Research ideas

Every idea extracted from recent arXiv mathematics papers — verified and unverified. Click an idea to open its full card; badges show the empirical verdict.

2 ideas found

Solves: Inference speedup ML: Transformer Math: Optimization Clear all

△ Mechanism confirmed, baseline not beaten 2026

Lipschitz-Certified Cache Refresh

Attach a certificate to a cached transformer KV state or recurrent latent state and refresh it only while its predicted certificate remains inside a latency-contracted admissible region. The controller uses a bound on certificate drift to guarantee that the state will remain admissible throughout the next sampling, communication, and execution delay, reducing unnecessary recomputation while exposing a measurable refresh boundary.

Useful7/10

Difficulty5/10

Novelty8/10

Dynamics Stability Inference speedup Transformer

Paper: CIPS: Maximal Certified Persistence in Cyber-Physical Systems arXiv:2608.06626

Unverified 2026

Powered Ratio Pruning for Heads and Channels

Replace an ordinary group-L1 penalty on structured neural components with a powered ratio-of-norms penalty applied to their nonnegative importance magnitudes. The ratio encourages importance to concentrate on a small number of heads, channels, or experts while being less sensitive to arbitrary rescaling of the underlying weights. After training, components with small importance can be physically removed and the model can be fine-tuned.

Useful6/10

Difficulty4/10

Novelty6/10

Regularization Memory Inference speedup Generalization

Paper: Separable Nonnegative Matrix Factorization Using Powered Ratio-of-Norms Regularization arXiv:2608.28799