Tianyi Lab @ UMD

Tianyi Lab @ UMD's repositories

Cherry_LLM

[NAACL'24] Self-data filtering of LLM instruction-tuning data using a novel perplexity-based difficulty score, without using any other models

Language:Python395 3 33

Reflection_Tuning

[ACL'24] Selective Reflection-Tuning: Student-Selected Data Recycling for LLM Instruction-Tuning

Language:Python364 4 6

HallusionBench

[CVPR'24] HallusionBench: You See What You Think? Or You Think What You See? An Image-Context Reasoning Benchmark Challenging for GPT-4V(ision), LLaVA-1.5, and Other Multi-modality Models

Language:PythonBSD-3-Clause310 5 13

Superfiltering

[ACL'24] Superfiltering: Weak-to-Strong Data Filtering for Fast Instruction-Tuning

Language:Python176 2 7

MoE-Embedding

Code for "Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free"

Language:Python83 2 5

MiP-Overthinking

[COLM'25] Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?

Language:PythonMIT3400

FaSTAR

Fast-Slow Toolpath Agent with Subroutine Mining for Efficient Multi-turn Image Editing

Language:Jupyter NotebookBSD-3-Clause2800

CoSTAR

Cost-Sensitive Toolpath Agent for Multi-turn Image Editing

Language:Jupyter NotebookBSD-3-Clause23 20

DEBATunE

[ACL'24] Can LLMs Speak For Diverse People? Tuning LLMs via Debate to Generate Controllable Controversial Statements

Language:Python23 10

ColorBench

Official repo for ColorBench

Language:PythonApache-2.02000

Mosaic-IT

[ACL'25] Mosaic-IT: Cost-Free Compositional Data Synthesis for Instruction Tuning

Language:Python20 10

C3PO

Code for "C3PO: Critical-Layer, Core-Expert, Collaborative Pathway Optimization for Test-Time Expert Re-Mixing"

Language:Jupyter NotebookApache-2.01800

R2-T2

[ICML 2025] Code for "R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts"

Language:PythonMIT1600

RuleR

[NAACL'25] RuleR: Improving LLM Controllability by Rule-based Data Recycling

Language:Python14 1 1

DisCL

Official repo for [ICCV 2025] Diffusion Curriculum (DisCL)

Language:Python12 20

BenTo

Code for "BENTO: benchmark reduction with in-context learning transferability"

Language:PythonApache-2.0400

mctune

[ACL'24] Multi-Objective Linguistic Control of Large Language Models

Language:Python2 10

MosT

Code for "Many-objective multi-solution transport"

Language:Python200