@HuggingPapers: Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance Naver AI eliminates unsta…
Summary
Naver AI introduces Stable-GFlowNet, a method to improve LLM red-teaming by eliminating unstable partition function estimation in Generative Flow Networks through contrastive trajectory balance.
View Cached Full Text
Cached at: 05/09/26, 08:16 PM
Stable-GFlowNet: Toward Diverse and Robust LLM Red-Teaming via Contrastive Trajectory Balance
Naver AI eliminates unstable partition function estimation in Generative Flow Networks via pairwise comparisons and robust masking, preventing mode collapse while maintaining diverse https://t.co/xRXREBVzmu
Similar Articles
$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data
This paper introduces a family of loss functions derived from f-divergences for training generative models like GFlowNets and LLMs, which are valid off-policy while matching on-policy gradients of the corresponding f-divergence. Applications include molecule discovery and asynchronous LLM tuning.
Generating Attacks for LLMs with GFlowNets
This paper proposes using GFlowNets to train an attacker LLM that automatically generates adversarial attacks against victim LLMs, enabling automated red teaming and robustness scoring in both English and Turkish.
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.
LeapAlign: Post-Training Flow Matching Models at Any Generation Step by Building Two-Step Trajectories
LeapAlign is a post-training method that improves flow matching model alignment with human preferences by reducing computational costs through two-step trajectory shortcuts while enabling stable gradient propagation to early generation steps. The method outperforms state-of-the-art approaches when fine-tuning Flux models across various image quality and text-alignment metrics.
@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.