La metamorfosis de la práctica científica: Del investigador individual a la hiperautoría e hiperinflación de papers --> https://publicaciones.unpa.edu.ar/index.php/RevITA/article/view/980 #ciencia #astronomía #papers #academia #investigación

Minimal envy mechanisms, by Julien Combe #EconTwitter #algorithms #matching #papers
http://marketdesigner.blogspot.com/2023/08/minimal-envy-mechanisms-by-julien-combe.html

Not sure if sb proposed this already: wouldn't adding #paragraph numbers to #research #papers make #academic #discussion easier? Instead of: "the idea in paper XYZ on page 294, left column, 3rd §, last sentence" you could say "the idea in paper XYZ in §3, last sentence". Moreover, I see potential benefits for #automatic #document #processing and even the writing process itself: authors would maybe stick more closely to the 1-paragraph-1-idea concept. #academicwriting Are there any such approaches to improve structure in the academic #discourse?

3/ REFERENCES

[1] Tianyi Zhang, Ganesha Upadhyaya, Anastasia Reinhardt, Hridesh Rajan, and Miryung Kim. 2018. Are code examples on an online Q&A forum reliable? a study of API misuse on stack overflow. In Proceedings of the 40th International Conference on Software Engineering, pages 886–896, Gothenburg, Sweden. Association for Computing Machinery. http://dx.doi.org/10.1145/3180155.3180260

#GenerativeAI #papers #NLP #NLProc #SoftwareDevelopment

2/ Since users of #CodeGeneration with particular APIs are usually relatively inexperienced in the said APIs, these inaccuracies may have grave consequences to the robustness and reliability of the resulting software.

(How would #CodeLlama fare?)

#GenerativeAI #papers #NLP #NLProc #SoftwareDevelopment

1/ How robust and reliable is the code generated by #LLMs, especially for real-world software development? A recent work [2] constructed a new benchmark based on [1] to evaluate if the generated code uses API correctly. Four popular #LLMs -- #GPT3.5, #GPT4, #Llama2, and #Vicuna -- are tested, and #GPT4 under zero-shot scored 62.09% misuse rate. Even with one-shot relevant examples the misuse rate of #GPT4 is 49.17%.

#GenerativeAI #papers #NLP #NLProc #SoftwareDevelopment

7/ REFERENCES

[1] Eric Mitchell, Charles Lin, Antoine Bosselut, Chelsea Finn, and Christopher D. Manning. 2021. Fast Model Editing at Scale. http://arxiv.org/abs/2110.11309

[2] Damai Dai, Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, and Furu Wei. 2021. Knowledge Neurons in Pretrained Transformers. http://arxiv.org/abs/2307.14988

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

6/ It’d be interesting to see how “complete” various model editing methods can achieve (screenshot), and how to achieve a better tradeoff between completeness and efficiency.

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

5/ Another related recent work is on how to properly evaluate model editing [5]. Instead of just assessing whether an individual fact has been successfully injected or if similar predictions for other subjects have not changed, the *consequences* of these updates, aka “ripple effects”, should also be evaluated (screenshot).

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

4/ They then propose PMET (Precise Model Editing in a Transformer), which effectively moves weight optimization upstream. And while trying to optimize both MHSA and FFN, it only updates on FFN weights to preserve specificity (screenshot 1). The result is much more stable performance as number of edits increases (screenshot 2), and better overall results (screenshot 3).

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

3/ By computing the similarities of the hidden states of MHSA and FFN before and after each layer, they observe FFN stabilizes much earlier than MHSA, thus concluding that MHSA encodes more general knowledge extraction patterns while FFN captures more factual aspect of knowledge (screenshot).

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

2/ Building on ROME/MEMIT, the authors of a more recent work [4] hypothesize that optimizing transformer layer hidden states to update knowledge may be too much, as these parameters contain simultaneously the effect from Multi-head Self-Attention (MHSA), feed-forward network (FFN), and residual connections.

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

1/ Can we "edit" #LLMs to update incorrect/outdated facts without costly retraining? Recent works such as training auxiliary models to predict weight changes in the main model (MEND) [1], locating "knowledge neurons" [2], using causal intervention to identify feed-forward network (FFN) weights to edit in ROME [3], and scaling up editing operations to thousands of associations in MEMIT [4] have proven it's doable, even practical.

#papers #NLP #NLProc #Knowledge #KnowledgeGraph

For accessibility reasons, arXiv is starting to publish HTML versions of papers. https://info.arxiv.org/about/accessibility_html_papers.html 🧵

#math #papers #openscience #academia #arXiv
@brembs
@lambo

I think this is interesting and welcome, especially on mobile devices. It is not without problems to want to quickly check some fact on your phone, download the PDF, go to landscape mode, find the right location in the paper, zoom in, etc.
1/4

@floe beats one of the reviews I got for a rejected paper when I was a PhD student, it went something like this, "Incorrect citation format... Reject". No other issue was raised by the reviewer.

#academic #reviews #papers

Freezing (and then carefully warming) donor organs breaks the organ transplant time barrier (for rats, for the time being) #EconTwitter #deceaseddonors #papers #transplantation
http://marketdesigner.blogspot.com/2023/08/freezing-and-then-carefully-warming.html

The principle of mutual recognition inside the Common European Asylum System https://www.meltingpot.org/2023/08/the-principle-of-mutual-recognition-inside-the-common-european-asylum-system/ #Tesidilaurea,ricercheestudi #RegolamentoDublino #Dirittodiasilo #UnioneEuropea #Papers

入国審査ゲーム「Papers, Please」が液晶ゲームに？　リリース10周年を記念したブラウザ向けデメイク作品「LCD, Please」公開中
https://www.4gamer.net/games/247/G024771/20230809004/
#4gamer #Papers #Please #サバイバル #パズル

The extraterritorial effect of refugee status https://www.meltingpot.org/2023/07/the-extraterritorial-effect-of-refugee-status/ #Tesidilaurea,ricercheestudi #Protezionesussidiaria #Statusdirifugiato #Dirittodiasilo #UnioneEuropea #Papers

On the evolution of natural product biosynthesis https://pubmed.ncbi.nlm.nih.gov/37507161/?utm_source=Other&utm_medium=rss&utm_campaign=None&utm_content=1BkdSGR7C2ZfUX8uF1iLxLu_FmeD43McnZfu-WWUWa2ZXtAobE&fc=None&ff=20230729070003&v=2.17.9.post6+86293ac #EvoDevo #Papers