Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements

AI By Crimson AI Hugging Face Papers 19 August 2026 · 00:00 7 views
Share: X Telegram

A new paper from Hugging Face introduces Agentic ESOpt, a framework using evolution strategies for full-parameter fine-tuning of long-horizon LLM agents, requiring only inference-level GPU memory and improving performance on WebArena-Lite.

Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements

Key points

Reinforcement learning (RL) has shown promise in single-turn LLM fine-tuning, but long-horizon agentic reasoning introduces branching interactions and sparse rewards, making RL less effective. The heavy backpropagation-based training stack of RL makes it impractical for larger models, and credit assignment becomes harder with longer trajectories.

In a new paper, researchers from Hugging Face argue that evolution strategies (ES) offer a better alternative for fine-tuning long-horizon agents. ES provides three key advantages: model scalability (full-parameter optimization with minimal inference-level GPU memory), flexibility (easy composition with prompt-space evolution), and long-horizon scalability (trajectory-level parameter attribution without decomposing rewards).

Based on this insight, they propose Agentic ESOpt, a full-parameter agentic fine-tuning framework that supports parameter-context co-evolution. At each step, it samples perturbations around the current parameters, evaluates the resulting agents with rewards, and applies an online reward-weighted update. A cosine decay schedule for the perturbation scale σ improves the exploration-adaptation trade-off.

On WebArena-Lite, full-parameter optimization of Qwen-3.5-27B improved the No Skill baseline by 6.69%. In test-time automatic heuristic design, Agentic ESOpt performed online prompt-parameter co-evolution, improving its matched baseline in 28 of 36 settings.

MetricValue
Improvement over No Skill baseline (WebArena-Lite)6.69%
Settings improved (test-time heuristic design)28 of 36
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 29 Aug 2026

Hugging Face Audit: 110 of 124 AI Evaluations Fail to Support Their Claims

A new commit-bound census of 124 Inspect Evals units reveals that 110 stop before deterministic inference due to missing historica...

4
Research paper
Hugging Face 29 Aug 2026

Aphanta: New Framework Diagnoses When Image Editing Boosts Multimodal Reasoning

Hugging Face researchers introduce Aphanta, a diagnostic framework that evaluates when image-editing intermediates improve multimo...

5
Research paper
Hugging Face 29 Aug 2026

EditaLive! Enables Real-Time Character Video Editing for Live Streaming

Hugging Face researchers introduce EditaLive, a framework for real-time human-centric video editing in live streams, achieving sta...

4