method · first seen 4 Oct 2026
evolution strategies
Also: ES
Gradient-free optimization that perturbs model parameters and aggregates rewards, enabling fine-tuning using only forward passes.
Mentions per week
PapersCodeDiscussion
Earliest sources
- DART-ES: Difficulty-Aware Reweighting and Targeted Replay for Fine-Tuning LLMs with Evolution StrategiesPapers · 4 Oct 2026 · Zhishen Sun, Hongzhan Wang, Sizhe Dang, Guang Dai, Haishan Ye
- DP-ES: Differentially Private Evolution Strategies for Prompt OptimizationPapers · 5 Oct 2026 · Ziniu Liu, Aiping Li, Yue Han, Han Yu, Junjian Zhang, Dong Zhu, Changjian Li, Sh…
- Evolutionary One-Step Generators: Fast and Diverse Sampling for Discrete DesignPapers · 6 Oct 2026 · Marcus Vukojevic, Erik Nielsen, Veronica Lachi, Andrea Passerini, Giovanni Iacca
- Micro Neural Policies for Safe Real-Time Robotic ControlPapers · 6 Oct 2026 · Hongpeng Cao, Riccardo Curcio, Daniele Ottaviano, Marco Caccamo
- Continual Test-Time Learning with Evolution StrategiesPapers · 16 Oct 2026 · Guangya Hao, Yiting Duan, Longyan Tan, Yunbo Long, Hao Wu, Ge Zheng, Alexandra B…
- Adversarial Post-Training of Foundation Models for High-Frequency TradingPapers · 19 Oct 2026 · Alexandre Bismuth, Kang Li, Aramis Marti-Shahandeh, Valentin Mohl, Jakob Foerste…
Latest sources
- The Evolution Strategies Paradox: Better Models from Worse PerturbationsPapers · 3 Nov 2026 · Roberto Dailey, Kajetan Schweighofer, Jamieson Warner, Conor F. Hayes, Elliot Me…
- Center-of-Mass Interpretation of Evolution StrategiesPapers · 26 Oct 2026 · Akshat Gupta
- Emergent Cooperative Thermoregulation in Lunar Swarm RoboticsPapers · 22 Oct 2026 · Anais Jeger, Harry Holt, Lennart Puck
- Adversarial Post-Training of Foundation Models for High-Frequency TradingPapers · 19 Oct 2026 · Alexandre Bismuth, Kang Li, Aramis Marti-Shahandeh, Valentin Mohl, Jakob Foerste…
- Continual Test-Time Learning with Evolution StrategiesPapers · 16 Oct 2026 · Guangya Hao, Yiting Duan, Longyan Tan, Yunbo Long, Hao Wu, Ge Zheng, Alexandra B…
- Micro Neural Policies for Safe Real-Time Robotic ControlPapers · 6 Oct 2026 · Hongpeng Cao, Riccardo Curcio, Daniele Ottaviano, Marco Caccamo
- Evolutionary One-Step Generators: Fast and Diverse Sampling for Discrete DesignPapers · 6 Oct 2026 · Marcus Vukojevic, Erik Nielsen, Veronica Lachi, Andrea Passerini, Giovanni Iacca
- DP-ES: Differentially Private Evolution Strategies for Prompt OptimizationPapers · 5 Oct 2026 · Ziniu Liu, Aiping Li, Yue Han, Han Yu, Junjian Zhang, Dong Zhu, Changjian Li, Sh…
- DART-ES: Difficulty-Aware Reweighting and Targeted Replay for Fine-Tuning LLMs with Evolution StrategiesPapers · 4 Oct 2026 · Zhishen Sun, Hongzhan Wang, Sizhe Dang, Guang Dai, Haishan Ye