User contributions for M54rahma
Jump to navigation
Jump to search
6 April 2025
- 20:2720:27, 6 April 2025 diff hist +114 stat946W25 →System Overview and Architecture
- 20:2620:26, 6 April 2025 diff hist +36 N File:pipeline.png WebGLM system pipeline current
- 20:2420:24, 6 April 2025 diff hist +1,964 stat946W25 →System Overview and Architecture
- 19:5119:51, 6 April 2025 diff hist +2,797 stat946W25 →Generator Module: Long-Form Answers with References
- 19:0819:08, 6 April 2025 diff hist +3,771 stat946W25 →WebGLM: Efficient Web-Enhanced Question Answering
- 18:5318:53, 6 April 2025 diff hist +54 N File:Evaluation.png Evaluation on LLM’s reference adoption current
- 18:4818:48, 6 April 2025 diff hist +44 N File:retriever time.png WebGLM retriever time analysis current
- 17:5517:55, 6 April 2025 diff hist +6 stat946W25 →Essential Background & Inspiration
- 15:3415:34, 6 April 2025 diff hist +2,306 stat946W25 →WebGLM: Efficient Web-Enhanced Question Answering
- 14:2214:22, 6 April 2025 diff hist +8 stat946W25 →WebGLM: Efficient Web-Enhanced Question Answering
- 14:0214:02, 6 April 2025 diff hist +1,926 stat946W25 →WebGLM: Efficient Web-Enhanced Question Answering
- 13:5413:54, 6 April 2025 diff hist +56 N File:WebGLM1.png WebGLM’s response to an example question current
- 12:0312:03, 6 April 2025 diff hist +28 stat946W25 →Model Architecture and Training Methods
- 11:5511:55, 6 April 2025 diff hist +141 stat946W25 →Topic 20: Diffusion Language Model
- 11:5011:50, 6 April 2025 diff hist +91 N File:tSNE plot.png A t-SNE plot of the learned word embeddings. Each word is colored by its POS. current
- 11:4711:47, 6 April 2025 diff hist +6,317 stat946W25 →Diffusion-LM: A Continuous Diffusion Model for Controllable Text Generation
5 April 2025
- 16:3516:35, 5 April 2025 diff hist +154 stat946W25 →Mathematical Formulation
- 16:2916:29, 5 April 2025 diff hist +89 N File:forward reverse diffusion processes.png A graphical model representing the forward and reverse diffusion processes. current
- 16:2716:27, 5 April 2025 diff hist +3,830 stat946W25 →Diffusion-LM: A Continuous Diffusion Model for Controllable Text Generation
- 15:1815:18, 5 April 2025 diff hist −1 stat946W25 →Controllable Generation Tasks
- 15:1715:17, 5 April 2025 diff hist −211 stat946W25 →Diffusion-LM: A Continuous Diffusion Model for Controllable Text Generation
- 14:0414:04, 5 April 2025 diff hist +702 stat946W25 →Controllable Generation with Diffusion-LM
- 12:3912:39, 5 April 2025 diff hist +21 stat946W25 →Exposure Bias in AR Models
29 March 2025
- 15:0615:06, 29 March 2025 diff hist +6 stat946W25 →H_2O: Efficient KV Cache Compression for Large Language Models
- 15:0515:05, 29 March 2025 diff hist +6,318 stat946W25 →Method
28 March 2025
- 15:5815:58, 28 March 2025 diff hist +3,625 stat946W25 →H_2O: Efficient KV Cache Compression for Large Language Models
- 15:1915:19, 28 March 2025 diff hist −27 stat946W25 →The Problem This Paper Tried to Address
- 15:1815:18, 28 March 2025 diff hist +286 stat946W25 →Method
- 14:5714:57, 28 March 2025 diff hist +1,106 stat946W25 →Key Contributions
- 13:1513:15, 28 March 2025 diff hist −5 stat946W25 →The Problem This Paper Tried to Address
- 13:1313:13, 28 March 2025 diff hist +191 stat946W25 →The Problem This Paper Tried to Address
- 13:0813:08, 28 March 2025 diff hist +803 stat946W25 →The Problem This Paper Tried to Address
- 13:0613:06, 28 March 2025 diff hist +39 N File:H2Q Figure1.png accuracy-memory trade-off current
- 12:0012:00, 28 March 2025 diff hist +28 stat946W25 →Limitations and Future Work
- 12:0012:00, 28 March 2025 diff hist +240 stat946W25 →Limitations and Future Work
- 00:1600:16, 28 March 2025 diff hist −26 stat946W25 →The Problem This Paper Tried to Address
- 00:0500:05, 28 March 2025 diff hist +181 stat946W25 →Background
21 March 2025
- 21:3821:38, 21 March 2025 diff hist +12 stat946W25 →1- Fine-grained Hardware-friendly Quantization Scheme
- 21:3321:33, 21 March 2025 diff hist +1,264 stat946W25 →1- Fine-grained Hardware-friendly Quantization Scheme
- 20:5920:59, 21 March 2025 diff hist +1,446 stat946W25 →1- Fine-grained Hardware-friendly Quantization Scheme
- 20:2920:29, 21 March 2025 diff hist +181 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
- 20:2220:22, 21 March 2025 diff hist +121 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
- 20:2120:21, 21 March 2025 diff hist +1,244 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
- 18:4118:41, 21 March 2025 diff hist +1,624 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
- 00:4400:44, 21 March 2025 diff hist +710 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
- 00:2500:25, 21 March 2025 diff hist +503 stat946W25 →Layer-by-Layer Knowledge Distillation (LKD)
- 00:2200:22, 21 March 2025 diff hist −4 stat946W25 →Topic 4: Quantization
- 00:2100:21, 21 March 2025 diff hist −1,093 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
- 00:0600:06, 21 March 2025 diff hist +2,155 stat946W25 →ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers