← Search

Dmitrii Tarasov

2 accepted papers

2026

Progressive Cramming: Reliable Token Compression and What It Reveals

ICML 2026poster

Token cramming compresses sequences into learned embeddings with near-perfect reconstruction, but prior work used fixed token budgets and 99\% accuracy thresholds, obscuring whether residual errors reflect optimization failures or fundamental limits. We introduce progressive cramming, which grows th…

Cited by 0SourceScholar
2026

Speech-to-LaTeX: New Models and Datasets for Converting Spoken Equations and Sentences

ICLR 2026poster

Conversion of spoken mathematical expressions is a challenging task that involves transcribing speech into a strictly structured symbolic representation while addressing the ambiguity inherent in the pronunciation of equations. Although significant progress has been achieved in automatic speech reco…

Cited by 0SourcecodeScholar