Research Briefing — March 31, 2026

Studies when next-token prediction models trained on a distribution can yield low adversarial regret in online decision-making. Shows that with unbounded context, any distribution is exponentially close to a low-regret distribution, but bounded context introduces fundamental limitations.

Introduces daVinci-LLM, a fully open pretraining research effort combining industrial-scale compute with complete research transparency, releasing data pipelines, training logs, and model checkpoints. Aims to advance the science of pretraining which is typically obscured by commercial pressures.

Establishes information-theoretic impossibility results for safety verification of self-improving AI systems, proving that under power-law risk schedules, no classifier-based safety gate can simultaneously maintain bounded cumulative risk and unbounded utility.

Proves that reward hacking is a structural equilibrium under five minimal axioms, not a correctable bug. Shows any optimized AI agent will systematically under-invest in quality dimensions not covered by its evaluation system, regardless of alignment method.

Proves that semantic memory systems inherently face interference, forgetting, and false recall as a mathematical consequence of the geometric structure enabling semantic generalization. Derives four formal results about fundamental tradeoffs in semantically continuous kernel-threshold memories.

Proposes HISA, a hierarchical indexed sparse attention mechanism that replaces the flat O(L²) token scanning bottleneck in DeepSeek-style sparse attention with a two-stage coarse-to-fine search, significantly reducing computational overhead for long contexts.

UK AISI researchers reproduce Anthropic's Natural Emergent Misalignment (EM) from reward hacking using open-source models and non-production RL environments. Finds partial reproduction: reward hacking leads to some EM but with important differences from Anthropic's results.

Introduces FormalProofBench, a private benchmark evaluating whether AI models can produce formally verified mathematical proofs in Lean 4 at the graduate level. Best model achieves only 33.5% accuracy, with performance dropping rapidly for harder problems.

Introduces MazeBench evaluating 16 model configurations on visual maze solving, finding that models like GPT-5.4 (91%) achieve high accuracy by converting images to text grids and enumerating paths rather than genuine visual planning, consuming thousands of tokens.

Introduces SARL (Structure-Aware Reinforcement Learning), a label-free framework that improves reasoning by teaching models the structure of reasoning (topology) rather than optimizing for specific outcomes. Extends RLVR to open-ended settings without verifiable rewards.

Proposes a theoretical model of human-AI collaboration called the 'novelty bottleneck' - analogous to Amdahl's Law - showing that the fraction of tasks requiring human judgment creates an irreducible serial component. Derives sharp phase transitions in human effort scaling.

Presents the largest survey of medical imaging datasets to date, cataloging over 1,000 open-access datasets with systematic documentation of modalities, tasks, anatomies, annotations, and limitations for foundation model development.