rve/learning-platform
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat/r42-accuracy-and-clear-history #34

Merged
rve merged 2 commits from feat/r42-accuracy-and-clear-history into main 2026-07-13 12:42:13 +00:00
Conversation 0 Commits 2 Files Changed 10
+357 -30
rve commented 2026-07-13 12:42:01 +00:00
Owner
Copy Link
Copy Source

Why

R42 was omitting knowledge-graph information (notably pension questions):
relevant facts existed in the graph but never reached the model.

Fixes

  • Retrieval — exact-token TF-IDF couldn't match Dutch compounds, so a
    pensioen query scored 0 against pensioenregeling / partnerpensioen
    and never retrieved them. Added a compound-word fallback (shared ≥6-char
    stem or containment, 0.4× weight) alongside exact matching.
  • Context building — deep article content was injected only for
    verbatim-mentioned topics; retrieved topics gave just a 200-char
    description. Now injects ~1000 chars for up to 5 topics (mentions first,
    then top-ranked retrieved); description widened to 320.
  • Prompt — new NAUWKEURIGHEID block (use all relevant facts, call
    lookup_topic before giving up); relaxed the 4-sentence cap for
    detail/list answers.

Clear history

Trash button in the chat header (confirm dialog) wipes
chat:thread:{userId} and reseeds the greeting via clearThread().

Tests

144 passing. Added compound-word retrieval tests + a new rag test for
deep-content injection. Lint + build clean.

## Why R42 was omitting knowledge-graph information (notably pension questions): relevant facts existed in the graph but never reached the model. ## Fixes - **Retrieval** — exact-token TF-IDF couldn't match Dutch compounds, so a `pensioen` query scored 0 against `pensioenregeling` / `partnerpensioen` and never retrieved them. Added a compound-word fallback (shared ≥6-char stem or containment, 0.4× weight) alongside exact matching. - **Context building** — deep article content was injected only for verbatim-mentioned topics; retrieved topics gave just a 200-char description. Now injects ~1000 chars for up to 5 topics (mentions first, then top-ranked retrieved); description widened to 320. - **Prompt** — new NAUWKEURIGHEID block (use all relevant facts, call `lookup_topic` before giving up); relaxed the 4-sentence cap for detail/list answers. ## Clear history Trash button in the chat header (confirm dialog) wipes `chat:thread:{userId}` and reseeds the greeting via `clearThread()`. ## Tests 144 passing. Added compound-word retrieval tests + a new rag test for deep-content injection. Lint + build clean.
rve added 2 commits 2026-07-13 12:42:01 +00:00
feat(r42): improve KB grounding accuracy and add clear-history 85452f66a7 
R42 was missing knowledge-graph information (e.g. pension questions)
because retrieval and context-building dropped relevant facts:

- retrieval: exact-token TF-IDF could not match Dutch compound words,
  so a "pensioen" query scored 0 against "pensioenregeling" /
  "partnerpensioen" and never retrieved them. Add a compound-word
  fallback (shared >=6-char stem or containment, 0.4x weight) alongside
  exact matching.
- rag: deep article content was only injected for verbatim-mentioned
  topics; retrieved topics contributed just a 200-char description.
  Inject ~1000 chars of content for up to 5 topics (mentions first,
  then top-ranked retrieved) and widen the description snippet to 320.
- prompts: add a NAUWKEURIGHEID block (use all relevant facts, call
  lookup_topic before giving up) and relax the 4-sentence cap for
  detail/list answers so complete facts aren't summarised away.

Also add a clear-history control: a trash button in the chat header
(confirm dialog) wipes chat🧵{userId} and reseeds the greeting
via clearThread() in useChat.

Tests: compound-word matching + rag deep-content injection. Spec updated.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
feat: add guide for calculating and processing pension premiums in proforma payslips
On Pull Request to Main / test (pull_request) Successful in 44s
Details
On Pull Request to Main / publish (pull_request) Successful in 1m14s
Details
On Pull Request to Main / deploy-dev (pull_request) Successful in 3m15s
Details
17cb30003d
rve merged commit 39e448e807 into main 2026-07-13 12:42:13 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No labels
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
eho (Edwin van den Houdt) lmi (Leroy Mies) lpr (Leon de Pruyssenaere de la Woestijne) not (Niek Otten) rdi (Robert van Diest) rve (Raymond Verhoef)
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: rve/learning-platform#34