pith. sign in

Tokenization and the Noiseless Channel

3 Pith papers cite this work. Polarity classification is still indexing.

3 Pith papers citing it

fields

cs.CL 3

years

2026 3

representative citing papers

Tokenisation via Convex Relaxations

cs.CL · 2026-05-21 · unverdicted · novelty 7.0

ConvexTok uses convex relaxation of tokenization to a linear program, improving intrinsic metrics, bits-per-byte, and some downstream tasks while certifying near-optimality within 1% at typical vocabulary sizes.

Faster Superword Tokenization

cs.CL · 2026-04-06 · accept · novelty 7.0

Frequency aggregation of supermerge candidates and a two-phase formulation make BoundlessBPE and SuperBPE training over 600x faster on 1GB data while preserving identical results, with open-source Python and Rust code.

citing papers explorer

Showing 3 of 3 citing papers.

  • Tokenisation via Convex Relaxations cs.CL · 2026-05-21 · unverdicted · none · ref 21

    ConvexTok uses convex relaxation of tokenization to a linear program, improving intrinsic metrics, bits-per-byte, and some downstream tasks while certifying near-optimality within 1% at typical vocabulary sizes.

  • Faster Superword Tokenization cs.CL · 2026-04-06 · accept · none · ref 15

    Frequency aggregation of supermerge candidates and a two-phase formulation make BoundlessBPE and SuperBPE training over 600x faster on 1GB data while preserving identical results, with open-source Python and Rust code.

  • Tokenization with Split Trees cs.CL · 2026-05-21 · unreviewed · ref 78