technique

Conditional Generative Flow Networks

techniqueactiveprovisionalconditional-generative-flow-networks-a41fece9·1 events·first seen 10h ago

Aliases: Conditional Generative Flow Networks

Co-occurring entities

Feature-wise Linear Modulation Trajectory Balance FlowPipe

More like this (12)

Posterior Refinement: Fast Language Generation via Any-Order Flow Maps causal generative modeling generative models Latent Reasoning with Normalizing Flows diffusion-based generative models generative language modeling Flow Map Language Models Self-Flow Generative Adversarial Networks FlowEdit: Associative Memory for Lifelong Pronunciation Adaptation in Flow-Matching TTS Transformer Grammars generative AI

Recent events (1)

4arXiv · cs.LG·10h ago·source ↗

FlowPipe: LLM-conditioned Generative Flow Networks for automated data preparation pipeline construction

FlowPipe is a new framework that frames ML data preparation pipeline synthesis as conditional probabilistic flow generation over a directed acyclic graph, using Conditional Generative Flow Networks (C-GFlowNets) with a Trajectory Balance objective. LLM-derived semantic priors are injected into the policy via Feature-wise Linear Modulation (FiLM), and a failure-aware flow objective steers search away from invalid states. Evaluated on 74 real-world datasets across two benchmark suites, FlowPipe improves accuracy by 11.96% on average over SOTA baselines and achieves 12.5x faster training convergence. The work addresses long-standing limitations in automated data pipeline construction including weak credit assignment and inefficient exploration.

Agent and Tool Ecosystem Feature-wise Linear Modulation Trajectory Balance Conditional Generative Flow Networks +1 more