[logseq-plugin-git:commit] 2025-12-27T19:07:25.270Z
This commit is contained in:
@@ -0,0 +1,37 @@
|
||||
progress:: {{renderer :todomaster}}
|
||||
icon:: 📅
|
||||
boox-notes:: 
|
||||
|
||||
- ### Tasks
|
||||
-
|
||||
- ### Notes
|
||||
- **Meeting with [[SoftwareMiningGroup]]**
|
||||
type:: [[meeting]]
|
||||
- Emeliot paper - Completing the evaluation
|
||||
- On the 4 december next call at 9:00
|
||||
- Target is IST journal
|
||||
- The second research question needs to be finalizaed
|
||||
- [[Phuong T. Nguyen]]
|
||||
- Working on the metagente paper
|
||||
- Almost done with all the comments. We can submit by tomorrow
|
||||
- TESORO revision
|
||||
- The student is working on it not in an intensive manner. There are experiments to be done.
|
||||
- Idea of Phuong is checking in LLM have been trained with malicious repositories
|
||||
- Following the flow of memorization
|
||||
- [https://arxiv.org/pdf/2412.13459](https://arxiv.org/pdf/2412.13459)
|
||||
- Involve the student from Vietnam that was involved on the CWE paper
|
||||
- The message of the paper is giving evidence that LLMs are trained with possibly malicious repositories.
|
||||
- Target is FSE 2026
|
||||
- It will be in Canada I guess.
|
||||
- We should also aim for [[ICSME]]
|
||||
- The deadline is on February.
|
||||
- Possibly idea is machine UNlearning!
|
||||
- Remove for instance the code if we notice that it is malicious
|
||||
- Update on [[Benchmarks]] #PROJECTS/MOSAICO #P1
|
||||
- LanguFuese e' rilasciato con licenza MIT
|
||||
- Alcune componenti hanno licenze enterprise. Componenti che in realtà non stiamo utilizzando
|
||||
- Collaborazione con Antonio
|
||||
- MCP *resources* vs *tools* (es calcolatrice per fare la somma) vs *prompt* (?)
|
||||
- Attualmente abbiamo fornito resources
|
||||
- La ricerca semantica sarà il tool
|
||||
-
|
||||
Reference in New Issue
Block a user