Skip to content

Commit c10e1ca

Browse files
committed
Updated on 2024-11-26
1 parent b25006a commit c10e1ca

File tree

2 files changed

+31
-12
lines changed

2 files changed

+31
-12
lines changed

papers/list.json

Lines changed: 14 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,16 @@
11
[
2+
{
3+
"title": "A Simple Early Exiting Framework for Accelerating Sampling in Diffusion Models",
4+
"author": "Taehong Moon et al",
5+
"year": "2024",
6+
"topic": "diffusion, early exit",
7+
"venue": "ICML",
8+
"description": "This paper presents Adaptive Score Estimation (ASE), a novel framework that accelerates diffusion model sampling by adaptively allocating computational resources based on the time step being processed. The authors observe that score estimation near the noise distribution (t→1) requires less computational power than estimation near the data distribution (t→0), leading them to develop a time-dependent early-exiting scheme where more neural network blocks are skipped during the noise-phase sampling steps. Their approach differs between architectures - for DiT models they skip entire blocks, while for U-ViT models they preserve the linear layers connected to skip connections while dropping other block components to maintain the residual pathway information. The authors fine-tune their models using a specially designed training procedure that employs exponential moving averages and weighted coefficients to ensure minimal information updates near t→0 while allowing more updates near t→1.",
9+
"link": "https://arxiv.org/pdf/2408.05927"
10+
},
211
{
312
"title": "Active Prompting with Chain-of-Thought for Large Language Models",
4-
"author": "Shizhe Diao, et al",
13+
"author": "Shizhe Diao et al",
514
"year": "2023",
615
"topic": "prompting, cot",
716
"venue": "Arxiv",
@@ -10,7 +19,7 @@
1019
},
1120
{
1221
"title": "RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment",
13-
"author": "Hanze Dong, et al",
22+
"author": "Hanze Dong et al",
1423
"year": "2023",
1524
"topic": "watermark, offset learning",
1625
"venue": "TMLR",
@@ -19,7 +28,7 @@
1928
},
2029
{
2130
"title": "Finding needles in a haystack: A Black-Box Approach to Invisible Watermark Detection",
22-
"author": "Minzhou Pan, et al",
31+
"author": "Minzhou Pan et al",
2332
"year": "2024",
2433
"topic": "watermark, offset learning",
2534
"venue": "Arxiv",
@@ -28,7 +37,7 @@
2837
},
2938
{
3039
"title": "Mitigating the Alignment Tax of RLHF",
31-
"author": "Yong Lin, et al",
40+
"author": "Yong Lin et al",
3241
"year": "2024",
3342
"topic": "rlhf, alignment",
3443
"venue": "Arxiv",
@@ -37,7 +46,7 @@
3746
},
3847
{
3948
"title": "AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising",
40-
"author": "Zigeng Chen, et al",
49+
"author": "Zigeng Chen et al",
4150
"year": "2024",
4251
"topic": "diffusion, parallelization, denoising",
4352
"venue": "Arxiv",

papers_read.html

Lines changed: 17 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -75,10 +75,10 @@ <h1>Here's where I keep a list of papers I have read.</h1>
7575
I typically use this to organize papers I found interesting. Please feel free to do whatever you want with it. Note that this is not every single paper I have ever read, just a collection of ones that I remember to put down.
7676
</p>
7777
<p id="paperCount">
78-
So far, we have read 180 papers. Let's keep it up!
78+
So far, we have read 181 papers. Let's keep it up!
7979
</p>
8080
<small id="searchCount">
81-
Your search returned 180 papers. Nice!
81+
Your search returned 181 papers. Nice!
8282
</small>
8383

8484
<div class="search-inputs">
@@ -105,9 +105,19 @@ <h1>Here's where I keep a list of papers I have read.</h1>
105105
</thead>
106106
<tbody>
107107

108+
<tr>
109+
<td>A Simple Early Exiting Framework for Accelerating Sampling in Diffusion Models</td>
110+
<td>Taehong Moon et al</td>
111+
<td>2024</td>
112+
<td>diffusion, early exit</td>
113+
<td>ICML</td>
114+
<td>This paper presents Adaptive Score Estimation (ASE), a novel framework that accelerates diffusion model sampling by adaptively allocating computational resources based on the time step being processed. The authors observe that score estimation near the noise distribution (t→1) requires less computational power than estimation near the data distribution (t→0), leading them to develop a time-dependent early-exiting scheme where more neural network blocks are skipped during the noise-phase sampling steps. Their approach differs between architectures - for DiT models they skip entire blocks, while for U-ViT models they preserve the linear layers connected to skip connections while dropping other block components to maintain the residual pathway information. The authors fine-tune their models using a specially designed training procedure that employs exponential moving averages and weighted coefficients to ensure minimal information updates near t→0 while allowing more updates near t→1.</td>
115+
<td><a href="https://arxiv.org/pdf/2408.05927" target="_blank">Link</a></td>
116+
</tr>
117+
108118
<tr>
109119
<td>Active Prompting with Chain-of-Thought for Large Language Models</td>
110-
<td>Shizhe Diao, et al</td>
120+
<td>Shizhe Diao et al</td>
111121
<td>2023</td>
112122
<td>prompting, cot</td>
113123
<td>Arxiv</td>
@@ -117,7 +127,7 @@ <h1>Here's where I keep a list of papers I have read.</h1>
117127

118128
<tr>
119129
<td>RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment</td>
120-
<td>Hanze Dong, et al</td>
130+
<td>Hanze Dong et al</td>
121131
<td>2023</td>
122132
<td>watermark, offset learning</td>
123133
<td>TMLR</td>
@@ -127,7 +137,7 @@ <h1>Here's where I keep a list of papers I have read.</h1>
127137

128138
<tr>
129139
<td>Finding needles in a haystack: A Black-Box Approach to Invisible Watermark Detection</td>
130-
<td>Minzhou Pan, et al</td>
140+
<td>Minzhou Pan et al</td>
131141
<td>2024</td>
132142
<td>watermark, offset learning</td>
133143
<td>Arxiv</td>
@@ -137,7 +147,7 @@ <h1>Here's where I keep a list of papers I have read.</h1>
137147

138148
<tr>
139149
<td>Mitigating the Alignment Tax of RLHF</td>
140-
<td>Yong Lin, et al</td>
150+
<td>Yong Lin et al</td>
141151
<td>2024</td>
142152
<td>rlhf, alignment</td>
143153
<td>Arxiv</td>
@@ -147,7 +157,7 @@ <h1>Here's where I keep a list of papers I have read.</h1>
147157

148158
<tr>
149159
<td>AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising</td>
150-
<td>Zigeng Chen, et al</td>
160+
<td>Zigeng Chen et al</td>
151161
<td>2024</td>
152162
<td>diffusion, parallelization, denoising</td>
153163
<td>Arxiv</td>

0 commit comments

Comments
 (0)