Mingzhe Du's picture

Mingzhe Du PRO

Elfsong

·

https://mingzhe.space

Elfsong

AI & ML interests

Code Generation / Preference Alignment / Bias Mitigation

Recent Activity

upvoted a paper 1 day ago

Reasoning with Confidence: Efficient Verification of LLM Reasoning Steps via Uncertainty Heads

upvoted a paper about 1 month ago

Towards Agentic RAG with Deep Reasoning: A Survey of RAG-Reasoning Systems in LLMs

upvoted a paper about 1 month ago

ExGRPO: Learning to Reason from Experience

View all activity

Organizations

Elfsong 's datasets 79

Elfsong/swe-perf-reasoning

Viewer • Updated Sep 5 • 292 • 72 • 1

Elfsong/swe-perf

Updated Aug 26 • 5

Elfsong/cisco_cr

Viewer • Updated Aug 25 • 104 • 3

Elfsong/Venus

Viewer • Updated Aug 18 • 9.3k • 98 • 6

Elfsong/new_cisco_cr_uc3

Viewer • Updated Aug 11 • 39 • 9

Elfsong/Venus_Anotation

Viewer • Updated Jul 28 • 12 • 42

Elfsong/cisco-cr-changes

Viewer • Updated Jul 23 • 52 • 5

Elfsong/cisco-cr

Viewer • Updated Jul 21 • 39 • 11

Elfsong/Venus_Python

Viewer • Updated Apr 16 • 1.28k • 45

Elfsong/Venus_Python_GRPO_Reasoning_Cold_Start

Viewer • Updated Apr 14 • 898 • 10

Elfsong/APPS_New

Updated Apr 3 • 24

Elfsong/apps_verified

Viewer • Updated Mar 21 • 3.37k • 4

Elfsong/APPS

Viewer • Updated Mar 20 • 12.3k • 66

Elfsong/Cisco_Response_03_14

Viewer • Updated Mar 19 • 550 • 32

Elfsong/JITT

Viewer • Updated Mar 16 • 2.39k • 5

Elfsong/Cisco_Prompts

Viewer • Updated Mar 14 • 55 • 5

Elfsong/Cisco_Response_New

Viewer • Updated Mar 13 • 440 • 3

Elfsong/Cisco_Response

Viewer • Updated Mar 12 • 385 • 6

Elfsong/DenseRuntime

Viewer • Updated Mar 11 • 768k • 4

Elfsong/QG-DPO-New

Viewer • Updated Mar 11 • 4.57k • 8

Elfsong/Llama8B-SFT-Questions

Viewer • Updated Mar 5 • 3.57k • 5

Elfsong/Llama8B-DPO-Questions

Viewer • Updated Mar 5 • 3.47k • 8

Elfsong/Llama8B-DPO-QG

Updated Mar 4 • 4

Elfsong/Llama8B-SFT-QG

Updated Mar 4 • 24

Elfsong/apps_generation

Updated Jan 31 • 595 • 1

Elfsong/Llama-3.3-70B-Instruct-sft-response

Viewer • Updated Jan 7 • 256 • 3

Elfsong/Llama-3.3-70B-Instruct-dpo-response

Viewer • Updated Jan 7 • 256 • 3

Elfsong/Llama-3.3-70B-Instruct-response

Viewer • Updated Jan 7 • 256 • 4

Elfsong/Llama-3.1-8B-Instruct-dpo-response

Viewer • Updated Jan 7 • 256 • 6

Elfsong/Llama-3.1-8B-Instruct-response

Viewer • Updated Jan 7 • 256 • 7