mlsanityData to 14 Oct 2026

Learning to Construct Knowledge through Sparse Reference Selection with Reinforcement Learning

arXiv 2509.05874 ↗First posted 7 Sep 2025cs.LG cs.AI cs.IRcited by 0

On the map: Deep reinforcement learning methods → RL training of language and agent models

Abstract (arXiv)

The rapid expansion of scientific literature makes it increasingly difficult to acquire new knowledge, particularly in specialized domains where reasoning is complex, full-text access is restricted, and target references are sparse among a large set of candidates. We present a Deep Reinforcement Learning framework for sparse reference selection that emulates human knowledge construction, prioritizing which papers to read under limited time and cost. Evaluated on drug--gene relation discovery with access restricted to titles and abstracts, our approach demonstrates that both humans and machines can construct knowledge effectively from partial information.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods. Builds on 6 of its 13 references that are in the map. First posted on arXiv 7 Sep 2025.

Builds on

6 of 13 references.

References · 6

  1. 2014Convolutional Neural Networks for Sentence Classificationcited by 1,127
  2. 2017DeepPath: A Reinforcement Learning Method for Knowledge Graph Reasoningcited by 266
  3. 2017Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learningcited by 161
  4. 2019Recovering Banditscited by 29
  5. 2019Blocking Banditscited by 25
  6. 2016ReasoNet: Learning to Stop Reading in Machine Comprehensioncited by 21

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Science Out of Its Ivory Tower: Improving Accessibility with Reinforcement Learning
  2. 2026Shaping Scientific Explanations to Expert Perspectives with Persona-Conditioned Reinforcement Learning
  3. 2026GIANTS: Generative Insight Anticipation from Scientific Literature
  4. 2026EvoIdeator: Evolving Scientific Ideas through Checklist-Grounded Reinforcement Learning
  5. 2026OUTLINEFORGE: Hierarchical Reinforcement Learning with Explicit States for Scientific Writing
  6. 2026Debate as Reward: A Multi-Agent Reward System for Scientific Ideation via RL Post-Training
  7. 2026COGTRL: Training LLMs for Scientific Discovery Assistance using Cognitive Traces via Reinforcement Learning
  8. 2025Beyond path selection: Better LLMs for Scientific Information Extraction with MimicSFT and Relevance and Rule-induced(R2)GRPO