繁體中文
English Indonesia Türkçe Deutsch Español Português Pусский Français Italiano Nederlands Polski العربية ไทย Tiếng Việt 繁體中文 简体中文 日本語 한국어

#knowledgeediting 搜尋結果

qx_dong's profile picture. PhD student @PKU1898. Research Intern @MSFTResearch Asia.
Qingxiu Dong
 @qx_dong
2023年5月23日

The In-context Learning Editing (IKE) method proposed in this paper showcases impressive performance, even in complex multi-hop related scenarios. #knowledgeediting

qx_dong's tweet image. The In-context Learning Editing (IKE) method proposed in this paper showcases impressive performance, even in complex multi-hop related scenarios. #knowledgeediting
0
0
7
1
539
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2024年8月1日

Welcome to attend the tutorial on "Knowledge Editing for Large Language Models" at IJCAI 2024 @IJCAIconf . Here are the details #NLP #AI #KnowledgeEditing #LLMs : Tutorial: Knowledge Editing for Large Language Models Location: Room 1F-Yeongju B Time: 9:00-12:30, August 3, 2024…

zxlzr's tweet image. Welcome to attend the tutorial on "Knowledge Editing for Large Language Models" at IJCAI 2024 @IJCAIconf . Here are the details #NLP #AI #KnowledgeEditing #LLMs : Tutorial: Knowledge Editing for Large Language Models Location: Room 1F-Yeongju B Time: 9:00-12:30, August 3, 2024…
1
13
70
24
6千
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年1月12日

Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…

zxlzr's tweet image. Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…
2
3
45
9
5千
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2023年10月14日

Our EasyEdit now supports editing multi-modal LLMs. Currently, you can edit MiniGPT-4 and Blip2. Check out the details here: github.com/zjunlp/EasyEdi…. Feel free to follow us for updates. Related paper: arxiv.org/abs/2310.08475. 🚀📚 #EasyEdit #ModelEditing #KnowledgeEditing…

zxlzr's tweet image. Our EasyEdit now supports editing multi-modal LLMs. Currently, you can edit MiniGPT-4 and Blip2. Check out the details here: github.com/zjunlp/EasyEdi…. Feel free to follow us for updates. Related paper: arxiv.org/abs/2310.08475. 🚀📚 #EasyEdit #ModelEditing #KnowledgeEditing…
0
9
39
9
6千
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2024年5月21日

The tutorial of Knowledge Editing for Large Language Models at LREC-COLING 2024 @LrecColing is in progress. #NLP #LLMs #KnowledgeEditing

zxlzr's tweet image. The tutorial of Knowledge Editing for Large Language Models at LREC-COLING 2024 @LrecColing is in progress. #NLP #LLMs #KnowledgeEditing
0
1
19
0
1千
longevityboris's profile picture. Founder @ 199 Biotechnologies, 199 Clinic, 199 Diagnostics & 199 Health | Investor & futurist advancing longevity, partial reprogramming & preventive medicine.
Boris Djordjevic
@longevityboris
年7月2日

Weight editors like ROME & MEMIT can overwrite facts, and ELDER stacks LoRA adapters for lifelong tweaks—but “ripple effects” still flip 20-30 % of related answers. #KnowledgeEditing

longevityboris's tweet image. Weight editors like ROME & MEMIT can overwrite facts, and ELDER stacks LoRA adapters for lifelong tweaks—but “ripple effects” still flip 20-30 % of related answers. #KnowledgeEditing
1
0
0
0
21
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年9月10日

Very glad that our work EasyEdit2 has been accepted to EMNLP 2025 Demonstration @emnlpmeeting , and looking forward to exchanging ideas together in Suzhou #EMNLP2025 #NLP #KnowledgeEditing #LLM #Steering ! Github: github.com/zjunlp/EasyEdi… Paper: arxiv.org/abs/2504.15133

zxlzr's tweet card. [ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs. - zjunlp/EasyEdit
EasyEdit/README_2.md at main · zjunlp/EasyEdit
來源: github.com
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年5月9日

🚀 Excited to introduce EasyEdit2 — a powerful upgrade to EasyEdit, now redesigned for unified, plug-and-play LLM behavior steering at inference time! #EasyEdit #LLM #ModelSteering #ModelEditing #KnowledgeEditing #EasyEdit2 #AI #InferenceTimeControl ✨ No retraining — just…

zxlzr's tweet image. 🚀 Excited to introduce EasyEdit2 — a powerful upgrade to EasyEdit, now redesigned for unified, plug-and-play LLM behavior steering at inference time! #EasyEdit #LLM #ModelSteering #ModelEditing #KnowledgeEditing #EasyEdit2 #AI #InferenceTimeControl ✨ No retraining — just…
zxlzr's tweet image. 🚀 Excited to introduce EasyEdit2 — a powerful upgrade to EasyEdit, now redesigned for unified, plug-and-play LLM behavior steering at inference time! #EasyEdit #LLM #ModelSteering #ModelEditing #KnowledgeEditing #EasyEdit2 #AI #InferenceTimeControl ✨ No retraining — just…
zxlzr's tweet image. 🚀 Excited to introduce EasyEdit2 — a powerful upgrade to EasyEdit, now redesigned for unified, plug-and-play LLM behavior steering at inference time! #EasyEdit #LLM #ModelSteering #ModelEditing #KnowledgeEditing #EasyEdit2 #AI #InferenceTimeControl ✨ No retraining — just…
zxlzr's tweet image. 🚀 Excited to introduce EasyEdit2 — a powerful upgrade to EasyEdit, now redesigned for unified, plug-and-play LLM behavior steering at inference time! #EasyEdit #LLM #ModelSteering #ModelEditing #KnowledgeEditing #EasyEdit2 #AI #InferenceTimeControl ✨ No retraining — just…
0
13
54
14
7千
1
4
22
5
4千
wangyiw33973985's profile picture. Assistant Professor at University of California, Merced
Yiwei Wang
 @wangyiw33973985
2024年4月3日

🔔Be careful when you use MQuAKE-3k to evaluate Knowledge Editing of LLMs. ❗ One-third of labels may not work due to knowledge conflicts. 💡 🔍Consider using our MQuAKE-2002 and MQuAKE-hard for more precise evaluations: wangywust.github.io/deepedit.io/ #LLM #KnowledgeEditing

wangyiw33973985's tweet image. 🔔Be careful when you use MQuAKE-3k to evaluate Knowledge Editing of LLMs. ❗ One-third of labels may not work due to knowledge conflicts. 💡 🔍Consider using our MQuAKE-2002 and MQuAKE-hard for more precise evaluations: wangywust.github.io/deepedit.io/ #LLM #KnowledgeEditing
0
3
12
3
1千
wangyiw33973985's profile picture. Assistant Professor at University of California, Merced
Yiwei Wang
 @wangyiw33973985
2024年4月2日

Our new paper proposes a new knowledge editing method for LLMs. 🚀 No retraining or prompt engineering, you only need our decoding method to achieve knowledge editing. 🎯 Feel free to check it at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #DecodingMethod #nlp2024

wangyiw33973985's tweet image. Our new paper proposes a new knowledge editing method for LLMs. 🚀 No retraining or prompt engineering, you only need our decoding method to achieve knowledge editing. 🎯 Feel free to check it at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #DecodingMethod #nlp2024
1
10
18
9
3千
GoatstackAI's profile picture. AI Agent that reads scientific papers for you and crafts personalized newsletters
GoatStack.AI
 @GoatstackAI
2024年3月31日

Explore an advanced multi-hop question-answering framework that updates real-time knowledge in LLMs through retrieval-augmented editing. #LLMs #KnowledgeEditing #QuestionAnswering

GoatstackAI's tweet image. Explore an advanced multi-hop question-answering framework that updates real-time knowledge in LLMs through retrieval-augmented editing. #LLMs #KnowledgeEditing #QuestionAnswering
1
1
3
0
224
wangyiw33973985's profile picture. Assistant Professor at University of California, Merced
Yiwei Wang
 @wangyiw33973985
2024年4月2日

New benchmarks for evaluating knowledge editing methods of LLMs. Precise evaluations through our benchmarks MQuAKE-2002 and MQuAKE-hard. 👮 See we remove the annotation mistakes due to knowledge conflicts at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #nlp2024

wangyiw33973985's tweet image. New benchmarks for evaluating knowledge editing methods of LLMs. Precise evaluations through our benchmarks MQuAKE-2002 and MQuAKE-hard. 👮 See we remove the annotation mistakes due to knowledge conflicts at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #nlp2024
1
7
15
2
2千
thealpha_dev's profile picture. A community of Dev, QA, and DevOps Engineers coming together to learn, grow, and connect
thealpha.dev
@thealpha_dev
年8月23日

🧠 Revolutionary: Edit LLM knowledge in real-time! EasyEdit lets you update facts, remove biases & insert knowledge WITHOUT retraining. UltraEdit processes 20K edits in 5 minutes! ACL 2024 published, 2.5k stars ⭐ github.com/zjunlp/EasyEdit #AI #LLM #KnowledgeEditing

thealpha_dev's tweet card. [ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs. - zjunlp/EasyEdit
GitHub - zjunlp/EasyEdit: [ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs.
來源: github.com
0
0
1
0
33
fei__sun's profile picture. Associate Professor @UCAS1978, working on RecSys, NLP, AI safety. My tweets are my own.
Sun Fei@EMNLP2025🚄
@fei__sun
年5月16日

We need more practical evaluation for #modelEditing #KnowledgeEditing 🤔 #LLM #AI #ACL2025

10k_miles_yang's profile picture. First-year PhD student at Institute of Computing Technology, Chinese Academy of Sciences, focusing on NLP, Knowledge Editing/Updating.
Wanli Yang
 @10k_miles_yang
年5月16日

😯To assess the real-world effectiveness of model editing techniques, we evaluated them on practical QA tasks and found that current editing methods perform substantially worse than previously reported (38.5% vs. 96%).

10k_miles_yang's tweet image. 😯To assess the real-world effectiveness of model editing techniques, we evaluated them on practical QA tasks and found that current editing methods perform substantially worse than previously reported (38.5% vs. 96%).
0
0
2
0
223
0
0
3
0
149
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2023年8月31日

Just dropped a quick update: EasyEdit now supports using the official fine-tuning API of gpt-3.5-turbo to customize ChatGPT for your editing cases. Try it out! 🔥📝 #EasyEdit #ChatGPT #KnowledgeEditing #NLP #AI #LLM Github: github.com/zjunlp/EasyEdit ChatGPT FT Setting:…

0
0
8
1
632
techyrushabh's profile picture. Coding, Roadmap, Solutions, Projects and more...
Techy Rushabh
 @techyrushabh
年3月21日

Want LLMs that *actually* learn from knowledge updates & don't just memorize? 🤔 Check out "CaKE" - Circuit-aware Knowledge Editing! It guides models to *reason* with new info, not just parrot it back. ➡️ [link to paper] #AI #KnowledgeEditing

1
0
0
0
4
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年4月3日

This is a systematic study on technical AGI safety and security. Interpretability techniques like steering vectors and circuit analysis can help us understand and improve LLM safety—but they can also be misused. #Safety #ModelEditing #KnowledgeEditing #LLM #NLP

sebkrier's profile picture. 🪼 AGI policy dev lead @GoogleDeepMind | rekkid junkie, dimensional glider, deep ArXiv dweller, interstellar fugitive, uncertain | 🛸
Séb Krier
@sebkrier
年4月2日

Excited to share @GoogleDeepMind's AGI safety and security strategy to tackle risks like misuse and misalignment. Rather than high-level principles, this 145-page paper outlines a concrete, defense-in-depth technical approach: proactively evaluating & restricting dangerous…

sebkrier's tweet image. Excited to share @GoogleDeepMind's AGI safety and security strategy to tackle risks like misuse and misalignment. Rather than high-level principles, this 145-page paper outlines a concrete, defense-in-depth technical approach: proactively evaluating & restricting dangerous…
19
94
494
298
43千
2
1
12
2
911
dsmall2apple1's profile picture. Research Fellow at NUS Research Interests: NLP, Structured Prediction, IE, KG, Neuro Symbolic Reasoning, Multi-Agent Collaboration, Knowledge Editing for LLMs
Shumin Deng
 @dsmall2apple1
年1月12日

Impressive summary and outlook of #KnowledgeEditing 😇

zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年1月12日

Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…

zxlzr's tweet image. Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…
2
3
45
9
5千
0
0
2
0
259
SongWang_SW's profile picture. Hi, I'm real Song Wang, a Ph.D. student at the Department of ECE, University of Virginia, advised by Prof. Jundong Li.
Song Wang
 @SongWang_SW
2023年10月27日

📜Join us on this journey to revolutionize the way we keep our language models relevant and up-to-date. Your input and insights are invaluable! 🙌 #KnowledgeEditing #LargeLanguageModels #Survey #AI #NLP #Innovation

0
0
3
0
232
thealpha_dev's profile picture. A community of Dev, QA, and DevOps Engineers coming together to learn, grow, and connect
thealpha.dev
@thealpha_dev
年8月23日

🧠 Revolutionary: Edit LLM knowledge in real-time! EasyEdit lets you update facts, remove biases & insert knowledge WITHOUT retraining. UltraEdit processes 20K edits in 5 minutes! ACL 2024 published, 2.5k stars ⭐ github.com/zjunlp/EasyEdit #AI #LLM #KnowledgeEditing

thealpha_dev's tweet card. [ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs. - zjunlp/EasyEdit
GitHub - zjunlp/EasyEdit: [ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs.
來源: github.com
0
0
1
0
33
longevityboris's profile picture. Founder @ 199 Biotechnologies, 199 Clinic, 199 Diagnostics & 199 Health | Investor & futurist advancing longevity, partial reprogramming & preventive medicine.
Boris Djordjevic
@longevityboris
年7月2日

Weight editors like ROME & MEMIT can overwrite facts, and ELDER stacks LoRA adapters for lifelong tweaks—but “ripple effects” still flip 20-30 % of related answers. #KnowledgeEditing

longevityboris's tweet image. Weight editors like ROME & MEMIT can overwrite facts, and ELDER stacks LoRA adapters for lifelong tweaks—but “ripple effects” still flip 20-30 % of related answers. #KnowledgeEditing
1
0
0
0
21
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年4月3日

This is a systematic study on technical AGI safety and security. Interpretability techniques like steering vectors and circuit analysis can help us understand and improve LLM safety—but they can also be misused. #Safety #ModelEditing #KnowledgeEditing #LLM #NLP

sebkrier's profile picture. 🪼 AGI policy dev lead @GoogleDeepMind | rekkid junkie, dimensional glider, deep ArXiv dweller, interstellar fugitive, uncertain | 🛸
Séb Krier
@sebkrier
年4月2日

Excited to share @GoogleDeepMind's AGI safety and security strategy to tackle risks like misuse and misalignment. Rather than high-level principles, this 145-page paper outlines a concrete, defense-in-depth technical approach: proactively evaluating & restricting dangerous…

sebkrier's tweet image. Excited to share @GoogleDeepMind's AGI safety and security strategy to tackle risks like misuse and misalignment. Rather than high-level principles, this 145-page paper outlines a concrete, defense-in-depth technical approach: proactively evaluating & restricting dangerous…
19
94
494
298
43千
2
1
12
2
911
techyrushabh's profile picture. Coding, Roadmap, Solutions, Projects and more...
Techy Rushabh
 @techyrushabh
年3月21日

Want LLMs that *actually* learn from knowledge updates & don't just memorize? 🤔 Check out "CaKE" - Circuit-aware Knowledge Editing! It guides models to *reason* with new info, not just parrot it back. ➡️ [link to paper] #AI #KnowledgeEditing

1
0
0
0
4
dsmall2apple1's profile picture. Research Fellow at NUS Research Interests: NLP, Structured Prediction, IE, KG, Neuro Symbolic Reasoning, Multi-Agent Collaboration, Knowledge Editing for LLMs
Shumin Deng
 @dsmall2apple1
年1月12日

Impressive summary and outlook of #KnowledgeEditing 😇

zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年1月12日

Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…

zxlzr's tweet image. Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…
2
3
45
9
5千
0
0
2
0
259
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年1月12日

Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…

zxlzr's tweet image. Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…
2
3
45
9
5千
Ofirlin's profile picture. Assistant professor at Bar Ilan University. Primary research is Machine learning, computational biology, and signal processing.
Ofir Lindenbaum
@Ofirlin
2024年11月12日

Results: State-of-the-art performance across datasets! Big shoutout to my amazing students @AmitRozner and Barak Battas for their amazing work! Join us and check out the paper here 👉 arxiv.org/pdf/2406.09920 #EMNLP #AIResearch #KnowledgeEditing #MachineLearning #NLP

1
1
3
0
227
prod42net's profile picture. 🌟Experienced #SolutionArchitect mastering #dotNET & more. From #Desktop to #Web, diving into #AI. Let's connect! 🚀#DevLife #TechJunkie #dotnet #cpp #pas #SQL
prod42net
@prod42net
2024年9月23日

