collaborators

6 papers

cs.LG2026

K2-V2: A 360-Open, Reasoning-Enhanced LLM

K2 Team, Zhengzhong Liu, Liping Tang +36

We introduce K2-V2, a 360-open LLM built from scratch as a superior base for reasoning adaptation, in addition to functions such as conversation and knowledge retrieval from genera…

cs.LG2025

PTPP-Aware Adaptation Scaling Laws: Predicting Domain-Adaptation Performance at Unseen Pre-Training Budgets

Etienne Goffinet, Shane Bergsma, Avraham Sheinin +4

Continual pre-training (CPT) for domain adaptation must balance target-domain gains with stability on the base domain. Existing CPT scaling laws typically assume a fixed pre-traini…

cs.LG2025

Power Lines: Scaling Laws for Weight Decay and Batch Size in LLM Pre-training

Shane Bergsma, Nolan Dey, Gurpreet Gosal +3

Efficient LLM pre-training requires well-tuned hyperparameters (HPs), including learning rate and weight decay . We study scaling laws for HPs: formulas for how to scale HPs…

cs.CL2025

Llama-3-Nanda-10B-Chat: An Open Generative Large Language Model for Hindi

Monojit Choudhury, Shivam Chauhan, Rocktim Jyoti Das +27

Developing high-quality large language models (LLMs) for moderately resourced languages presents unique challenges in data availability, model adaptation, and evaluation. We introd…

cs.CL2025

Sherkala-Chat: Building a State-of-the-Art LLM for Kazakh in a Moderately Resourced Setting

Fajri Koto, Rituraj Joshi, Nurdaulet Mukhituly +31

Llama-3.1-Sherkala-8B-Chat, or Sherkala-Chat (8B) for short, is a state-of-the-art instruction-tuned open generative large language model (LLM) designed for Kazakh. Sherkala-Chat (…

cs.LG2025

Straight to Zero: Why Linearly Decaying the Learning Rate to Zero Works Best for LLMs

Shane Bergsma, Nolan Dey, Gurpreet Gosal +3

LLMs are commonly trained with a learning rate (LR) warmup, followed by cosine decay to 10% of the maximum (10x decay). In a large-scale empirical study, we show that under an opti…