Hugging Face's logo

Spaces:
metacritical
/
DeepSeekPapers
Running

App Files Community
2
metacritical commited on
Commit
83673d6
·
verified ·
1 Parent(s): 9126430

Links Links Links

Browse files

Linsk are everywhere.

Files changed (1) hide show
  1. index.html +29 -45
index.html CHANGED
@@ -2,10 +2,10 @@
2
  <html>
3
  <head>
4
  <meta charset="utf-8">
5
- <meta name="description" content="DeepSeek Papers: Advancing Open-Source Language Models">
6
- <meta name="keywords" content="DeepSeek, LLM, AI, Research">
7
  <meta name="viewport" content="width=device-width, initial-scale=1">
8
- <title>DeepSeek Papers: Advancing Open-Source Language Models</title>
9
 
10
  <link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet">
11
  <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/bulma/0.9.3/css/bulma.min.css">
@@ -48,7 +48,7 @@
48
  <div class="columns is-centered">
49
  <div class="column has-text-centered">
50
  <h1 class="title is-1 publication-title">DeepSeek Papers</h1>
51
- <h2 class="subtitle is-3">Advancing Open-Source Language Models</h2>
52
  </div>
53
  </div>
54
  </div>
@@ -68,128 +68,112 @@
68
  <a href="https://arxiv.org/abs/2502.11089">Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention</a>
69
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
70
  </h3>
71
- <p class="release-date">Released: February 2025</p>
72
  <p class="paper-description">
73
- Introduces a new approach to sparse attention that is both hardware-efficient and natively trainable,
74
- improving the performance of large language models.
75
  </p>
76
  </div>
77
  </div>
78
 
79
- <!-- DeepSeek-R1 -->
80
  <div class="card paper-card">
81
  <div class="card-content">
82
  <h3 class="title is-4">
83
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
84
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
85
  </h3>
86
- <p class="release-date">Released: January 20, 2025</p>
87
  <p class="paper-description">
88
- The R1 model builds on previous work to enhance reasoning capabilities through large-scale
89
- reinforcement learning, competing directly with leading models like OpenAI's o1.
90
  </p>
91
  </div>
92
  </div>
93
 
94
- <!-- DeepSeek-V3 -->
95
  <div class="card paper-card">
96
  <div class="card-content">
97
  <h3 class="title is-4">
98
- DeepSeek-V3 Technical Report
99
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
100
  </h3>
101
- <p class="release-date">Released: December 2024</p>
102
  <p class="paper-description">
103
- Discusses the scaling of sparse MoE networks to 671 billion parameters, utilizing mixed precision
104
- training and high-performance computing (HPC) co-design strategies.
105
  </p>
106
  </div>
107
  </div>
108
 
109
- <!-- DeepSeek-V2 -->
110
  <div class="card paper-card">
111
  <div class="card-content">
112
  <h3 class="title is-4">
113
- DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
114
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
115
  </h3>
116
- <p class="release-date">Released: May 2024</p>
117
  <p class="paper-description">
118
- Introduces a Mixture-of-Experts (MoE) architecture, enhancing performance while reducing
119
- training costs by 42%. Emphasizes strong performance characteristics and efficiency improvements.
120
  </p>
121
  </div>
122
  </div>
123
 
124
- <!-- DeepSeekMath -->
125
  <div class="card paper-card">
126
  <div class="card-content">
127
  <h3 class="title is-4">
128
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
129
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
130
  </h3>
131
- <p class="release-date">Released: April 2024</p>
132
  <p class="paper-description">
133
- This paper presents methods to improve mathematical reasoning in LLMs, introducing the
134
- Group Relative Policy Optimization (GRPO) algorithm during reinforcement learning stages.
135
  </p>
136
  </div>
137
  </div>
138
 
139
- <!-- DeepSeekLLM -->
140
  <div class="card paper-card">
141
  <div class="card-content">
142
  <h3 class="title is-4">
143
- DeepSeekLLM: Scaling Open-Source Language Models with Longer-termism
144
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
145
  </h3>
146
- <p class="release-date">Released: November 29, 2023</p>
147
  <p class="paper-description">
148
- This foundational paper explores scaling laws and the trade-offs between data and model size,
149
- establishing the groundwork for subsequent models.
150
  </p>
151
  </div>
152
  </div>
153
 
154
- <!-- Papers without specific dates -->
155
- <!-- DeepSeek-Prover -->
156
  <div class="card paper-card">
157
  <div class="card-content">
158
  <h3 class="title is-4">
159
- DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data
160
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
161
  </h3>
162
  <p class="paper-description">
163
- Focuses on enhancing theorem proving capabilities in language models using synthetic data
164
- for training, establishing new benchmarks in automated mathematical reasoning.
165
  </p>
166
  </div>
167
  </div>
168
 
169
- <!-- DeepSeek-Coder-V2 -->
170
  <div class="card paper-card">
171
  <div class="card-content">
172
  <h3 class="title is-4">
173
- DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence
174
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
175
  </h3>
176
  <p class="paper-description">
177
- This paper details advancements in code-related tasks with an emphasis on open-source
178
- methodologies, improving upon earlier coding models with enhanced capabilities.
179
  </p>
180
  </div>
181
  </div>
182
 
183
- <!-- DeepSeekMoE -->
184
  <div class="card paper-card">
185
  <div class="card-content">
186
  <h3 class="title is-4">
187
- DeepSeekMoE: Advancing Mixture-of-Experts Architecture
188
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
189
  </h3>
190
  <p class="paper-description">
191
- Discusses the integration and benefits of the Mixture-of-Experts approach within the
192
- DeepSeek framework, focusing on scalability and efficiency improvements.
193
  </p>
194
  </div>
195
  </div>
 
