Almanac
benchmark

bits-per-byte (BpB)

benchmarkactivebits-per-byte-bpb--dbf6a839·2 events·first seen 26d ago

Aliases: bits-per-byte (BpB), bits-per-byte

Co-occurring entities

More like this (12)

Recent events (2)

5arXiv · cs.LG·26d ago·source ↗

ConvexTok: Tokeniser Construction via Linear Programming and Convex Optimisation

This paper proposes ConvexTok, a new tokenisation algorithm that formulates vocabulary construction as a linear program solved via convex optimisation, replacing the greedy approaches used by BPE and Unigram. ConvexTok consistently improves intrinsic tokenisation metrics and bits-per-byte (BpB) for language models, with less consistent gains on downstream tasks. A key feature is the ability to certify proximity to optimality via a lower bound, with empirical results showing the algorithm is within 1% of optimal at common vocabulary sizes.

6arXiv · cs.LG·19d ago·source ↗

HullFT: Efficient Test-Time Finetuning of LLMs via Convex Reconstruction and Gradient Caching

HullFT is a new method for test-time finetuning (TTFT) of language models that addresses the dual bottlenecks of retrieval quality and per-query finetuning cost. It represents query embeddings as sparse convex combinations of training sequences using Frank-Wolfe optimization, yielding diverse and relevant support sets without expensive diversity-aware search. A geometric integerization step converts fractional weights into integer multiplicities, enabling a Gradient Reuse scheme that amortizes forward-backward computation across repeated examples. Experiments show improved quality-efficiency tradeoffs over prior TTFT methods, measured in bits-per-byte at lower total runtime.