Trending Research
Embedding Atlas: Low-Friction, Interactive Embedding Visualization
Embedding projections are popular for visualizing large datasets and models.
Practical Efficiency of Muon for Pretraining
We demonstrate that Muon, the simplest instantiation of a second-order optimizer, explicitly expands the Pareto frontier over AdamW on the compute-time tradeoff.
XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation
bytedance/xverse
•
•
Achieving fine-grained control over subject identity and semantic attributes (pose, style, lighting) in text-to-image generation, particularly for multiple subjects, often undermines the editability and coherence of Diffusion Transformers (DiTs).
R-KV: Redundancy-aware KV Cache Compression for Training-Free Reasoning Models Acceleration
To address this, we propose Redundancy-aware KV Cache Compression for Reasoning models (R-KV), a novel method specifically targeting redundant tokens in reasoning models.
SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Models
xid32/soundmind
•
•
While large language models have shown reasoning capabilities, their application to the audio modality, particularly in large audio-language models (ALMs), remains significantly underdeveloped.
MinerU: An Open-Source Solution for Precise Document Content Extraction
Document content analysis has been a crucial research area in computer vision.
TCSinger 2: Customizable Multilingual Zero-shot Singing Voice Synthesis
To overcome these challenges, we introduce TCSinger 2, a multi-task multilingual zero-shot SVS model with style transfer and style control based on various prompts.
Energy-Based Transformers are Scalable Learners and Thinkers
alexiglad/EBT
•
•
Further, we find that EBTs achieve better results than existing models on most downstream tasks given the same or worse pretraining performance, suggesting that EBTs generalize better than existing approaches.
Cautious Optimizers: Improving Training with One Line of Code
AdamW has been the default optimizer for transformer pretraining.
$τ^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
Existing benchmarks for conversational AI agents simulate single-control environments, where only the AI agent can use tools to interact with the world, while the user remains a passive information provider.
