Anton Baumann

[Sasha Rush @srush_nlp]

On-Policy Distillation is the most active new research direction being explored in RL for LLMs. Had the chance to discuss how it works with Dwarkesh and why it fits so nicely into large-scale pipelines.

[Ronak Malde @rronak_]

We have been exploring new algorithmic frontiers and are excited to share our contributions to Self Distillation Policy Optimization (SDPO) for agentic continual learning, check out our blog post here:

https://t.co/5xjL02jtUz

[Sasha Rush @srush_nlp]

Been working on text feedback / OPSD in Composer. Really interesting space, and much more to be explored.

[Jonas Hübotter @jonashubotter]

Self-distillation for long-horizon training at scale!

[Jonas Hübotter @jonashubotter]

Today and tomorrow we’ll be presenting self-distillation with orals at ICLR in Rio 🇧🇷

1. “Self-Distillation enables Continual Learning” at lifelong agents workshop (Sun 11:30am)
2. “Reinforcement Learning via Self-Distillation” at scaling post-training workshop (Mon 2:40pm)
3. “Test-Time Self-Distillation” at test-time updates workshop (Mon 4:15pm)

[Jonas Hübotter @jonashubotter]

Just came across this great discussion of self-distillation on @latentspacepod! Really good run down by Ted Kyi and we’re every bit excited about what’s next as he is!

https://t.co/G5LrWlOT8B

[Explainable Machine Learning @ExplainableML]

3/
Post-hoc Probabilistic Vision-Language Models
@_antonbaumann, @ruili_pml, Marcus Klasson, Santeri Mentu, @ShyamgopalKart1, @zeynepakata, @arnosolin, Martin Trapp
[Paper]: https://t.co/kYxKZSe1x9
[Project]: https://t.co/p7TzhvZQrn
[Code]: https://t.co/UPnWQZmlvw

[Jonas Hübotter @jonashubotter]

Training LLMs with verifiable rewards uses 1bit signal per generated response. This hides why the model failed.

Today, we introduce a simple algorithm that enables the model to learn from any rich feedback!
And then turns it into dense supervision.

(1/n)

[Anton Baumann @_antonbaumann]

@OverwatchEU Habt ihr schon an Junkensteins Tür geklopft? Es gibt eine Overwatch-PS4 zu gewinnen! https://t.co/8bgTYj5kuM #OWHalloween3