Manas Mehta
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
NLP×1
Frequent co-authors
Research Timeline
2026
Randomized YaRN Improves Length Generalization for Long-Context Reasoning
The paper proposes Randomized YaRN, a training method that improves length generalization in large language models by exposing them to out-of-distribution positional representations during training on short context data.
Highlighted terms show continued research focus across papers