zagreus-0.4B post-trained to 0.4787 on ITALIC (base 0.28, prior leader 0.372) via committee soft answer-KD, under 10 H100h.
Ruggero Marino Lazzaroni
ruggsea
AI & ML interests
NLP in any form
Recent Activity
updated a collection about 9 hours ago
ITALIC Post-Training Challenge submission updated a collection about 9 hours ago
ITALIC Post-Training Challenge submission updated a collection about 9 hours ago
ITALIC Post-Training Challenge submissionOrganizations
models 18
ruggsea/dante-zero-2025-03-05
Text Generation • 0.4B • Updated • 39
ruggsea/reappraisal-bt-reward-model
Updated
ruggsea/qwen3-4b-reappraisal-grpo-v4
Updated
ruggsea/reappraisal-reward-model
2B • Updated • 3
ruggsea/roman-mosaics-lora
Text-to-Image • Updated • 10 • • 1
ruggsea/qwen-2.5-7b-wsdm-cot-continued
Updated • 2
ruggsea/Llama70B-CoT-WSDM-adapter
Updated • 3
ruggsea/Qwen72B-CoT-WSDM-adapter
Updated • 2
ruggsea/Qwen72B-CoT-WSDM
Text Generation • 73B • Updated • 11
ruggsea/Llama70B-CoT-WSDM
Text Generation • 71B • Updated • 16 •
datasets 12
ruggsea/infini-news-index
Updated • 7.4k • 1
ruggsea/infini-news-corpus
Viewer • Updated • 1.36B • 48.4k • 30
ruggsea/social-sim-bench-gens
Viewer • Updated • 1.2k • 1.26k
ruggsea/stanford-encyclopedia-of-philosophy_instruct
Viewer • Updated • 11.9k • 81 • 15
ruggsea/continual-eval
Viewer • Updated • 1.63k • 66
ruggsea/bluesky-journalist-classification
Viewer • Updated • 1.19k • 108 • 1
ruggsea/wsdm2024-deepseek-cot
Viewer • Updated • 55.8k • 51 • 1
ruggsea/wsdm2024-cot-dataset
Viewer • Updated • 48.1k • 30 • 1
ruggsea/stanford-encyclopedia-of-philosophy_chat_multi_turn_mistral_large
Viewer • Updated • 11.9k • 16 • 5
ruggsea/stanford-encyclopedia-of-philosophy_chat_multi_turn_athene
Viewer • Updated • 11.9k • 16 • 2