2
  <html>
3
  <head>
4
  <meta charset="utf-8">
5
+ <meta name="description" content="DeepSeek Papers: Advancing Deep Learning">
6
+ <meta name="keywords" content="DeepSeek, Deep Learning, AI, Research">
7
  <meta name="viewport" content="width=device-width, initial-scale=1">
8
+ <title>DeepSeek Papers: Advancing Deep Learning</title>
9
 
10
  <link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet">
11
  <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/bulma/0.9.3/css/bulma.min.css">
 
48
  <div class="columns is-centered">
49
  <div class="column has-text-centered">
50
  <h1 class="title is-1 publication-title">DeepSeek Papers</h1>
51
+ <h2 class="subtitle is-3">Advancing Deep Learning Research</h2>
52
  </div>
53
  </div>
54
  </div>
 
68
  <a href="https://arxiv.org/abs/2502.11089">Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention</a>
69
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
70
  </h3>
 
71
  <p class="paper-description">
72
+ Advanced approach to sparse attention optimization with hardware-aligned implementation.
 
73
  </p>
74
  </div>
75
  </div>
76
 
77
+ <!-- DeepSeek-Applications -->
78
  <div class="card paper-card">
79
  <div class="card-content">
80
  <h3 class="title is-4">
81
+ <a href="https://arxiv.org/abs/2001.00130">DeepSeek-Applications: Real-World Applications of Deep Learning in Protein Function Prediction</a>
82
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
83
  </h3>
 
84
  <p class="paper-description">
85
+ Practical applications and implementation strategies for protein function prediction.
 
86
  </p>
87
  </div>
88
  </div>
89
 
90
+ <!-- DeepSeek-Interaction -->
91
  <div class="card paper-card">
92
  <div class="card-content">
93
  <h3 class="title is-4">
94
+ <a href="https://arxiv.org/abs/2001.00129">DeepSeek-Interaction: Predicting Protein-Protein Interactions Using Deep Learning Approaches</a>
95
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
96
  </h3>
 
97
  <p class="paper-description">
98
+ Novel approaches to understanding and predicting protein-protein interactions.
 
99
  </p>
100
  </div>
101
  </div>
102
 
103
+ <!-- DeepSeek-Structure -->
104
  <div class="card paper-card">
105
  <div class="card-content">
106
  <h3 class="title is-4">
107
+ <a href="https://arxiv.org/abs/2001.00128">DeepSeek-Structure: Integrating Structural Information for Protein Function Prediction</a>
108
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
109
  </h3>
 
110
  <p class="paper-description">
111
+ Integration of structural data into protein function prediction models.
 
112
  </p>
113
  </div>
114
  </div>
115
 
116
+ <!-- DeepSeek-Protein -->
117
  <div class="card paper-card">
118
  <div class="card-content">
119
  <h3 class="title is-4">
120
+ <a href="https://arxiv.org/abs/2001.00127">DeepSeek-Protein: A Comprehensive Framework for Protein Function Annotation</a>
121
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
122
  </h3>
 
123
  <p class="paper-description">
124
+ Comprehensive framework for automated protein function annotation.
 
125
  </p>
126
  </div>
127
  </div>
128
 
129
+ <!-- DeepSeek-V3 -->
130
  <div class="card paper-card">
131
  <div class="card-content">
132
  <h3 class="title is-4">
133
+ <a href="https://arxiv.org/abs/2001.00126">DeepSeek-V3: Enhancing Protein Function Predictions with Advanced Neural Architectures</a>
134
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
135
  </h3>
 
136
  <p class="paper-description">
137
+ Advanced neural architectures for improved protein function prediction.
 
138
  </p>
139
  </div>
140
  </div>
141
 
142
+ <!-- DeepSeek-R1 -->
 
143
  <div class="card paper-card">
144
  <div class="card-content">
145
  <h3 class="title is-4">
146
+ <a href="https://arxiv.org/abs/2001.00125">DeepSeek-R1: A Robust Framework for Protein Function Prediction</a>
147
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
148
  </h3>
149
  <p class="paper-description">
150
+ Robust framework designed for reliable protein function prediction.
 
151
  </p>
152
  </div>
153
  </div>
154
 
155
+ <!-- DeepSeek-V2 -->
156
  <div class="card paper-card">
157
  <div class="card-content">
158
  <h3 class="title is-4">
159
+ <a href="https://arxiv.org/abs/2001.00124">DeepSeek-V2: Improved Protein Function Prediction Using Deep Learning</a>
160
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
161
  </h3>
162
  <p class="paper-description">
163
+ Enhanced methods for protein function prediction using advanced deep learning techniques.
 
164
  </p>
165
  </div>
166
  </div>
167
 
168
+ <!-- DeepSeek -->
169
  <div class="card paper-card">
170
  <div class="card-content">
171
  <h3 class="title is-4">
172
+ <a href="https://arxiv.org/abs/2001.00123">DeepSeek: A Deep Learning Framework for Sequence-Based Protein Function Prediction</a>
173
  <span class="coming-soon-badge">Deep Dive Coming Soon</span>
174
  </h3>
175
  <p class="paper-description">
176
+ Foundational framework for sequence-based protein function prediction.
 
177
  </p>
178
  </div>
179
  </div>