Classification Done Right for Vision-Language Pre-Training
2024-11-05 Classification Done Right for Vision-Language Pre-Training Download PDF
Video Instruction Tuning With Synthetic Data
Oct 3, 2024 · The development of video large multimodal models (LMMs) has been hindered by the difficulty of curating large amounts of high-quality raw data from the web. To address this, we …
Seed-Thinking-v1.5: Advancing Superb Reasoning Models with ...
Apr 10, 2025 · Seed-Thinking-v1.5 achieves 86.7 on AIME 2024, 55.0 on Codeforces and 77.3 on GPQA, demonstrating excellent reasoning abilities in STEM and coding. Beyond reasoning tasks, …
MaskBit: Embedding-free Image Generation via Bit Tokens
Dec 8, 2024 · ABSTRACT Masked transformer models for class-conditional image generation have become a compelling alternative to diffusion models. Typically comprising two stages - an initial …
KOR-Bench: Benchmarking Language Models on Knowledge …
2024-10-09 KOR-Bench: Benchmarking Language Models on Knowledge-Orthogonal Reasoning Tasks Download PDF
DSTC: Direct Preference Learning with Only Self-Generated Tests and ...
2024-11-20 DSTC: Direct Preference Learning with Only Self-Generated Tests and Code to Improve Code LMs Download PDF
How Far is Video Generation from World Model: A Physical Law ...
2024-11-04 How Far is Video Generation from World Model: A Physical Law Perspective Download PDF
DiG: Scalable and Efficient Diffusion Models with Gated Linear ...
2024-11-26 DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention Download Paper Github
Reward-Augmented Data Enhances Direct Preference Alignment of LLMs
2024-10-10 Reward-Augmented Data Enhances Direct Preference Alignment of LLMs Download PDF Github
Hyper-Connections - Publications - ByteDance Seed Team
Mar 18, 2025 · We present hyper-connections, a simple yet effective method that can serve as an alternative to residual connections. This approach specifically addresses common drawbacks …