From a75cb7bc9b9b9e39284d75f9c9e941a584dd24cf Mon Sep 17 00:00:00 2001 From: xpli Date: Thu, 25 May 2023 10:06:16 +0800 Subject: [PATCH 1/3] add some new papers --- README.md | 48 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 35 insertions(+), 13 deletions(-) diff --git a/README.md b/README.md index 9ba1c0e..17deae3 100644 --- a/README.md +++ b/README.md @@ -17,21 +17,30 @@ This is a collection of research and review papers of Model Editing. Any suggest - Eric Mitchell, Charles Lin, Antoine Bosselut, Christopher D. Manning, Chelsea Finn.
**Memory-Based Model Editing at Scale**. (ICML 2022) [[paper](https://arxiv.org/abs/2206.06520)] [[code](https://github.com/eric-mitchell/serac)] [[demo](https://sites.google.com/view/serac-editing)] - - Shikhar Murty, Christopher D. Manning, Scott M. Lundberg, Marco Túlio Ribeiro.
**Fixing Model Bugs with Natural Language Patches**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2211.03318)] [[code](https://github.com/MurtyShikhar/LanguagePatching)] - - Aman Madaan, Niket Tandon, Peter Clark, Yiming Yang.
**MemPrompt: Memory-assisted Prompt Editing with User Feedback**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2201.06009)] [[code](https://github.com/madaan/memprompt)] [[page](https://memprompt.com/)] [[video](https://www.youtube.com/watch?v=Ld7R02bOiNQ&t=1s)] - - Daliang Li, Ankit Singh Rawat, Manzil Zaheer, Xin Wang, Michal Lukasik, Andreas Veit, Felix Yu, Sanjiv Kumar.
**Large Language Models with Controllable Working Memory**. [[paper](https://arxiv.org/abs/2211.05110)] - - Qingxiu Dong, Damai Dai, Yifan Song, Jingjing Xu, Zhifang Sui, Lei Li.
-**CaliNet: Calibrating Factual Knowledge in Pretrained Language Models**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2210.03329)] [[code](https://github.com/dqxiu/CaliNet)] - +**Calibrating Factual Knowledge in Pretrained Language Models**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2210.03329)] [[code](https://github.com/dqxiu/CaliNet)] - Zeyu Huang, Yikang Shen, Xiaofeng Zhang, Jie Zhou, Wenge Rong, Zhang Xiong.
-**Transformer-Patcher: One Mistake worth One Neuron**. (ICLR 2023) [[paper](https://arxiv.org/abs/2301.09785)] [[code](https://github.com/ZeroYuHuang/Transformer-Patcher)] + **Transformer-Patcher: One Mistake worth One Neuron**. (ICLR 2023) [[paper](https://arxiv.org/abs/2301.09785)] [[code](https://github.com/ZeroYuHuang/Transformer-Patcher)] +- Thomas Hartvigsen, Swami Sankaranarayanan, Hamid Palangi, Yoon Kim, Marzyeh Ghassemi.
+ **Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors**. [[paper](https://arxiv.org/abs/2211.11031)] [[code](https://github.com/thartvigsen/grace)] + +#### Change LM's representation space + +- Evan Hernandez, Belinda Z. Li, Jacob Andreas.
+ + **Measuring and Manipulating Knowledge Representations in Language Models**.[[paper](http://arxiv.org/abs/2304.00740)] [[code](https://github.com/evandez/REMEDI)] + +#### Memory extension + +- Damai Dai, Wenbin Jiang, Qingxiu Dong, Yajuan Lyu, Qiaoqiao She, Zhifang Sui.
+ + **Neural Knowledge Bank for Pretrained Transformers**.[[paper](http://arxiv.org/abs/2208.00399)] ### Modify Parameter @@ -67,13 +76,14 @@ This is a collection of research and review papers of Model Editing. Any suggest **Mass-Editing Memory in a Transformer**. [[paper](https://arxiv.org/abs/2210.07229)] [[code](https://github.com/kmeng01/memit)] [[page](https://memit.baulab.info/)] [[demo](https://memit.baulab.us/#/)] - Peter Hase, Mona Diab, Asli Celikyilmaz, Xian Li, Zornitsa Kozareva, Veselin Stoyanov, Mohit Bansal, Srinivasan Iyer.
-**Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs**. [[paper](https://arxiv.org/pdf/2111.13654.pdf)] [[code](https://github.com/peterbhase/SLAG-Belief-Updating)] - -- Thomas Hartvigsen, Swami Sankaranarayanan, Hamid Palangi, Yoon Kim, Marzyeh Ghassemi.
