{"arxiv_id":"2305.17400","incoming":{"papers":4,"sentences":8,"groups":[{"kind":"method_or_tool","label":"Uses its method or tool","papers":1,"sentences":1,"quotes":[{"arxiv_id":"2506.14648","title":"SENIOR: Efficient Query Selection and Preference-Guided Exploration in Preference-based Reinforcement Learning","url":"/paper/2506.14648","section":"Implementation Details","sentence":"For all methods, we use unsupervised pre-training proposed by PEBBLE and the same hyper-parameters and network architectures as in original papers [ref] .","intent_p":1.0,"contests_p":null}]},{"kind":"background","label":"Cites it as background","papers":2,"sentences":3,"quotes":[{"arxiv_id":"2506.14648","title":"SENIOR: Efficient Query Selection and Preference-Guided Exploration in Preference-based Reinforcement Learning","url":"/paper/2506.14648","section":"INTRODUCTION","sentence":"To address this problem, various query selection schemes have been proposed, such as entropy-based query selection [ref] , disagreement-based query selection [ref] , policy-aligned query selection [ref] , etc.","intent_p":0.85,"contests_p":null},{"arxiv_id":"2507.00611","title":"Residual Reward Models: Leveraging Prior Knowledge for Efficient Preference-based Reinforcement Learning in Robotics","url":"/paper/2507.00611","section":"Reward Learning from Human Preferences","sentence":"Prior work has studied improving query-efficiency by sampling the most informative behaviors to obtain preferences, such as maximizing behavior entropy [ref] , aligning the agent's actions with the queries [ref] , and allowing the agent to actively ask questions [ref] .","intent_p":0.86,"contests_p":null}]}],"unlabelled_sentences":4},"outgoing":{"papers":5,"sentences":7,"groups":[{"kind":"background","label":"Cites as background","papers":2,"sentences":2,"quotes":[{"arxiv_id":"2109.10862","title":"Recursively Summarizing Books with Human Feedback","url":"/paper/2109.10862","section":"Related Work","sentence":"PbRL provides a natural approach for humans (oracle overseer) to communicate desired behaviors with RL agents by making relative judgments between a pair of behaviors [ref] .","intent_p":0.98,"contests_p":null},{"arxiv_id":"2004.13649","title":"Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels","url":"/paper/2004.13649","section":"Data Augmentation for Reward Learning","sentence":"Data augmentation has been widely used in many deep RL [ref] algorithms, and also has been seamlessly integrated into previous PbRL algorithms for consistency regularization [ref] .","intent_p":0.8,"contests_p":null}]}],"unlabelled_sentences":5},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2305.17400"}}