🧠 Discover the latest in enhancing Large Language Models with innovative Knowledge Editing Techniques by Mike Young. Learn how KME can update models efficiently without losing valuable knowledge. #AI #NLP #KnowledgeEditing 📚 ift.tt/cjyp5d0

dev.to

Enhancing Large Language Models: A Survey of Knowledge Editing Techniques

Enhancing Large Language Models: A Survey of Knowledge Editing Techniques

來源: dev.to
0
0
1
0
18
bonybean's profile picture. Life Captured - Day by Day
Bony Bean
 @bonybean
2024年9月17日

Introducing OneEdit: A groundbreaking neural-symbolic system offering seamless integration and conflict resolution in knowledge graphs and large language models. Read the full blog post at: ift.tt/rThS4DA #AI #KnowledgeEditing #NeuralSymbolic

bonybean's tweet card. OneEdit: A Neural-Symbolic Collaborative Knowledge Editing System for Seamless Integration and Conflict Resolution in Knowledge Graphs and Large Language Models
OneEdit: A Neural-Symbolic Collaborative Knowledge Editing System for Seamless Integration and...
來源: marktechpost.com
0
0
2
0
33
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2024年8月1日

Welcome to attend the tutorial on "Knowledge Editing for Large Language Models" at IJCAI 2024 @IJCAIconf . Here are the details #NLP #AI #KnowledgeEditing #LLMs : Tutorial: Knowledge Editing for Large Language Models Location: Room 1F-Yeongju B Time: 9:00-12:30, August 3, 2024…

zxlzr's tweet image. Welcome to attend the tutorial on "Knowledge Editing for Large Language Models" at IJCAI 2024 @IJCAIconf . Here are the details #NLP #AI #KnowledgeEditing #LLMs : Tutorial: Knowledge Editing for Large Language Models Location: Room 1F-Yeongju B Time: 9:00-12:30, August 3, 2024…
1
13
70
24
6千
未找到 "#knowledgeediting" 的結果
qx_dong's profile picture. PhD student @PKU1898. Research Intern @MSFTResearch Asia.
Qingxiu Dong
 @qx_dong
2023年5月23日

The In-context Learning Editing (IKE) method proposed in this paper showcases impressive performance, even in complex multi-hop related scenarios. #knowledgeediting

qx_dong's tweet image. The In-context Learning Editing (IKE) method proposed in this paper showcases impressive performance, even in complex multi-hop related scenarios. #knowledgeediting
0
0
7
1
539
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2024年8月1日

Welcome to attend the tutorial on "Knowledge Editing for Large Language Models" at IJCAI 2024 @IJCAIconf . Here are the details #NLP #AI #KnowledgeEditing #LLMs : Tutorial: Knowledge Editing for Large Language Models Location: Room 1F-Yeongju B Time: 9:00-12:30, August 3, 2024…

zxlzr's tweet image. Welcome to attend the tutorial on "Knowledge Editing for Large Language Models" at IJCAI 2024 @IJCAIconf . Here are the details #NLP #AI #KnowledgeEditing #LLMs : Tutorial: Knowledge Editing for Large Language Models Location: Room 1F-Yeongju B Time: 9:00-12:30, August 3, 2024…
1
13
70
24
6千
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2024年5月21日

The tutorial of Knowledge Editing for Large Language Models at LREC-COLING 2024 @LrecColing is in progress. #NLP #LLMs #KnowledgeEditing

zxlzr's tweet image. The tutorial of Knowledge Editing for Large Language Models at LREC-COLING 2024 @LrecColing is in progress. #NLP #LLMs #KnowledgeEditing
0
1
19
0
1千
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
2023年10月14日

Our EasyEdit now supports editing multi-modal LLMs. Currently, you can edit MiniGPT-4 and Blip2. Check out the details here: github.com/zjunlp/EasyEdi…. Feel free to follow us for updates. Related paper: arxiv.org/abs/2310.08475. 🚀📚 #EasyEdit #ModelEditing #KnowledgeEditing…

zxlzr's tweet image. Our EasyEdit now supports editing multi-modal LLMs. Currently, you can edit MiniGPT-4 and Blip2. Check out the details here: github.com/zjunlp/EasyEdi…. Feel free to follow us for updates. Related paper: arxiv.org/abs/2310.08475. 🚀📚 #EasyEdit #ModelEditing #KnowledgeEditing…
0
9
39
9
6千
zxlzr's profile picture. Associate Professor @ZJU_China. Research interests include NLP, LLM, KG, Agent, Knowledge Editing.
Ningyu Zhang@ZJU
@zxlzr
年1月12日

Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…

zxlzr's tweet image. Over the past year, #KnowledgeEditing has experienced rapid development. As the new year begins, I’ve taken some time to reflect on the progress of this field and share my thoughts on its future directions. I look forward to discussing and collaborating with everyone to further…
2
3
45
9
5千
longevityboris's profile picture. Founder @ 199 Biotechnologies, 199 Clinic, 199 Diagnostics & 199 Health | Investor & futurist advancing longevity, partial reprogramming & preventive medicine.
Boris Djordjevic
@longevityboris
年7月2日

Weight editors like ROME & MEMIT can overwrite facts, and ELDER stacks LoRA adapters for lifelong tweaks—but “ripple effects” still flip 20-30 % of related answers. #KnowledgeEditing

longevityboris's tweet image. Weight editors like ROME & MEMIT can overwrite facts, and ELDER stacks LoRA adapters for lifelong tweaks—but “ripple effects” still flip 20-30 % of related answers. #KnowledgeEditing
1
0
0
0
21
GoatstackAI's profile picture. AI Agent that reads scientific papers for you and crafts personalized newsletters
GoatStack.AI
 @GoatstackAI
2024年3月31日

Explore an advanced multi-hop question-answering framework that updates real-time knowledge in LLMs through retrieval-augmented editing. #LLMs #KnowledgeEditing #QuestionAnswering

GoatstackAI's tweet image. Explore an advanced multi-hop question-answering framework that updates real-time knowledge in LLMs through retrieval-augmented editing. #LLMs #KnowledgeEditing #QuestionAnswering
1
1
3
0
224
wangyiw33973985's profile picture. Assistant Professor at University of California, Merced
Yiwei Wang
 @wangyiw33973985
2024年4月3日

🔔Be careful when you use MQuAKE-3k to evaluate Knowledge Editing of LLMs. ❗ One-third of labels may not work due to knowledge conflicts. 💡 🔍Consider using our MQuAKE-2002 and MQuAKE-hard for more precise evaluations: wangywust.github.io/deepedit.io/ #LLM #KnowledgeEditing

wangyiw33973985's tweet image. 🔔Be careful when you use MQuAKE-3k to evaluate Knowledge Editing of LLMs. ❗ One-third of labels may not work due to knowledge conflicts. 💡 🔍Consider using our MQuAKE-2002 and MQuAKE-hard for more precise evaluations: wangywust.github.io/deepedit.io/ #LLM #KnowledgeEditing
0
3
12
3
1千
wangyiw33973985's profile picture. Assistant Professor at University of California, Merced
Yiwei Wang
 @wangyiw33973985
2024年4月2日

Our new paper proposes a new knowledge editing method for LLMs. 🚀 No retraining or prompt engineering, you only need our decoding method to achieve knowledge editing. 🎯 Feel free to check it at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #DecodingMethod #nlp2024

wangyiw33973985's tweet image. Our new paper proposes a new knowledge editing method for LLMs. 🚀 No retraining or prompt engineering, you only need our decoding method to achieve knowledge editing. 🎯 Feel free to check it at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #DecodingMethod #nlp2024
1
10
18
9
3千
wangyiw33973985's profile picture. Assistant Professor at University of California, Merced
Yiwei Wang
 @wangyiw33973985
2024年4月2日

New benchmarks for evaluating knowledge editing methods of LLMs. Precise evaluations through our benchmarks MQuAKE-2002 and MQuAKE-hard. 👮 See we remove the annotation mistakes due to knowledge conflicts at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #nlp2024

wangyiw33973985's tweet image. New benchmarks for evaluating knowledge editing methods of LLMs. Precise evaluations through our benchmarks MQuAKE-2002 and MQuAKE-hard. 👮 See we remove the annotation mistakes due to knowledge conflicts at: wangywust.github.io/deepedit.io/ #KnowledgeEditing #LLMs #nlp2024
1
7
15
2
2千

Something went wrong.


Something went wrong.


United States Trends