Implement Relevance Scoring in Episodic Memory
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 73
- Forks
- 89
- Avg merge
- 14d 58m
- Merged PRs (30d)
- 2
Description
**What's the problem this feature will solve?**
The current EpisodicMemory module successfully implements recency and LLM-graded importance scoring, but is missing the relevance logic that is needed, since the entire memory architecture is inspired from the "Generative Agents" paper, We need to implement vector similarity scoring to complete the "Generative Agents" retrieval architecture.
**Describe the solution you'd like**
Adding a relevancy based scoring logic to the entire file based on a query string which is obtained from the observations stored.
**Are you planning to open a PR for this?**
Yes I'm working on a PR for this issue :)
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start in the EpisodicMemory module and read the existing recency and LLM-graded importance scoring logic. Trace how observations produce the query string, then inspect the retrieval flow to determine where vector similarity belongs. Done means relevance scoring is integrated with the existing retrieval architecture described in the issue.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100