Files
logseq/journals/2026_01_29.md
T

4.9 KiB

progress:: {{renderer :todomaster}} icon:: 📅 boox-notes:: BOOX Notes

- ### Tasks
	- {{query (and (task TODO) [[29-01-2026]])}}
	  query-table:: true
	  query-properties:: [:block]
	- ...
- ### Notes
	- Meeting CM3 [[PROJECTS/MOSAICO]]
	  type:: [[meeting]]
		- WP4
			- ![image.png](../assets/image_1769671414296_0.png)
			- ![image.png](../assets/image_1769671513452_0.png)
			- ![image.png](../assets/image_1769671605759_0.png)
			- ![image.png](../assets/image_1769671708456_0.png)
			- ![image.png](../assets/image_1769671875745_0.png)
			- ![image.png](../assets/image_1769672175935_0.png)
				- How can we position BDI with what has been done so far? [[question]]
			- ![image.png](../assets/image_1769672400208_0.png)
				- Keep an eye on the deliverable [[Deliverable D3.2]] which is on collaboration pattern as shown below
			- ![image.png](../assets/image_1769672456745_0.png)
			- ![image.png](../assets/image_1769672504982_0.png)
				- How do we link with languages like BPMN? [[question]]
					- Ok answered below
			- ![image.png](../assets/image_1769672608482_0.png)
			- ![image.png](../assets/image_1769672674672_0.png)
			- ![image.png](../assets/image_1769672788052_0.png)
				- This seems to be like too high-level
			- ![image.png](../assets/image_1769672886870_0.png)
			- ![image.png](../assets/image_1769672971263_0.png)
			- ![image.png](../assets/image_1769673005517_0.png)
				- You said that the datasets are necessary to train the model. But it is sure that the existing models are nothing to do with agents or multi-agent systems. Isn't it? [[question]]
			- ![image.png](../assets/image_1769673122545_0.png)
				- I think you should focus on the workflow benchmarks [[question]]
			- ![image.png](../assets/image_1769673209665_0.png)
				- This should be looked also for our [[Benchmarking framework]]
				- How do you see this linked with the benchmarking functionalities of the repository? [[question]]
				- https://zjunlp.github.io/project/WorFBench/
			- ![image.png](../assets/image_1769673564213_0.png){:height 538, :width 921}
			- ![image.png](../assets/image_1769673870219_0.png)
			- ![image.png](../assets/image_1769673930716_0.png)
			- ![image.png](../assets/image_1769674026611_0.png)
			- ![image.png](../assets/image_1769674068587_0.png)
			- ![image.png](../assets/image_1769674158359_0.png)
			- ![image.png](../assets/image_1769674200531_0.png)
				- How would you expect the query shown at slides 36 from working w.r.t. what we have seen yesterday? [[question]]
					- This needs to be discusssed but the constraint based specifications will be trasleted as requests to the repository API
			- ![image.png](../assets/image_1769674423475_0.png)
				- For the consensus at certain point, humans might be involved. The reference agents will take care of this human-in-the-loop step
			- ![image.png](../assets/image_1769675036866_0.png)
			- ![image.png](../assets/image_1769675098624_0.png)
				- ACL - Agent Communication Library
			- ![image.png](../assets/image_1769675185370_0.png)
			- ![image.png](../assets/image_1769675290508_0.png)
			- ![image.png](../assets/image_1769675381847_0.png)
			- ![image.png](../assets/image_1769675404747_0.png)
				- They have developed a small DSL to specify what the agent does and they have also developed a transformation to generate the corresponding agent card.
				- It seems we can simplify the agent data model by removing attributes related to BDI [[Notes]]
		- WP4
			- ![image.png](../assets/image_1769677103771_0.png)
			- ![image.png](../assets/image_1769677121726_0.png)
			- ![image.png](../assets/image_1769677210875_0.png)
			- ![image.png](../assets/image_1769677271502_0.png)
			- ![image.png](../assets/image_1769677335081_0.png)
			- ![image.png](../assets/image_1769677469777_0.png)
			- ![image.png](../assets/image_1769677481214_0.png)
				- "Run-time" metamodel
			- ![image.png](../assets/image_1769677543799_0.png)
			- ![image.png](../assets/image_1769677563832_0.png)
				- Kubernetes becasuse they have an explicit governance process.
			- ![image.png](../assets/image_1769677720969_0.png)
			- ![image.png](../assets/image_1769677755937_0.png)
				- That's interesting. Very few projects have an explicit governance process.
			- ![image.png](../assets/image_1769677843753_0.png)
				- Maybe this was not in 2015
			- ![image.png](../assets/image_1769677975417_0.png)
			- ![image.png](../assets/image_1769678283609_0.png)
			- ![image.png](../assets/image_1769678392403_0.png)
			- ![image.png](../assets/image_1769678441623_0.png)
				- The idea is to translate this graphical notation towards more specific technologies like Claude-Flow, OpenAI Swarm, Langchain Deep Agents, etc.
		- ### **Questions**
			- {{query (and [[question]] [[29-01-2026]] )[[question]]}}
			  query-table:: true
			  query-properties:: [:block]