Bryan Dai

ModularRSI is proposed, a benchmark-disjoint, contrastive, and modular framework for generalizable harness evolution that contrasts successful and failed trajectories for the same task and aggregates evidence across tasks to identify recurring behavioral deficiencies.

SpectralShift is proposed, a spectral reparameterization approach for long-context continual pretraining of GDNs that reparameterizes the alpha projections initialization to reshape the decay spectrum by enhancing slow propagation capacity, and further introduces a learning-rate scaling for alpha projections to facilitate long-context training.

OPD transfers reasoning skills rather than answers, revealing a complex interplay between teacher-student origins that can either enhance or hinder model capabilities.

Black-box RL can boost agent performance by nearly 15 points on complex tasks, revealing a new frontier for scalable optimization.

SWD achieves high-fidelity circuit extraction with less than 1% of the data used by traditional methods, revolutionizing interpretability in pretrained transformers.

BoE transforms candidate selection by leveraging partial verification, significantly enhancing outcomes in vision-language tasks where complete evaluations are unattainable.

Loopie outperforms traditional Transformers by leveraging a novel architecture that maximizes efficiency without sacrificing reasoning power, achieving gold-medal performance in competitive settings.

A two-loop configuration in LoopCoder-v2 boosts code generation performance by over 50% compared to a non-looped baseline, while more loops actually hinder results.

HyperTool boosts multi-step tool use accuracy by over 100% in LLMs, transforming how agents interact with complex tool workflows.

Building agents that can reliably automate complex, multi-step workflows over local files and tools just got a whole lot easier.

By dynamically injecting frequency-aware n-gram features, X-GRAM achieves state-of-the-art accuracy with smaller embedding tables, offering a practical path to scaling memory-augmented architectures.