Multi-Turn On-Policy Distillation with Prefix Replay
Paper • 2607.04763 • Published • 13
AI & ML interests
None defined yet.
baohao
submitted a
paper to Daily Papers about 1 month ago
karimknaebel
authored a
paper 3 months ago
karimknaebel
submitted a
paper to Daily Papers 3 months ago
baohao
submitted a
paper to Daily Papers 7 months ago
karimknaebel
authored a
paper 7 months ago
baohao
authored 6
papers about 1 year ago
Make Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning
Paper • 2306.00477 • Published • 2
ApiQ: Finetuning of 2-Bit Quantized Large Language Model
Paper • 2402.05147 • Published
Mask More and Mask Later: Efficient Pre-training of Masked Language Models by Disentangling the [MASK] Token
Paper • 2211.04898 • Published
3-in-1: 2D Rotary Adaptation for Efficient Finetuning, Efficient Batching and Composability
Paper • 2409.00119 • Published • 1
Unilogit: Robust Machine Unlearning for LLMs Using Uniform-Target Self-Distillation
Paper • 2505.06027 • Published • 18
Lost at the Beginning of Reasoning
Paper • 2506.22058 • Published • 1
baohao
authored 2
papers over 1 year ago
karimknaebel
authored 3
papers almost 2 years ago
Benchmarks and Challenges in Pose Estimation for Egocentric Hand Interactions with Objects
Paper • 2403.16428 • Published
Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think
Paper • 2409.11355 • Published • 31
Point2Vec for Self-Supervised Representation Learning on Point Clouds
Paper • 2303.16570 • Published
baohao
authored a
paper about 2 years ago