-**Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors**. [[paper](https://arxiv.org/abs/2211.11031)] [[code](https://github.com/thartvigsen/grace)] + **Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs**. [[paper](https://arxiv.org/pdf/2111.13654.pdf)] [[code](https://github.com/peterbhase/SLAG-Belief-Updating)] - Peter Hase, Mohit Bansal, Been Kim, Asma Ghandeharioun.
-**Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models.** [[paper](https://arxiv.org/pdf/2301.04213.pdf)] [[code](https://github.com/google/belief-localization)] + **Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models.** [[paper](https://arxiv.org/pdf/2301.04213.pdf)] [[code](https://github.com/google/belief-localization)] + +- Damai Dai , Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, Furu Wei.
+ + **Knowledge Neurons in Pretrained Transformers**.(ACL 2022)[[paper](http://arxiv.org/abs/2104.08696)] [[code](https://github.com/Hunter-DDM/knowledge-neurons)] [[code by EleutherAI](https://github.com/EleutherAI/knowledge-neurons)] ### More Papers Related @@ -90,4 +100,16 @@ This is a collection of research and review papers of Model Editing. Any suggest **Prompting GPT-3 To Be Reliable**. [[paper](https://arxiv.org/abs/2210.09150)] - Gabriel Ilharco, Mitchell Wortsman, Samir Yitzhak Gadre, Shuran Song, Hannaneh Hajishirzi, Simon Kornblith, Ali Farhadi, Ludwig Schmidt.
-**Patching open-vocabulary models by interpolating weights**. (NeurIPS 2022) [[paper](https://arxiv.org/abs/2208.05592)] [[code](https://github.com/mlfoundations/patching)] + **Patching open-vocabulary models by interpolating weights**. (NeurIPS 2022) [[paper](https://arxiv.org/abs/2208.05592)] [[code](https://github.com/mlfoundations/patching)] + +- Xin Cheng, Yankai Lin, Xiuying Chen, Dongyan Zhao, Rui Yan.
+ + **Decouple knowledge from paramters for plug-and-play language modeling**. (ACL2023 Findings)[[paper](http://arxiv.org/abs/2305.11564)] [[code](https://github.com/Hannibal046/PlugLM)] + + + +[ (https://github.com/Hunter-DDM/knowledge-neurons)]: +[(http://arxiv.org/abs/2104.08696)]: +[(https://github.com/Hannibal046/PlugLM)]: + +[(http://arxiv.org/abs/2305.11564)]: From 19277eee0b3d0001311386d84a27c9af210be1c3 Mon Sep 17 00:00:00 2001 From: xpli Date: Thu, 25 May 2023 10:12:46 +0800 Subject: [PATCH 2/3] fix format --- README.md | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/README.md b/README.md index 17deae3..196eb56 100644 --- a/README.md +++ b/README.md @@ -17,14 +17,15 @@ This is a collection of research and review papers of Model Editing. Any suggest - Eric Mitchell, Charles Lin, Antoine Bosselut, Christopher D. Manning, Chelsea Finn.
**Memory-Based Model Editing at Scale**. (ICML 2022) [[paper](https://arxiv.org/abs/2206.06520)] [[code](https://github.com/eric-mitchell/serac)] [[demo](https://sites.google.com/view/serac-editing)] + - Shikhar Murty, Christopher D. Manning, Scott M. Lundberg, Marco Túlio Ribeiro.
-**Fixing Model Bugs with Natural Language Patches**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2211.03318)] [[code](https://github.com/MurtyShikhar/LanguagePatching)] + **Fixing Model Bugs with Natural Language Patches**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2211.03318)] [[code](https://github.com/MurtyShikhar/LanguagePatching)] - Aman Madaan, Niket Tandon, Peter Clark, Yiming Yang.
-**MemPrompt: Memory-assisted Prompt Editing with User Feedback**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2201.06009)] [[code](https://github.com/madaan/memprompt)] [[page](https://memprompt.com/)] [[video](https://www.youtube.com/watch?v=Ld7R02bOiNQ&t=1s)] + **MemPrompt: Memory-assisted Prompt Editing with User Feedback**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2201.06009)] [[code](https://github.com/madaan/memprompt)] [[page](https://memprompt.com/)] [[video](https://www.youtube.com/watch?v=Ld7R02bOiNQ&t=1s)] - Daliang Li, Ankit Singh Rawat, Manzil Zaheer, Xin Wang, Michal Lukasik, Andreas Veit, Felix Yu, Sanjiv Kumar.
-**Large Language Models with Controllable Working Memory**. [[paper](https://arxiv.org/abs/2211.05110)] + **Large Language Models with Controllable Working Memory**. [[paper](https://arxiv.org/abs/2211.05110)] - Qingxiu Dong, Damai Dai, Yifan Song, Jingjing Xu, Zhifang Sui, Lei Li.
