Picture for Ang Li

Ang Li

EdgeLoRA: An Efficient Multi-Tenant LLM Serving System on Edge Devices

Add code
Jul 02, 2025
Viaarxiv icon

Convergence of Spectral Principal Paths: How Deep Networks Distill Linear Representations from Noisy Inputs

Add code
Jun 10, 2025
Viaarxiv icon

Towards provable probabilistic safety for scalable embodied AI systems

Add code
Jun 05, 2025
Viaarxiv icon

Avoid Forgetting by Preserving Global Knowledge Gradients in Federated Learning with Non-IID Data

Add code
May 26, 2025
Viaarxiv icon

Embracing Imperfection: Simulating Students with Diverse Cognitive Levels Using LLM-based Agents

Add code
May 26, 2025
Viaarxiv icon

Invisible Tokens, Visible Bills: The Urgent Need to Audit Hidden Operations in Opaque LLM Services

Add code
May 24, 2025
Viaarxiv icon

G1: Teaching LLMs to Reason on Graphs with Reinforcement Learning

Add code
May 24, 2025
Viaarxiv icon

Twin-2K-500: A dataset for building digital twins of over 2,000 people based on their answers to over 500 questions

Add code
May 23, 2025
Viaarxiv icon

AppealCase: A Dataset and Benchmark for Civil Case Appeal Scenarios

Add code
May 22, 2025
Viaarxiv icon

Identification of Probabilities of Causation: A Complete Characterization

Add code
May 21, 2025
Viaarxiv icon