arXiv AI
DART-ES: Difficulty-Aware Reweighting and Targeted Replay for Fine-Tuning LLMs with Evolution Strategies
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Jul 8
Evaluating Fine-Tuning and Metrics for Neural Decompilation of Dart AOT Binaries
arXiv:2607. 06125v1 Announce Type: cross Abstract: Neural decompilation is increasingly studied as a code-generation problem, yet its evaluation methodology remains underdeveloped for modern languages.
arXiv AI
Aug 14
$\varepsilon$-MemEvo: Adaptive Cross-Task Memory Transfer for LLM Program Evolution
arXiv:2608. 12522v1 Announce Type: new Abstract: LLM-based program evolution systems such as FunSearch and AlphaEvolve have shown strong ability to discover novel algorithms, but typically optimize each task in isolation, discarding search experience after completion.
arXiv Machine Learning
Aug 19
Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements
Agentic ESOpt proposes using evolution strategies (ES) instead of reinforcement learning to fine‑tune large language‑model agents for long‑horizon tasks. ES offers model scalability, flexibility, and better long‑horizon credit assignment, enabling full‑parameter optimization with minimal GPU memory. The framework samples parameter perturbations, evaluates agents with rewards, and updates online, achieving notable performance gains on WebArena‑Lite and in test‑time prompt‑parameter co‑evolution.
arXiv AI
Sep 10
ReST-RL: Reinforcing LLM Reasoning through Unified Self-Training and Value-Guided Search
ReST‑RL introduces a unified Reinforced Self‑Training (ReST) policy‑value framework that enhances large language model (LLM) reasoning by combining an optimized ReST‑style GRPO algorithm with a value‑guided search (VM‑MCTS). The ReST‑GRPO component reshapes trajectory distributions to increase reward variance and expose policies to more informative partial states, improving training efficiency. VM‑MCTS trains a Value Model from self‑collected Monte‑Carlo Tree Search targets and uses it during inference to provide precise process signals and verification scores, boosting reasoning accuracy across coding benchmarks and out‑of‑domain math and science tasks.
arXiv Machine Learning
Aug 26
Mitigating Exploration Bias in RL for Multi-Instruction Following
arXiv:2608.23830v1 Announce Type: cross Abstract: RL has emerged as a powerful paradigm for enhancing the instruction following capabilities of LLMs. While existing training recipes achieve substanti...
arXiv AI
Aug 11
DistillCache: KL-Guided Adaptive KV-Cache Eviction for Memory-Efficient LLM Inference
arXiv:2608. 08878v1 Announce Type: cross Abstract: Transformer-based large language models (LLMs) achieve strong performance across many tasks, but their Key-Value (KV) cache grows linearly with sequence length, creating a severe memory bottleneck for long-context inference.