Merge-Based Vocabulary Extension Stalls at the Pre-Tokenizer: An English-Preserving Repair for Nepali, Tested in Continued Pretraining
arXiv 2610.33016v1•cb7a2e552f539d9c287e99856dabf492afb03d2efff087f4a229f1b713f5aeb7
Paper metadata
- arXiv ID
- 2610.33016
- Version
- v1
- Category
- cs.AI, cs.CL
- Authors
- Aarjan Chaudhary (Naamche Labs)
- Publication date
- 2026-10-27T04:00:00Z
- Source identifier
- 2610.33016v1
- Public record ID
- record:sha256:cb7a2e552f539d9c287e99856dabf492afb03d2efff087f4a229f1b713f5aeb7
The PDF link points to arxiv.org. Baitaphish does not expose a private stored PDF.
This is source-provided metadata, not an enriched summary or an impact assessment. Follow the canonical source link for the published material.
Evidence and limitations
- Source ID
- arxiv_research
- Record identifier
- cb7a2e552f539d9c287e99856dabf492afb03d2efff087f4a229f1b713f5aeb7
- Record type
- Source metadata
This record may overlap with other records. Source metadata can be incomplete or change. Validate consequential decisions against the linked source and your own environment.