Google DeepMind proposes Decoupled DiLoCo, an extension of the DiLoCo distributed training framework designed for resilient training across heterogeneous or unreliable compute. No content snippet available beyond the title, but DiLoCo variants address the core challenge of large-scale training without tight synchronization.
MEM1 trains agents end-to-end via RL to compress and update an internal memory state at each step, maintaining constant context size across arbitrarily long multi-turn tasks. Unlike RAG or full-context retention, the memory management policy itself is learned. Demonstrated on multi-turn web and tool-use tasks; from MIT, accepted ICLR 2026.
Intuitor (ICLR 2026) trains LLMs to improve reasoning using only self-certainty as a reward signal—no labeled data, no external verifier, no human-crafted reward. The companion code release (RLIF framework) enables direct reproduction of the result that models can self-improve on reasoning benchmarks from internal feedback alone. Practically significant because it removes the dependency on curated verifiable datasets.