@svlevine: Flow reversal steering allows "steering" diffusion-based VLAs with high-level actions, for example from VLM reasoning. …
Summary
Flow reversal steering enables steering diffusion-based vision-language-action models with high-level actions, such as from VLM reasoning, and allows RL in diffusion noise space for task exploration.
View Cached Full Text
Cached at: 06/12/26, 04:51 AM
Flow reversal steering allows “steering” diffusion-based VLAs with high-level actions, for example from VLM reasoning. This also lets us run RL in the diffusion noise space with exploration guided by high-level reasoning: think through a task, then practice it! 👇 https://t.co/T9hgTozpuR
Similar Articles
Beyond Steering Vector: Flow-based Activation Steering for Inference-Time Intervention
This paper introduces FLAS, a flow-based activation steering method that learns a concept-conditioned velocity field to steer language model activations at inference time. On the AxBench benchmark, FLAS is the first learned method to consistently outperform in-context prompting on held-out concepts without per-concept tuning.
@svlevine: A new way to do off-policy RL with diffusion: if we have off-policy data, we need to figure out what the diffusion late…
A new method for off-policy reinforcement learning with diffusion models, using flow reversal to handle off-policy data by reversing the diffusion process on it.
@svlevine: Diffusion (or flow) makes for excellent policies, but training them with RL is notoriously hard: BPTT is unstable, RL o…
New paper shows how to optimize flow matching actors for reinforcement learning by approximating the Jacobian of the flow denoising process with the identity matrix, making training feasible.
@aditya_oberai: What if we treat flow steps as RL actions? Combined with our “flow reversal” technique, this leads to a really clean & …
Proposes treating flow steps as RL actions combined with a 'flow reversal' technique for flow offline reinforcement learning.
UniSteer: Text-Guided Flow Matching in Activation Space for Versatile LLM Steering
UniSteer introduces a text-guided activation flow matching method to learn a universal conditional velocity field in activation space, enabling versatile LLM behavior control and classification tasks without task-specific intervention modules.