← Back to all news
arXiv Computation and Language

Lineage Is Memory: Parent Inheritance and Feedback Disclosure in Short-Budget LLM-Guided Optimization

The paper investigates how iterative language‑model optimizers improve prompts and programs by evaluating candidates, exposing failing cases to a proposer model, and retaining revisions that enhance performance. By conducting 1,750 pre‑registered experiments across nine benchmarks, the authors disentangle the effects of parent inheritance (keeping a surviving parent) from feedback disclosure (providing failure summaries) under equal computational budgets. They find that greedy parent inheritance alone drives most optimization gains, outperforming seed‑reset strategies even when the latter receive comprehensive failure summaries or larger proposer models, and that the cost of proposer reflection is minimal compared to validation and minibatch gating.

Related stories

arXiv AI
Aug 11

Janus: An Algorithm-Evaluator Co-Evolution Framework for LLM-Driven Discovery under Expensive Evaluation Budgets

arXiv:2608. 08189v1 Announce Type: new Abstract: LLM-driven program discovery relies on rapid evaluator feedback, but many scientific and engineering tasks require high-fidelity simulations, hardware execution, or physical experiments, making each evaluation expensive.

By Ximeng Liu, Qianlong Wang, Yingming Mao, Annan Li, Yatao Li, Shizhen Zhao, Jianmin Wu, Dawei Yin, Dou Shen
llms
More like this →
arXiv Computation and Language
Oct 21

Explore More, Validate Less: Efficient Reflective Prompt Optimization

The paper introduces a two‑stage framework for reflective prompt optimization that separates candidate generation from validation. In the construction stage, multiple parents are sampled and their training responses reused across parallel reflection branches, with each revision screened on a small minibatch to allow locally improved candidates to propagate without full validation. In the exploration stage, a fixed full‑validation budget is used to select the best prompt from a frozen candidate pool, enabling independent control of pool size and validation effort. Experiments on five benchmarks with Qwen3.5‑9B and Llama‑3.1‑8B‑Instruct show that retaining 100 candidates and performing 20 full validations can improve performance by up to 11.22 percentage points over the initial solver, and learned selection consistently outperforms the initial solver across all comparisons.

By Xiaofan Zhou, Lu Cheng
llmsbenchmarks
More like this →
arXiv Machine Learning
Oct 22

HGM-Dual: Lineage-Aware Guided Agent Harness Optimization for Long-Horizon Planning

HGM-Dual is a diagnostic‑guided framework that improves harness optimization for planning agents by representing agent evolution as lineage‑aware context and using error‑conditioned dual‑path child generation. It records ancestor changes, behavior outcomes, and ineffective modifications, enabling explicit edit‑outcome attribution. The method achieves up to 14.45% relative improvement in case accuracy while using 2×–4× fewer evaluation budgets compared to the strongest baseline.

By Sudipta Paul, Pengrui Quan, Aounon Kumar, Nicholas Tzou, Vivek Kulkarni, Srinivas Chappidi
llmsagentssafety
More like this →
arXiv Machine Learning
Aug 3

Overcoming the Weakest-Link Effect in LLM-Driven Program Optimization via Heterogeneous Edit Recombination

arXiv:2607. 28947v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to solve complex problems by searching over program space, offering a general paradigm for scientific problems that can be naturally represented and solved as programs.

By Jingwen Fu, Zhen Liu, Yuhan Liu, He Zhang, Nanning Zheng
llmsagentsrobotics
More like this →
arXiv AI
Oct 9

How Should a Prompt Optimizer Spend a Tight Budget? BudgetAPO with Noise-Adaptive Evaluation

The paper introduces BudgetAPO, a single‑stage prompt optimizer designed for tight‑budget scenarios where only a few hundred model calls are feasible. BudgetAPO adapts the evaluation slice size to each task’s noise level, uses a fixed slice to turn accept/reject decisions into paired comparisons, and applies a reflective operator to jointly rewrite reasoning strategy and output format. Across seven benchmarks and five subject models, BudgetAPO outperforms existing methods, ranking first on every subject and requiring far fewer calls to achieve comparable or superior performance.

By Haoyue Liu, Zhichao Wang, Huanyu Yan, Xiaoying Tang
llmsbenchmarks
More like this →