Decision Transformer
PulseAugur coverage of Decision Transformer — every cluster mentioning Decision Transformer across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New framework TAF improves decision-sequence learning with misaligned data
Researchers have developed a new framework called Target-Aligned Fusion (TAF) to improve decision-sequence learning when using external data that may not perfectly align with the target environment. TAF addresses dynami…
-
New HOBA framework enhances online advertising bidding with hierarchical RL
Researchers have developed HOBA, a novel hierarchical reinforcement learning framework designed to improve online advertising bidding systems. This system decouples strategic reasoning, model selection, and bid executio…
-
New SOV-CAD framework reconstructs CAD sequences using stepwise visual feedback
Researchers have developed SOV-CAD, a new framework for reconstructing Computer-Aided Design (CAD) modeling sequences from images. Unlike previous methods that generate entire sequences at once, SOV-CAD mimics human des…
-
New GLAN framework enhances personalized landing page recommendations
Researchers have developed GLAN, a novel sequence modeling framework designed to improve personalized landing page recommendations on online platforms. GLAN addresses limitations in previous reinforcement learning appro…
-
OnDeFog enhances reinforcement learning for frame-dropping environments
Researchers have introduced OnDeFog, an advancement in reinforcement learning designed to handle frame dropping, a common issue in real-world applications due to communication delays or sensor failures. This new method …
-
SlimDT paper proposes injecting RTG outside sequential modeling
Researchers have developed SlimDT, a modification of the Decision Transformer (DT) model for offline reinforcement learning. SlimDT removes the Return-to-Go (RTG) token from the autoregressive sequence, instead injectin…
-
QHyer model enhances offline goal-conditioned RL with adaptive history compression
Researchers have developed QHyer, a novel approach for offline goal-conditioned reinforcement learning that addresses challenges posed by partially observable and history-dependent datasets. QHyer utilizes a Q-estimator…
-
Gemma 4 31B weights show cross-modal transfer via thin trainable interface
Researchers have demonstrated that frozen weights from the Gemma 4 31B text-pretrained model can be effectively reused across different modalities, including robotics and associative recall tasks. By employing a thin, t…