AlphaMemo: Structured Search-Process Memory for Self-Evolving Alpha Mining Agents

• URL: http://arxiv.org/abs/2606.20625v1
• Date: Tue, 26 May 2026 15:48:09 GMT
• Title: AlphaMemo: Structured Search-Process Memory for Self-Evolving Alpha Mining Agents
• Abstract summary: AlphaMemo is a self-evolving alpha mining agent with Structured Search-Process Memory.<n>It records reusable evidence about which edit motifs work or fail under specific parent-factor contexts.<n>Experiments on CSI 500 and S&P 500 show improved out-of-sample performance and fixed-budget discovery efficiency.
• Score: 87.95001910627825
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: LLM agents are promising for alpha mining via combining financial priors, symbolic reasoning, executable factor generation, and feedback-driven refinement. Yet, they face a combinatorial search space, noisy non-stationary feedback, redundant discoveries, and overfitting risks from naively reusing past successes. To address these challenges, we propose AlphaMemo, a self-evolving alpha mining agent with Structured Search-Process Memory. Rather than memorizing only final factors or full trajectories, AlphaMemo records reusable evidence about which edit motifs work or fail under specific parent-factor contexts. It extracts motifs from Abstract Syntax Tree (AST) differences, applies confidence-gated residual memory on top of a search-ledger prior, and uses asymmetric veto control to suppress high-confidence failure patterns. Experiments on CSI 500 and S\&P 500 show improved out-of-sample performance and fixed-budget discovery efficiency, with ablations validating the roles of residual learning, confidence gating, AST-diff motifs, and veto memory. Code is at https://github.com/jarrettyu/AlphaMemo.

Related papers

• AutoScientist-Quant: Self-Evolving Coding Agents for Automatic Research in Quantitative Investment [28.36446889018906]
  We present AutoScientist-Quant, a self evolving search process that regards quantitative research as one budgeted search problem.<n>A single controller conditions every decision on the remaining budget, choosing at each round whether to improve, combine, pivot, or stop, which node to expand, how many alphas to generate, and how to retrieve past trajectories from the shared memory.<n>The framework attains the best value of nearly every metric in every setting, and these conclusions hold across several backbones and markets.
  arXiv  Detail & Related papers  (2026-08-05T23:34:46Z)
• RING: Retrieval-Internalized Generation for Continual Large-Scale Knowledge Injection [88.56816599695391]
  Retrieval-augmented generation (RAG) improves factuality but adds latency and engineering overhead at serving time.<n>We propose RING (Retrieval-Internalized Generation), a holistic paradigm spanning both architecture and training.
  arXiv  Detail & Related papers  (2026-08-03T03:00:43Z)
• AttriMem: Attribution-Guided Process Feedback for Agent Memory Learning [13.898789808338634]
  We propose an attribution-guided process-feedback framework for learning memory-construction policies with RL.<n>AttriMem augments the global outcome reward with local rewards derived from token-level contributions to the final answer.
  arXiv  Detail & Related papers  (2026-07-23T09:35:34Z)
• XALPHA: A Memory-Driven AI Quant Researcher for Hypothesis-to-Code Alpha Discovery [11.611093390158908]
  XAlpha is a memory-driven AI Quant Researcher for continuous hypothesis-to-code alpha discovery.<n>It integrates report-grounded financial knowledge with discovery feedback from prior generations and research cycles.<n>XAlpha turns alpha mining from isolated factor generation into a closed-loop research process.
  arXiv  Detail & Related papers  (2026-07-09T10:17:22Z)
• Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents [0.0]
  coding agents increasingly rely on external memory to reuse prior experience, repair traces, and repository-local operational knowledge.<n>This paper reframes issue-memory use as a selective, risk-sensitive control problem rather than a pure top-k retrieval problem.<n>We introduce RSCB-MC, a risk-sensitive contextual bandit memory controller that decides whether an agent should use no memory, inject the top resolution, summarize multiple candidates, perform high-precision or high-recall retrieval, abstain, or ask for feedback.
  arXiv  Detail & Related papers  (2026-04-30T00:32:53Z)
• Memory-Augmented LLM-based Multi-Agent System for Automated Feature Generation on Tabular Data [52.80567461420316]
  We propose a Memory-Augmented LLM-based Multi-Agent System (textbfMALMAS) for automated feature generation.<n>MALMAS decomposes the generation process into agents with distinct responsibilities, and a Router Agent activates an appropriate subset of agents per iteration.<n>We further integrate a memory module comprising procedural memory, feedback memory, and conceptual memory, enabling iterative refinement that adaptively guides subsequent feature generation.
  arXiv  Detail & Related papers  (2026-04-22T07:09:30Z)
• Latent Abstraction for Retrieval-Augmented Generation [4.6096940605642915]
  Existing systems rely on generating natural language queries at each hop.<n>We propose textbfLAnR (Latent Abstraction for RAG), a unified framework in which a single LLM jointly performs encoding, retrieval, and generation entirely within its own latent space.
  arXiv  Detail & Related papers  (2026-04-20T06:26:13Z)
• NextMem: Towards Latent Factual Memory for LLM-based Agents [58.35585202907478]
  NextMem is a latent factual memory framework that utilizes an autoregressive autoencoder to efficiently construct latent memory.<n>Extensive experiments demonstrate that NextMem achieves superior performance.
  arXiv  Detail & Related papers  (2026-02-26T14:35:27Z)
• Agent Alpha: Tree Search Unifying Generation, Exploration and Evaluation for Computer-Use Agents [6.357946414392768]
  Agent Alpha is a unified framework that synergizes generation, exploration, and evaluation through step-level Monte Carlo Tree Search (MCTS)<n>On the OSWorld benchmark, Agent Alpha achieves a state-of-the-art success rate of $sim 77%$, significantly outperforming trajectory-level baselines under equivalent compute.
  arXiv  Detail & Related papers  (2026-02-03T02:07:12Z)
• Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models [96.0074341403456]
  Inference-time compute has re-emerged as a practical way to improve LLM reasoning.<n>Most test-time scaling (TTS) algorithms rely on autoregressive decoding.<n>We propose Prism, an efficient TTS framework for dLLMs.
  arXiv  Detail & Related papers  (2026-02-02T09:14:51Z)
• Alpha-R1: Alpha Screening with LLM Reasoning via Reinforcement Learning [28.326583684637853]
  Signal decay and regime shifts pose recurring challenges for data-driven investment strategies in non-stationary markets.<n>Existing factor-based methods typically reduce alphas to numerical time series, overlooking the semantic rationale that determines when a factor is economically relevant.<n>We propose Alpha-R1, an 8B- parameter reasoning model trained via reinforcement learning for context-aware alpha screening.
  arXiv  Detail & Related papers  (2025-12-29T14:50:23Z)
• Cognitive Alpha Mining via LLM-Driven Code-Based Evolution [29.71597480304934]
  We introduce the Cognitive Alpha Mining Framework (CogAlpha), which combines code-level alpha representation with LLM-driven reasoning and evolutionary search.<n>Treating LLMs as adaptive cognitive agents, our framework iteratively refines, mutates, and recombines alpha candidates through prompts and financial feedback.<n>Experiments on A-share equities demonstrate that CogAlpha consistently discovers alphas with superior predictive accuracy, robustness, and generalization over existing methods.
  arXiv  Detail & Related papers  (2025-11-24T07:45:59Z)
• TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework [62.66056331998838]
  TeaRAG is a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps.<n>Our reward function evaluates the knowledge sufficiency by a knowledge matching mechanism, while penalizing excessive reasoning steps.
  arXiv  Detail & Related papers  (2025-11-07T16:08:34Z)
• Context Enhancement with Reconstruction as Sequence for Unified Unsupervised Anomaly Detection [68.74469657656822]
  Unsupervised anomaly detection (AD) aims to train robust detection models using only normal samples. Recent research focuses on a unified unsupervised AD setting in which only one model is trained for all classes. We introduce a novel Reconstruction as Sequence (RAS) method, which enhances the contextual correspondence during feature reconstruction.
  arXiv  Detail & Related papers  (2024-09-10T07:37:58Z)

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.