@sherryyangML: Machine learning engineering (MLE) is the new agentic frontier. I'll be sharing our work on scaling RL for MLE agents a…
Summary
Two ICLR 2026 papers show how small RL-trained agents outperform frontier models on machine-learning engineering tasks and how MLE-Smith automatically scales MLE workloads.
Similar Articles
@ethantsliu: meta-rl induces agentic LLM exploration traditional RL trains LLM agents to use fixed policies that struggle with activ…
A new research paper reframes LLM agent training as a cross-episode Meta-RL problem, using critic-free policy gradients to enable in-context adaptation without gradient updates. The LAMER framework improves test-time performance by 11-19% over standard RL baselines on long-horizon tasks and generalizes better to unseen environments.
MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering
OpenAI introduces MLE-bench, a benchmark of 75 Kaggle ML competitions to evaluate AI agents on real-world ML engineering tasks. The best setup, o1-preview with AIDE scaffolding, achieves at least a Kaggle bronze medal in 16.9% of competitions.
@sheriyuo: Meta-RL really does feel like a very promising direction
A researcher highlights that meta-RL is a promising direction for training LLM agents, reframing agent training as a cross-episode meta-RL problem to enable active exploration and trial-and-error adaptation.
LFM2.5-2.6B: Deploy Agents Everywhere (8 minute read)
Liquid AI releases LFM2.5-2.6B, a compact agentic model designed to run entirely on-device, enabling free inference, low latency, and privacy. The post details its training pipeline including SFT, teacher specialization, distillation, and agentic RL.
@dair_ai: New research from NVIDIA. They just dropped a PyTorch-native training framework for agentic RL. (bookmark it) Paper sum…
NVIDIA released Molt, a PyTorch-native agentic RL framework designed for compactness and readability, with performance comparable to Megatron-based stacks. The framework is open-source and includes a paper.