Picture for Tianyu Shi

Tianyu Shi

Department of Computer Science, University of Toronto

Enhancing Low-Cost Video Editing with Lightweight Adaptors and Temporal-Aware Inversion

Add code
Jan 08, 2025
Viaarxiv icon

Enhancing Code LLMs with Reinforcement Learning in Code Generation

Add code
Dec 29, 2024
Viaarxiv icon

SCBench: A Sports Commentary Benchmark for Video LLMs

Add code
Dec 23, 2024
Viaarxiv icon

FASIONAD : FAst and Slow FusION Thinking Systems for Human-Like Autonomous Driving with Adaptive Feedback

Add code
Nov 27, 2024
Figure 1 for FASIONAD : FAst and Slow FusION Thinking Systems for Human-Like Autonomous Driving with Adaptive Feedback
Figure 2 for FASIONAD : FAst and Slow FusION Thinking Systems for Human-Like Autonomous Driving with Adaptive Feedback
Figure 3 for FASIONAD : FAst and Slow FusION Thinking Systems for Human-Like Autonomous Driving with Adaptive Feedback
Figure 4 for FASIONAD : FAst and Slow FusION Thinking Systems for Human-Like Autonomous Driving with Adaptive Feedback
Viaarxiv icon

Infant Agent: A Tool-Integrated, Logic-Driven Agent with Cost-Effective API Usage

Add code
Nov 02, 2024
Viaarxiv icon

FALCON: Feedback-driven Adaptive Long/short-term memory reinforced Coding Optimization system

Add code
Oct 28, 2024
Viaarxiv icon

Traj-Explainer: An Explainable and Robust Multi-modal Trajectory Prediction Approach

Add code
Oct 22, 2024
Figure 1 for Traj-Explainer: An Explainable and Robust Multi-modal Trajectory Prediction Approach
Figure 2 for Traj-Explainer: An Explainable and Robust Multi-modal Trajectory Prediction Approach
Figure 3 for Traj-Explainer: An Explainable and Robust Multi-modal Trajectory Prediction Approach
Figure 4 for Traj-Explainer: An Explainable and Robust Multi-modal Trajectory Prediction Approach
Viaarxiv icon

Role-RL: Online Long-Context Processing with Role Reinforcement Learning for Distinct LLMs in Their Optimal Roles

Add code
Sep 26, 2024
Figure 1 for Role-RL: Online Long-Context Processing with Role Reinforcement Learning for Distinct LLMs in Their Optimal Roles
Figure 2 for Role-RL: Online Long-Context Processing with Role Reinforcement Learning for Distinct LLMs in Their Optimal Roles
Figure 3 for Role-RL: Online Long-Context Processing with Role Reinforcement Learning for Distinct LLMs in Their Optimal Roles
Figure 4 for Role-RL: Online Long-Context Processing with Role Reinforcement Learning for Distinct LLMs in Their Optimal Roles
Viaarxiv icon

LanguaShrink: Reducing Token Overhead with Psycholinguistics

Add code
Sep 01, 2024
Viaarxiv icon

Self-evolving Agents with reflective and memory-augmented abilities

Add code
Sep 01, 2024
Viaarxiv icon