awesome-peft-for-foundation-models.github.io
Paper: arxiv.org/abs/2501.13787
Repo: github.com/THUDM/Awesom...
"Parameter-Efficient Fine-Tuning for Foundation Models"
Project: awesome-peft-for-foundation-models.github.io
Paper: arxiv.org/abs/2501.13787
Repo: github.com/THUDM/Awesom...
rly like what i've done with my hair :)
May 15, 2024 at 8:52 AM
rly like what i've done with my hair :)
New short paper, to appear at NoDaLiDa 2025! I’ve written up findings from ablation studies on language adaptation of small LLMs with limited data/compute using different PEFT methods and setups: arxiv.org/abs/2412.12674
Train More Parameters But Mind Their Placement: Insights into Language Adaptation with PEFT
Smaller LLMs still face significant challenges even in medium-resourced languages, particularly when it comes to language-specific knowledge -- a problem not easily resolved with machine-translated da...
arxiv.org
December 18, 2024 at 9:15 AM
New short paper, to appear at NoDaLiDa 2025! I’ve written up findings from ablation studies on language adaptation of small LLMs with limited data/compute using different PEFT methods and setups: arxiv.org/abs/2412.12674
ChatGPT is short for Chat Grand Peft Thauto
June 11, 2025 at 11:37 AM
ChatGPT is short for Chat Grand Peft Thauto
Are you still using LoRA to fine-tune your LLM? 2024 has seen an explosion of new parameter-efficient fine tuning technique (PEFT), thanks to clever uses of the singular value decomposition (SVD). Let's dive into the alphabet soup: SVF, SVFT, MiLoRA, PiSSA, LoRA-XS 🤯...
February 20, 2025 at 12:38 PM
Are you still using LoRA to fine-tune your LLM? 2024 has seen an explosion of new parameter-efficient fine tuning technique (PEFT), thanks to clever uses of the singular value decomposition (SVD). Let's dive into the alphabet soup: SVF, SVFT, MiLoRA, PiSSA, LoRA-XS 🤯...
I peft from the clouds
Και γιατί να λυπηθώ τον Αμερικάνο στρατιωτικό που εκτέλεσε την εντολή;
Κτήνη!
Και γιατί να λυπηθώ τον Αμερικάνο στρατιωτικό που εκτέλεσε την εντολή;
Κτήνη!
⭕ New York Times: Πύραυλος Tomahawk των ΗΠΑ πίσω από τον βομβαρδισμό σχολείου στο Ιράν
New York Times: Πύραυλος Tomahawk των ΗΠΑ πίσω από τον βομβαρδισμό σχολείου στο Ιράν - Documento
Το βίντεο, αναρτήθηκε την Κυριακή από το πρακτορείο ειδήσεων Mehr News Agency του Ιράν.
www.documentonews.gr
March 9, 2026 at 9:48 AM
I peft from the clouds
Και γιατί να λυπηθώ τον Αμερικάνο στρατιωτικό που εκτέλεσε την εντολή;
Κτήνη!
Και γιατί να λυπηθώ τον Αμερικάνο στρατιωτικό που εκτέλεσε την εντολή;
Κτήνη!
🚨I’m traveling to #NeurIPS2024 next week to present these papers.
(1/3) 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐝 𝐔𝐧𝐫𝐞𝐬𝐭𝐫𝐢𝐜𝐭𝐞𝐝-𝐑𝐚𝐧𝐤 𝐌𝐚𝐭𝐫𝐢𝐜𝐞𝐬 𝐟𝐨𝐫 𝐏𝐄𝐅𝐓
A new PEFT method replacing low-rank matrices (LoRA) with more expressive structured matrices
arxiv.org/abs/2406.17740
(1/3) 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐝 𝐔𝐧𝐫𝐞𝐬𝐭𝐫𝐢𝐜𝐭𝐞𝐝-𝐑𝐚𝐧𝐤 𝐌𝐚𝐭𝐫𝐢𝐜𝐞𝐬 𝐟𝐨𝐫 𝐏𝐄𝐅𝐓
A new PEFT method replacing low-rank matrices (LoRA) with more expressive structured matrices
arxiv.org/abs/2406.17740
December 6, 2024 at 7:24 PM
🚨I’m traveling to #NeurIPS2024 next week to present these papers.
(1/3) 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐝 𝐔𝐧𝐫𝐞𝐬𝐭𝐫𝐢𝐜𝐭𝐞𝐝-𝐑𝐚𝐧𝐤 𝐌𝐚𝐭𝐫𝐢𝐜𝐞𝐬 𝐟𝐨𝐫 𝐏𝐄𝐅𝐓
A new PEFT method replacing low-rank matrices (LoRA) with more expressive structured matrices
arxiv.org/abs/2406.17740
(1/3) 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐝 𝐔𝐧𝐫𝐞𝐬𝐭𝐫𝐢𝐜𝐭𝐞𝐝-𝐑𝐚𝐧𝐤 𝐌𝐚𝐭𝐫𝐢𝐜𝐞𝐬 𝐟𝐨𝐫 𝐏𝐄𝐅𝐓
A new PEFT method replacing low-rank matrices (LoRA) with more expressive structured matrices
arxiv.org/abs/2406.17740
Sakana.ai's Transformer² arxiv.org/abs/2501.06252 paper features a cool new parameter-efficient fine-tuning (PEFT) technique that makes tuned models composable! Let’s dive in 💦.