-**Calibrating Factual Knowledge in Pretrained Language Models**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2210.03329)] [[code](https://github.com/dqxiu/CaliNet)] + **Calibrating Factual Knowledge in Pretrained Language Models**. (EMNLP 2022) [[paper](https://arxiv.org/abs/2210.03329)] [[code](https://github.com/dqxiu/CaliNet)] - Zeyu Huang, Yikang Shen, Xiaofeng Zhang, Jie Zhou, Wenge Rong, Zhang Xiong.
**Transformer-Patcher: One Mistake worth One Neuron**. (ICLR 2023) [[paper](https://arxiv.org/abs/2301.09785)] [[code](https://github.com/ZeroYuHuang/Transformer-Patcher)] - Thomas Hartvigsen, Swami Sankaranarayanan, Hamid Palangi, Yoon Kim, Marzyeh Ghassemi.
@@ -33,13 +34,11 @@ This is a collection of research and review papers of Model Editing. Any suggest #### Change LM's representation space - Evan Hernandez, Belinda Z. Li, Jacob Andreas.
- **Measuring and Manipulating Knowledge Representations in Language Models**.[[paper](http://arxiv.org/abs/2304.00740)] [[code](https://github.com/evandez/REMEDI)] #### Memory extension - Damai Dai, Wenbin Jiang, Qingxiu Dong, Yajuan Lyu, Qiaoqiao She, Zhifang Sui.
- **Neural Knowledge Bank for Pretrained Transformers**.[[paper](http://arxiv.org/abs/2208.00399)] ### Modify Parameter @@ -81,9 +80,8 @@ This is a collection of research and review papers of Model Editing. Any suggest - Peter Hase, Mohit Bansal, Been Kim, Asma Ghandeharioun.
**Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models.** [[paper](https://arxiv.org/pdf/2301.04213.pdf)] [[code](https://github.com/google/belief-localization)] -- Damai Dai , Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, Furu Wei.
+- Damai Dai , Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, Furu Wei.
**Knowledge Neurons in Pretrained Transformers**.(ACL 2022)[[paper](http://arxiv.org/abs/2104.08696)] [[code](https://github.com/Hunter-DDM/knowledge-neurons)] [[code by EleutherAI](https://github.com/EleutherAI/knowledge-neurons)] - **Knowledge Neurons in Pretrained Transformers**.(ACL 2022)[[paper](http://arxiv.org/abs/2104.08696)] [[code](https://github.com/Hunter-DDM/knowledge-neurons)] [[code by EleutherAI](https://github.com/EleutherAI/knowledge-neurons)] ### More Papers Related @@ -102,9 +100,7 @@ This is a collection of research and review papers of Model Editing. Any suggest - Gabriel Ilharco, Mitchell Wortsman, Samir Yitzhak Gadre, Shuran Song, Hannaneh Hajishirzi, Simon Kornblith, Ali Farhadi, Ludwig Schmidt.
**Patching open-vocabulary models by interpolating weights**. (NeurIPS 2022) [[paper](https://arxiv.org/abs/2208.05592)] [[code](https://github.com/mlfoundations/patching)] -- Xin Cheng, Yankai Lin, Xiuying Chen, Dongyan Zhao, Rui Yan.
- - **Decouple knowledge from paramters for plug-and-play language modeling**. (ACL2023 Findings)[[paper](http://arxiv.org/abs/2305.11564)] [[code](https://github.com/Hannibal046/PlugLM)] +- Xin Cheng, Yankai Lin, Xiuying Chen, Dongyan Zhao, Rui Yan.
**Decouple knowledge from paramters for plug-and-play language modeling**. (ACL2023 Findings)[[paper](http://arxiv.org/abs/2305.11564)] [[code](https://github.com/Hannibal046/PlugLM)] From a16eff77a47c6bd84c98cf6ccb949680e86aa27f Mon Sep 17 00:00:00 2001 From: xpli Date: Thu, 25 May 2023 10:15:03 +0800 Subject: [PATCH 3/3] fix redundant link --- README.md | 7 ------- 1 file changed, 7 deletions(-) diff --git a/README.md b/README.md index 196eb56..6b680c7 100644 --- a/README.md +++ b/README.md @@ -102,10 +102,3 @@ This is a collection of research and review papers of Model Editing. Any suggest - Xin Cheng, Yankai Lin, Xiuying Chen, Dongyan Zhao, Rui Yan.
**Decouple knowledge from paramters for plug-and-play language modeling**. (ACL2023 Findings)[[paper](http://arxiv.org/abs/2305.11564)] [[code](https://github.com/Hannibal046/PlugLM)] - - -[ (https://github.com/Hunter-DDM/knowledge-neurons)]: -[(http://arxiv.org/abs/2104.08696)]: -[(https://github.com/Hannibal046/PlugLM)]: - -[(http://arxiv.org/abs/2305.11564)]: