AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Zero-Shot Adaptation of Behavioral Foundation Models to Unseen Dynamics

Maksim Bobrin, Ilya Zisman, Alexander Nikulin, Vladislav Kurenkov, Dmitry V. Dylov

ICLR 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Robust fine-tuning of zero-shot models · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add transformer belief encoder to FB (Belief-FB)Vanilla-FB · Randomized-FourRooms (test)zero-shot return: 0.15 -> 0.4Improved
add transformer belief encoder to FB (Belief-FB)Vanilla-FB · Randomized-PointMass (test)zero-shot return: 0.1 -> 0.45Improved
add transformer belief encoder to FB (Belief-FB)Vanilla-FB · Ant-Wind (test)zero-shot return: 250 -> 550Improved
add vMF task-vector sampling around context (Rotation-FB) (ablation)Belief-FB · Randomized-FourRooms (test)zero-shot return: 0.4 -> 0.61Improved
add vMF task-vector sampling around context (Rotation-FB) (ablation)Belief-FB · Randomized-PointMass (test)zero-shot return: 0.45 -> 0.55Improved
add vMF task-vector sampling around context (Rotation-FB) (ablation)Belief-FB · Ant-Wind (test)zero-shot return: 550 -> 640Improved
add vMF task-vector sampling around context (Rotation-FB) (ablation)Belief-FB · OGBench Scene (test)zero-shot return: 0.45 -> 0.55Improved
use oracle environment ID instead of inferred beliefFB · Randomized-FourRoomszero-shot return: 0.90 -> 0.10 (train -> test)Mixed
increase number of training layouts beyond 25 (ablation)Belief-FB · Randomized-FourRoomszero-shot return: <3%No clear effect
increase transformer context length beyond a single episode (100 steps) (ablation)Belief-FB · Randomized-FourRooms and PointMasszero-shot returnNo clear effect
use context length shorter than one episode (50 vs 100 steps) (ablation)Belief-FB · Randomized-FourRooms and PointMasszero-shot returnHurt
condition backward network B on context (ablation)Belief-FB · Randomized-FourRoomszero-shot returnHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy