2 papers
cs.CL2026
Merge-Based Vocabulary Extension Stalls at the Pre-Tokenizer: An English-Preserving Repair for Nepali, Tested in Continued Pretraining
Aarjan Chaudhary
Llama-3, Qwen3 and many other open LLMs pre-tokenize text with a regular expression whose word class admits only Unicode letters. Devanagari vowel signs are combining marks, so thi…
cs.CR2026
Baselines Before Architecture: Evaluating Coding Agents for Autonomous Penetration Testing
Ananda Dhakal, Krish Neupane, Aarjan Chaudhary
Recent autonomous penetration testing papers report high benchmark scores while adding multi-component security harnesses around frontier LLMs. Because these systems often change b…