L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
L3 Lab
university
AI & ML interests
None defined yet.
Recent Activity
models 12
l3lab/VTR-2B
2B • Updated • 133
l3lab/L1-Qwen3-8B-Max
8B • Updated • 36
l3lab/L1-Qwen3-8B-Exact
8B • Updated • 41 • 1
l3lab/L1-Qwen-7B-Max
8B • Updated • 36
l3lab/L1-Qwen-7B-Exact
8B • Updated • 145 • 1
l3lab/L1-1.5B-Short
2B • Updated • 12
l3lab/all-distilroberta-v1-lr2e-4-bs256-nneg3-ml-ne2
Updated • 173
l3lab/L1-Qwen-1.5B-Exact
2B • Updated • 189 • 6
l3lab/L1-Qwen-1.5B-Max
2B • Updated • 189 • 15
l3lab/ntp-mathlib-context-deepseek-coder-1.3b
Text Generation • Updated • 191 • 3
datasets 10
l3lab/far-explorer-2026-09-30
Viewer • Updated • 103k • 55
l3lab/miniCTX-v2
Viewer • Updated • 668 • 347 • 3
l3lab/miniCTX-v2-data
Updated • 23
l3lab/Massive-Math-455K-Verified
Viewer • Updated • 455k • 97 • 1
l3lab/lean-premises
Updated • 233 • 3
l3lab/miniCTX
Viewer • Updated • 662 • 1.11k • 3
l3lab/ntp-mathlib-instruct-context-fullproof
Viewer • Updated • 144k • 42 • 2
l3lab/ntp-mathlib-instruct-context
Viewer • Updated • 614k • 169 • 2
l3lab/ntp-mathlib
Viewer • Updated • 213k • 219 • 2
l3lab/ntp-mathlib-instruct-st
Viewer • Updated • 307k • 140