AGI Lab

non-profit

AI & ML interests

None defined yet.

Recent Activity

Ningyu submitted a paper 5 days ago

How LoRA Remembers? A Parametric Memory Law for LLM Finetuning

Ningyu submitted a paper 6 days ago

Rethinking Memory as Continuously Evolving Connectivity

Ningyu submitted a paper 6 days ago

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems

View all activity

Papers

How LoRA Remembers? A Parametric Memory Law for LLM Finetuning

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems

View all Papers

authored a paper about 7 hours ago

Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents

Paper • 2605.29447 • Published 6 days ago • 17

authored 3 papers 4 days ago

D-CORE: Incentivizing Task Decomposition in Large Reasoning Models for Complex Tool Use

Paper • 2602.02160 • Published Feb 2 • 14

Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization

Paper • 2605.28109 • Published 7 days ago • 23

Pyramid Texture Filtering

Paper • 2305.06525 • Published May 11, 2023 • 1

submitted a paper to Daily Papers 5 days ago

How LoRA Remembers? A Parametric Memory Law for LLM Finetuning

Paper • 2605.30260 • Published 6 days ago • 37

submitted 2 papers to Daily Papers 6 days ago

Rethinking Memory as Continuously Evolving Connectivity

Paper • 2605.28773 • Published 7 days ago • 32

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems

Paper • 2605.28732 • Published 7 days ago • 39

submitted a paper to Daily Papers 6 days ago

Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization

Paper • 2605.28109 • Published 7 days ago • 23

submitted a paper to Daily Papers 9 days ago

Rethinking Cross-Layer Information Routing in Diffusion Transformers

Paper • 2605.20708 • Published 14 days ago • 109

authored a paper 12 days ago

Rethinking Cross-Layer Information Routing in Diffusion Transformers

Paper • 2605.20708 • Published 14 days ago • 109

xiaochonglinghu

submitted a paper to Daily Papers 12 days ago

Enhancing Train-Free Infinite-Frame Generation for Consistent Long Videos

Paper • 2605.18233 • Published 16 days ago • 91

authored a paper 13 days ago

Full Attention Strikes Back: Transferring Full Attention into Sparse within Hundred Training Steps

Paper • 2605.16928 • Published 18 days ago • 93

authored a paper 14 days ago

Inference-Scale Complexity in ANN-SNN Conversion for High-Performance and Low-Power Applications

Paper • 2409.03368 • Published Mar 5, 2025 • 1

authored a paper 15 days ago

BEAM: Binary Expert Activation Masking for Dynamic Routing in MoE

Paper • 2605.14438 • Published 20 days ago • 5

submitted a paper to Daily Papers 16 days ago

FashionChameleon: Towards Real-Time and Interactive Human-Garment Video Customization

Paper • 2605.15824 • Published 19 days ago • 64

submitted a paper to Daily Papers 18 days ago

BEAM: Binary Expert Activation Masking for Dynamic Routing in MoE

Paper • 2605.14438 • Published 20 days ago • 5

submitted a paper to Daily Papers 23 days ago

What Matters for Diffusion-Friendly Latent Manifold? Prior-Aligned Autoencoders for Latent Diffusion

Paper • 2605.07915 • Published 26 days ago • 8

submitted a paper to Daily Papers 26 days ago

Continuous-Time Distribution Matching for Few-Step Diffusion Distillation

Paper • 2605.06376 • Published 27 days ago • 26

submitted a paper to Daily Papers about 1 month ago

CoInteract: Physically-Consistent Human-Object Interaction Video Synthesis via Spatially-Structured Co-Generation

Paper • 2604.19636 • Published Apr 21 • 87

xiaochonglinghu

submitted a paper to Daily Papers about 1 month ago

Elucidating the SNR-t Bias of Diffusion Probabilistic Models

Paper • 2604.16044 • Published Apr 17 • 73