mlsanityData to 14 Oct 2026

CosmoCore-Evo: Evolutionary Dream-Replay Reinforcement Learning for Adaptive Code Generation

arXiv 2512.21351 ↗First posted 20 Dec 2025cs.SE cs.AI cs.NEcited by 0

Not on the map. Why?

Abstract (arXiv)

Building on the affective dream-replay reinforcement learning framework of CosmoCore, we introduce CosmoCore-Evo, an extension that incorporates evolutionary algorithms to enhance adaptability and novelty in code generation tasks. Inspired by anthropological aspects of human evolution, such as natural selection and adaptation in early hominids, CosmoCore-Evo treats RL trajectories as ``genomes'' that undergo mutation and selection during the nocturnal replay phase. This mechanism allows agents to break free from trained patterns, fostering emergent behaviors and improved performance in distribution-shifted environments, such as changing APIs or novel libraries. We augment the Dream Queue with evolutionary operations, including mutation of high-fitness trajectories and enterprise-tuned fitness functions that incorporate efficiency, compliance, and scalability metrics. Evaluated on extended benchmarks including HumanEval variants with shifts, BigCodeBench, and a custom PySpark pipeline simulation, CosmoCore-Evo achieves up to 35% higher novelty in solutions and 25% faster adaptation compared to the original CosmoCore and baselines like PPO and REAMER. Ablations confirm the role of evolutionary components in bridging the sentient gap for LLM agents. Code for replication, including a toy simulation, is provided.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 28 references that are in the map. First posted on arXiv 20 Dec 2025.

Builds on

20 of 28 references.

References · 20

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  3. 2017Deep reinforcement learning from human preferencescited by 3,912
  4. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  5. 2015Prioritized Experience Replaycited by 1,099
  6. 2021CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generationcited by 906
  7. 2017Evolution Strategies as a Scalable Alternative to Reinforcement Learningcited by 574
  8. 2024BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructionscited by 546
  9. 2022CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learningcited by 358
  10. 2018Distributed Prioritized Experience Replaycited by 269
  11. 2017Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learningcited by 150
  12. 2018Evolution-Guided Policy Gradient in Reinforcement Learningcited by 83
8 more
  1. 2024StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedbackcited by 77
  2. 2023Evolutionary Reinforcement Learning: A Surveycited by 29
  3. 2025Process-Supervised Reinforcement Learning for Code Generationcited by 22
  4. 2017Emotion in Reinforcement Learning Agents and Robots: A Surveycited by 21
  5. 2023Reinforcement Learning-assisted Evolutionary Algorithm: A Survey and Research Opportunitiescited by 8
  6. 2022Play with Emotion: Affect-Driven Reinforcement Learningcited by 6
  7. 2025Towards Emotionally Intelligent and Responsible Reinforcement Learningcited by 2
  8. 2025CosmoCore Affective Dream-Replay Reinforcement Learning for Code Generationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Teaching LLMs to Self-Evolve: Cultivating Core Meta-Skills with Reinforcement Learning
  2. 2026What Do Evolutionary Coding Agents Evolve?
  3. 2026Code Evolution for Control: Synthesizing Policies via LLM-Driven Evolutionary Search
  4. 2025Co-Evolving LLM Coder and Unit Tester via Reinforcement Learning
  5. 2026Mem2Evolve: Towards Self-Evolving Agents via Co-Evolutionary Capability Expansion and Experience Distillation
  6. 2025ReVeal: Self-Evolving Code Agents via Reliable Self-Verification
  7. 2026Live-Evo: Online Evolution of Agentic Memory from Continuous Feedback
  8. 2026EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning