Progressive 20-questions prompting lifts local Clef-Flash fax classification to 49/50
mvanhorn · x · 2026-10-06
mvanhorn shares a weekend experiment classifying faxes with locally-run Clef-Flash: direct classification was underwhelming, but switching to a 20-questions-style progressive prompting approach got 49/50 on hand-labeled ground truth — a practical pattern for boosting small local models on classification tasks.
Related event: October agentic engineering tips collection: personal agents take over(4 posts)→
More from Models
- Strata engine boosts RTX 3090 prefill 17x, reigniting the local LLM debate — Iory1998 · 2026-10-06
- Reflection unveils Beam: 501B-param open MoE model with 23B active, trained on 23.8T tokens — bigblueboo · 2026-10-06
- Apodex 1.1 Mini hits #4 on OpenRouter trending, 79.6B tokens processed in 4 days — SimonShaoleiDu · 2026-10-06
- "OCR is superhuman now" — a short take on document AI's leap — generativist · 2026-10-06
- Dev returns to Codex after a week on Opus 5.5: "basically unusable" — jacob_posel · 2026-10-06
- CMDB-1500: open-source multimodal benchmark with 1,500 decision-making tasks — Powerful_Buy_4616 · 2026-10-06