(They have stunning artwork on their website 🤩 too)
(They have stunning artwork on their website 🤩 too)
February 10, 2025 at 10:04 PM
Sakana.ai's Transformer² arxiv.org/abs/2501.06252 paper features a cool new parameter-efficient fine-tuning (PEFT) technique that makes tuned models composable! Let’s dive in 💦.
(They have stunning artwork on their website 🤩 too)
(They have stunning artwork on their website 🤩 too)
8/IPA is just a first step. We believe that a deeper understanding of the feature space is important to unlocking better model adaptation.💭
To try it out and find more details👇:
Arxiv: arxiv.org/abs/2509.04398
Code: github.com/valeoai/peft...
To try it out and find more details👇:
Arxiv: arxiv.org/abs/2509.04398
Code: github.com/valeoai/peft...
GitHub - valeoai/peft-ipa
Contribute to valeoai/peft-ipa development by creating an account on GitHub.
github.com
December 2, 2025 at 11:11 AM
8/IPA is just a first step. We believe that a deeper understanding of the feature space is important to unlocking better model adaptation.💭
To try it out and find more details👇:
Arxiv: arxiv.org/abs/2509.04398
Code: github.com/valeoai/peft...
To try it out and find more details👇:
Arxiv: arxiv.org/abs/2509.04398
Code: github.com/valeoai/peft...
Going through and scanning all of the Fisk and cone sketches I have made and with me here. Many were peft behind. But! For once you can see them in high detail instead of the phone captures in the past! It's very exciting but also heart warming to see that adventure unfold again.
July 5, 2025 at 6:34 PM
Going through and scanning all of the Fisk and cone sketches I have made and with me here. Many were peft behind. But! For once you can see them in high detail instead of the phone captures in the past! It's very exciting but also heart warming to see that adventure unfold again.
1/Serve your PEFT with a fresh IPA!🍺
Finetuning large models is cheaper thanks to LoRA, but is its random init optimal?🤔
Meet IPA: a feature-aware alternative to random projections
#NeurIPS2025 WS #CCFM Oral+Best Paper
Work w/
S. Venkataramanan @tuanhungvu.bsky.social @abursuc.bsky.social M. Cord
🧵
Finetuning large models is cheaper thanks to LoRA, but is its random init optimal?🤔
Meet IPA: a feature-aware alternative to random projections
#NeurIPS2025 WS #CCFM Oral+Best Paper
Work w/
S. Venkataramanan @tuanhungvu.bsky.social @abursuc.bsky.social M. Cord
🧵
December 2, 2025 at 11:11 AM
1/Serve your PEFT with a fresh IPA!🍺
Finetuning large models is cheaper thanks to LoRA, but is its random init optimal?🤔
Meet IPA: a feature-aware alternative to random projections
#NeurIPS2025 WS #CCFM Oral+Best Paper
Work w/
S. Venkataramanan @tuanhungvu.bsky.social @abursuc.bsky.social M. Cord
🧵
Finetuning large models is cheaper thanks to LoRA, but is its random init optimal?🤔
Meet IPA: a feature-aware alternative to random projections
#NeurIPS2025 WS #CCFM Oral+Best Paper
Work w/
S. Venkataramanan @tuanhungvu.bsky.social @abursuc.bsky.social M. Cord
🧵
Become an LLM expert with a free in-IDE course from JetBrains Academy. Learn how modern LLMs work and how to apply them in real projects, from NLP basics to RAG, vector databases, LoRA, and PEFT. Take the course. 👇
jb.gg/ecr9co
jb.gg/ecr9co
April 10, 2026 at 8:41 AM
Become an LLM expert with a free in-IDE course from JetBrains Academy. Learn how modern LLMs work and how to apply them in real projects, from NLP basics to RAG, vector databases, LoRA, and PEFT. Take the course. 👇
jb.gg/ecr9co
jb.gg/ecr9co
🆕 LoRA vs Adapter vs Prefix Tuning: PEFT Memory Comparison
#PaperReview #AI #LoRA #PEFT
https://tildalice.io/lora-vs-adapter-vs-prefix-tuning-peft-comparison/
#PaperReview #AI #LoRA #PEFT
https://tildalice.io/lora-vs-adapter-vs-prefix-tuning-peft-comparison/
May 22, 2026 at 9:05 PM
🆕 LoRA vs Adapter vs Prefix Tuning: PEFT Memory Comparison
#PaperReview #AI #LoRA #PEFT
https://tildalice.io/lora-vs-adapter-vs-prefix-tuning-peft-comparison/
#PaperReview #AI #LoRA #PEFT
https://tildalice.io/lora-vs-adapter-vs-prefix-tuning-peft-comparison/
why are we dooming about ymmt pacing again we still have 1.5 episodes peft to resolve viomiya
August 28, 2026 at 3:25 AM
why are we dooming about ymmt pacing again we still have 1.5 episodes peft to resolve viomiya