Yu li's picture

193 198

Yu li

Yukkkop

·

AI & ML interests

None yet

Recent Activity

liked a Space about 16 hours ago

wcy1122/MGM-Omni

reacted to Jaward's post with 🚀 about 16 hours ago

fascinating read! staying bullish on search with rl might just help us get rid of hallucination entirely. I really like their approach: 1) <think>on prompt/context && what u know </think> 2) self <search>when u don’t know</search> (iteratively) with no external tool 3) <information>cite sources to support claim(s)</information> 4) <answer>final answer</answer> their rl training was done cost efficiently too, see code: https://github.com/TsinghuaC3I/SSRL

liked a model 1 day ago

ServiceNow-AI/Apriel-Nemotron-15b-Thinker

View all activity

Organizations

None yet

upvoted 4 papers 16 days ago

AlphaAdam:Asynchronous Masked Optimization with Dynamic Alpha for Selective Updates

Paper • 2501.18094 • Published Jan 30 • 1

Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam

Paper • 2502.17055 • Published Feb 24 • 19

PixNerd: Pixel Neural Field Diffusion

Paper • 2507.23268 • Published 20 days ago • 50

Beyond Fixed: Variable-Length Denoising for Diffusion Large Language Models

Paper • 2508.00819 • Published 19 days ago • 62

upvoted 7 papers 22 days ago

Every FLOP Counts: Scaling a 300B Mixture-of-Experts LING LLM without Premium GPUs

Paper • 2503.05139 • Published Mar 7 • 4

Latent Flow Transformer

Paper • 2505.14513 • Published May 20 • 29

Differentiable Solver Search for Fast Diffusion Sampling

Paper • 2505.21114 • Published May 27 • 12

Aligning Latent Spaces with Flow Priors

Paper • 2506.05240 • Published Jun 5 • 27

Emerging Properties in Unified Multimodal Pretraining

Paper • 2505.14683 • Published May 20 • 133

Beyond Context Limits: Subconscious Threads for Long-Horizon Reasoning

Paper • 2507.16784 • Published 29 days ago • 116

Singular Value Decomposition on Kronecker Adaptation for Large Language Model

Paper • 2506.15251 • Published Jun 18 • 1

upvoted a paper 23 days ago

SingLoRA: Low Rank Adaptation Using a Single Matrix

Paper • 2507.05566 • Published Jul 8 • 110

upvoted 5 papers about 1 month ago

RiemannLoRA: A Unified Riemannian Framework for Ambiguity-Free LoRA Optimization

Paper • 2507.12142 • Published Jul 16 • 36

TC-Light: Temporally Consistent Relighting for Dynamic Long Videos

Paper • 2506.18904 • Published Jun 23 • 10

Neural-Driven Image Editing

Paper • 2507.05397 • Published Jul 7 • 26

Zebra-Llama: Towards Extremely Efficient Hybrid Models

Paper • 2505.17272 • Published May 22 • 1

Energy-Based Transformers are Scalable Learners and Thinkers

Paper • 2507.02092 • Published Jul 2 • 60

upvoted 2 papers about 2 months ago

RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale

Paper • 2505.03005 • Published May 5 • 35

Trillion 7B Technical Report

Paper • 2504.15431 • Published Apr 21 • 38

upvoted a paper 2 months ago

OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generation

Paper • 2506.07977 • Published Jun 9 • 41