The official repository of the paper: Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models
Dehai Min
ZhishanQ
AI & ML interests
PhD@UIC | NLP, LLM
Recent Activity
submitted a paper about 10 hours ago
Verifiable Rewards Beyond Math and Code: Lightweight Corpus-Grounded Process Supervision for Factual Question Answering updated a collection 4 days ago
PUMAOrganizations
models 6
ZhishanQ/qwen3-embedding-redundancy-detector-0.6B
Sentence Similarity • 0.6B • Updated • 51
ZhishanQ/QuCo-extractor-0.5B
Text Generation • 0.5B • Updated • 110 •
ZhishanQ/UniHGKR-7B
Feature Extraction • 7B • Updated • 1
ZhishanQ/UniHGKR-base-beir
Sentence Similarity • 0.1B • Updated • 4
ZhishanQ/UniHGKR-base
Sentence Similarity • 0.1B • Updated • 3
ZhishanQ/UniHGKR-7B-pretrained
Feature Extraction • 7B • Updated • 3
datasets 6
ZhishanQ/puma-rd-training-data
Viewer • Updated • 484k • 17
ZhishanQ/QuCo-RAG-es-data-archive
Updated • 11
ZhishanQ/UniHGKR_eval_data
Updated • 11
ZhishanQ/UniHGKR_training_data
Preview • Updated • 62
ZhishanQ/UniHGKR_Date_Text_Pairs
Viewer • Updated • 4.33M • 24 • 1
ZhishanQ/CompMix-IR
Updated • 46