CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad

• URL: http://arxiv.org/abs/2603.14575v1
• Date: Sun, 15 Mar 2026 19:53:31 GMT
• Title: CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad
• Abstract summary: We develop CausalEvolve, equipped with a causal scratchpad that leverages Large Language Models (LLMs) to identify and reason about guiding factors for evolution.<n>CaulEvolve first identifies outcome-level factors that offer complementary inspirations in improving the target objective.<n>During the evolution, CausalEvolve also inspects surprise patterns during the evolution and abductive reasoning to hypothesize new factors, which in turn offer novel directions.
• Score: 72.197582948888
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Evolve-based agent such as AlphaEvolve is one of the notable successes in using Large Language Models (LLMs) to build AI Scientists. These agents tackle open-ended scientific problems by iteratively improving and evolving programs, leveraging the prior knowledge and reasoning capabilities of LLMs. Despite the success, existing evolve-based agents lack targeted guidance for evolution and effective mechanisms for organizing and utilizing knowledge acquired from past evolutionary experience. Consequently, they suffer from decreasing evolution efficiency and exhibit oscillatory behavior when approaching known performance boundaries. To mitigate the gap, we develop CausalEvolve, equipped with a causal scratchpad that leverages LLMs to identify and reason about guiding factors for evolution. At the beginning, CausalEvolve first identifies outcome-level factors that offer complementary inspirations in improving the target objective. During the evolution, CausalEvolve also inspects surprise patterns during the evolution and abductive reasoning to hypothesize new factors, which in turn offer novel directions. Through comprehensive experiments, we show that CausalEvolve effectively improves the evolutionary efficiency and discovers better solutions in 4 challenging open-ended scientific tasks.

Related papers

• MeEvo: Metacognitive Evolution Combined with Natural Evolution for Automatic Heuristic Design [17.812076497570157]
  We propose MeEvo, a dual-layer Automatic Heuristic Design framework that cyclically couples Natural Evolution and Metacognitive Evolution.<n>MeEvo achieves stronger and more stable performance than existing LLM-based AHD architectures, especially on complex constrained tasks.
  arXiv  Detail & Related papers  (2026-06-12T07:36:38Z)
• EvoSci: A Bio-Inspired Multi-Agent Framework for the Evolution of Scientific Discovery [51.25847252305662]
  EvoSci is a multi-agent scientific collaboration framework, which integrates bio-inspired evolution with knowledge graph modeling.<n>By combining collaborative reasoning, shared memory, and evolutionary feedback, EvoSci significantly enhances the coherence and creativity of scientific exploration.
  arXiv  Detail & Related papers  (2026-05-20T04:58:49Z)
• Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents [58.69865074060139]
  We study the case where an agent's self-evolution deviates in unintended ways, leading to undesirable or even harmful outcomes.<n>Our empirical findings reveal that misevolution is a widespread risk, affecting agents built even on top-tier LLMs.<n>We discuss potential mitigation strategies to inspire further research on building safer and more trustworthy self-evolving agents.
  arXiv  Detail & Related papers  (2025-09-30T14:55:55Z)
• Towards Understanding How Knowledge Evolves in Large Vision-Language Models [55.82918299608732]
  We investigate how multimodal knowledge evolves and eventually induces natural languages in Large Vision-Language Models (LVLMs)<n>We identify two key nodes in knowledge evolution: the critical layers and the mutation layers, dividing the evolution process into three stages: rapid evolution, stabilization, and mutation.<n>Our research is the first to reveal the trajectory of knowledge evolution in LVLMs, providing a fresh perspective for understanding their underlying mechanisms.
  arXiv  Detail & Related papers  (2025-03-31T17:35:37Z)
• Learning Evolution via Optimization Knowledge Adaptation [50.280704114978384]
  Evolutionary algorithms (EAs) maintain populations through evolutionary operators to discover solutions for complex tasks.<n>We introduce an Optimization Knowledge Adaptation Evolutionary Model (OKAEM) to enhance its optimization capabilities.<n>OKAEM exploits prior knowledge for significant performance gains across various knowledge transfer settings.<n>It is capable of emulating principles of natural selection and genetic recombination.
  arXiv  Detail & Related papers  (2025-01-04T05:35:21Z)
• EvoWiki: Evaluating LLMs on Evolving Knowledge [72.92365627254063]
  EvoWiki is an evolving dataset designed to reflect knowledge evolution by categorizing information into stable, evolved, and uncharted states.<n>Our results indicate that current models often struggle with evolved knowledge, frequently providing outdated or incorrect responses.<n>EvoWiki provides a robust benchmark for advancing future research on the knowledge evolution capabilities of large language models.
  arXiv  Detail & Related papers  (2024-12-18T08:04:57Z)
• Evolutionary Reinforcement Learning: A Systematic Review and Future Directions [18.631418642768132]
  Evolutionary Reinforcement Learning (EvoRL) is a solution to the limitations of reinforcement learning and evolutionary algorithms (EAs) in complex problem-solving. EvoRL integrates EAs and reinforcement learning, presenting a promising avenue for training intelligent agents. This systematic review provides insights into the current state of EvoRL and offers a guide for advancing its capabilities in the ever-evolving landscape of artificial intelligence.
  arXiv  Detail & Related papers  (2024-02-20T02:07:57Z)
• Learning and evolution: factors influencing an effective combination [0.0]
  The mutual relationship between evolution and learning is a controversial argument among the artificial intelligence and neuro-evolution communities. The author investigates whether combining learning and evolution permits to find better solutions than those discovered by evolution alone.
  arXiv  Detail & Related papers  (2023-06-20T09:03:52Z)
• The Role of Evolution in Machine Intelligence [8.228029233058411]
  I argue that the alternative, evolution, is important to the development of machine intelligence. My first-order suggestion is to diversify research across a broader spectrum of evolutionary approaches.
  arXiv  Detail & Related papers  (2021-06-21T14:46:18Z)
• Incremental Evolution and Development of Deep Artificial Neural Networks [1.6631602844999722]
  We extend Fast Deep Evolutionary Network Structured Representation (Fast-DENSER) to incremental development. Results show that the average performance of the models generated by incremental development is statistically superior to the non-incremental average performance.
  arXiv  Detail & Related papers  (2020-04-01T09:23:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.