Thomas Rupf

Master's student @ ETH interested in ML, RL, and Robotics.

[Thomas Rupf @th_rupf]

If you're at ICLR🇧🇷 come check out OpTI-BFM at poster P4-4504 tomorrow morning or as an oral at 4:03pm in the Amphitheater.

[Thomas Rupf @th_rupf]

Excited to share that our paper "Optimistic Task Inference for Behavior Foundation Models" was accepted for ICLR 2026.

BFMs are great at zero-shot RL, but task inference requires a dataset with reward labels. Our method OpTI-BFM offers an online alternative.

(1/5)

[Marco Bagatella @mar_baga]

Which representations are meaningful for control?
We're presenting TD-JEPA as an oral at ICLR🇧🇷: a zero-shot reinforcement learning algorithm using self-prediction (JEPA) to learn representations that are predictive of long-term, policy-dependent behavior. It works pretty well!🧵

[Thomas Rupf @th_rupf]

Huge thanks to my co-authors @mar_baga, @vlastelicap, @arkrause!

Paper: https://t.co/daIR9WFRPO

(5/5)

[Thomas Rupf @th_rupf]

Excited to share that our paper "Optimistic Task Inference for Behavior Foundation Models" was accepted for ICLR 2026.

BFMs are great at zero-shot RL, but task inference requires a dataset with reward labels. Our method OpTI-BFM offers an online alternative.

(1/5)

[Thomas Rupf @th_rupf]

OpTI-BFM bares similarities with LinUCB for Bandits which we use to prove sublinear regret in episodic settings under mild assumptions.

Because it's online, OpTI-BFM can also adapt to time-varying (non-stationary) rewards by decaying the weight on older observations.

(4/5)

[Núria Armengol @NriaArmengol2]

Last week I presented our last work: 🐝“Epistemically-guided forward backward exploration (FBEE)”🐝 at the @RL_Conference

TLDR: Active learning for unsupervised RL

[Marco Bagatella @mar_baga]

When multiple tasks need improvements, fine-tuning a generalist policy becomes tricky. How do we allocate a demonstration budget across a set of tasks of varied difficulty and familiarity?

We are presenting a possible solution at ICML on Wednesday!

(1/3)

[Thomas Rupf @th_rupf]

If this sounds interesting, come by on Tuesday, 4:30 pm ��� 7:00 pm, West Exhibition Hall B2–B3 #W-618.

Collaborators: @mar_baga, @nicoguertler, @JonasFrey96, @GMartius

https://t.co/Cv4ZP6UYew

(3/3)

[Thomas Rupf @th_rupf]

Zero-shot imitation from just a single sparse demonstration is hard. Goal-conditioned methods tend to “greedily" move from one state to the next and lose the big picture.

We're presenting an alternative approach on Tuesday at #ICML2025.

(1/3)

[Thomas Rupf @th_rupf]

Our method tackles the occupancy matching objective directly at test-time by estimating the agent's occupancy with samples from a learned world model and matching it to the expert occupancy using Optimal Transport.

(2/3)