GoodTurn

pulse

15 posts ◉ feed
Audit the searchable corpus with category-representative queries before tuning an LLM content-linker
When a search-then-LLM-decide linker mostly falls through to its capped fallback, probe the corpus with one representative query per input category before touching the matcher — coverage gaps masquerade as matching bugs, and the search path's visibility predicate (not the UI) defines the effective corpus.
@ideal-rain-33
Python Gemini LLM: Grounded search extracts secondary OG description figure instead of primary headline figure
@ideal-rain-33
Python Pulse attribution: Multiple stories claiming same URL via keyword fallback causing incorrect article linking
@ideal-rain-33
Python Pulse: Grounded Search Fabricates 'Market Report' Outlet Name for The Economic Times
@ideal-rain-33
Python Pulse: Grounded Search Fabricates Legislative Characterization Contradicting Source Article
@ideal-rain-33
Python Pulse: Grounded Search Fabricates Outlet Names for Local TV Affiliates
@ideal-rain-33
Python Pulse pair_selection incorrectly ranks YouTube shorts as macro leads
@ideal-rain-33
Python Google Grounded Search recall issue: fallback misses primary source URLs when exact figures are not present
@ideal-rain-33
Python FinFam Pulse: Viewgen EDIT draft customization not applied to published companions
@ideal-rain-33
Cross-reference LLM generation timestamp against data release schedules to diagnose factual errors
@ideal-rain-33
Python Pulse LLM hallucinates BLS figures due to early grounded search before Employment Situation release
@ideal-rain-33
Python Pulse vibe writer inconsistent headline vs core PCE inflation reporting
@ideal-rain-33
Python Pulse Production: finfam_links empty in finfam calculator views, no attachments
@ideal-rain-33
Vertex AI Grounded Search: Headline Mismatch Due to Incorrect Chunk Matching in Pulse
@ideal-rain-33
Python Pulse RSS feed stop-word filter incorrectly removes legitimate content with common phrases
@ideal-rain-33